home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
io Programmo 17
/
IOPROG_17.ISO
/
soft
/
formats
/
filefmts.lst
< prev
next >
Encoding:
Amiga (detected)
Atari
Commodore
DOS
FM Towns/JPY
Macintosh
Macintosh JP
Macintosh to JP
NeXTSTEP
RISC OS/Acorn
Shift JIS
UTF-8
Wrap
File List
|
1998-07-02
|
283.2 KB
|
6,008 lines
**********************************************************************
* This file contains header_info about these file_formats : *
* *
* 669/IFF-voice/amiga exe/AMS/ARC/ARJ/AU/BGI/BMP/CEG/CEL/CHR/ *
* CMF/COL/COM/CORE/CPI/Dbase II/Dbase III/DBase IV/DMF/DMS/DWC/ *
* EFE/EXE/FAR/FLC/TWE/OPT/FLI/FON/FPT/FSM/GF1 PAT/GIF/GZIP/GKH/ *
* GL/GRIB/HSI/HYP/IFF/INS/JPG/LBM/LBR/LZH/MID/MOD/MSK/MTM/ *
* MZ-EXE/NE-EXE/NG/OBJ/OS2-hlp/PBM/PCX/PIC/PICpro/PLY/PS16/QFX/ *
* RAW/RDIB/RIFF/RIFX/S3I/S3M/SND/SQZ/SDK/SDS/SMP/STM/TAR/TIFF/ *
* TXW/UWF/ULT/WAVE/WKS/DOC/WP-file/WQ1/XM/ZIP/ZOO/Zyxel *
* *
**********************************************************************
File format list Release 1.00 Last change 05/28/95
This compilation is Copyright (c) 1994,1995 Max Maischein
This is only the beta version. Please do not redistribute, as the
format might still change.
--------!-CONTACT_INFO----------------------
If you notice any mistakes or omissions, please let me know! It is only
with YOUR help that the list can continue to grow. Please send
all changes to me rather than distributing a modified version of the list.
This file has been authored in the style of the INTERxxy.* file list
by Ralf Brown, and uses almost the same format.
Please read the file FILEFMTS.1ST before asking me any questions. You may find
that they have already been addressed.
Max Maischein
Max Maischein, 2:244/1106.17
Max_Maischein@spam.fido.de
corion@hera.rbi.uni-frankfurt.de
Corion on IRC
--------!-DISCLAIMER------------------------
DISCLAIMER: THIS MATERIAL IS PROVIDED "AS IS". I verify the information
contained in this list to the best of my ability, but I cannot be held
responsible for any problems caused by use or misuse of the information,
especially for those file formats foreign to the PC, like AMIGA or SUN file
formats. If an information it is marked "guesswork" or undocumented, you
should check it carefully to make sure your program will not break with
an unexpected value (and please let me know whether or not it works
the same way).
Information marked with "???" is known to be incomplete or guesswork.
Some file formats were not released by their creators, others are regarded
as proprietary, which means that if your programs deal with them, you might
be looking for trouble. I don't care about this.
--------!-FLAGS-----------------------------
One or more letters may follow the file format ID; they have the following
meanings:
Cx - Charset used :
7 - Unix 7-bit characters
A - Amiga charset (if there is one)
E - EBDIC character format
U - Unicode character set
W - Windows char set
Default is the 8-Bit IBM PC-II Charset. Note that Microsoft
introduced codepages which might be relevant with other
programs.
G - guesswork, incomplete, unreliable etc.
M - Motorola byte order
Default is Intel byte order
O - obsolete, valid only for version noted below
X - Synonym topic. See topic named under see also.
--------!-CATEGORIES------------------------
The ninth column of the divider line preceding an entry usually contains a
classification code for the application that uses those files.
The codes currently in use are:
! - User information ( not really a file format )
A - Archives
a - Animations
B - Binary files for compilers etc.
H - Help file
I - Images, bit maps
D - Data support files
E - Executable files
f - Generic file format. RIFF and IFF are generic file formats.
F - Font files (TTF)
G - General graphics file
M - Module music file (MIDI,MOD)
R - Resource data files
S - Sound files (WAV,VOC)
T - Text files (DOC,TXT)
W - Spreadsheet and related (WKS)
X - Database files (DBF)
--------!-FIELDS----------------------------
After a format description, you will sometimes find other keywords. The
meanings of these are :
EXTENSION:
This is the default extension of files of the given type.
On DOS systems, most files have a 3 letter extension.
On Amiga systems, the files are prefixed with something.
The DOS extensions are all uppercase, extensions for other systems
are in lower case chars. On other systems, which do not have the con-
cept of extensions, as the MAC, this is the file type.
OCCURENCES:
Where you are likely to encounter those files. This specifies
machines (like PC,AMIGA) or operating systems (like UNIX).
PROGRAMS:
Programs which either create, use or convert files of this format.
Some might be used for validation or conversion.
REFERENCE:
A reference to a file or an article in a magazine which is mandatory
or recommended for further understanding of the matter.
SEE ALSO:
A cross reference to a topic which might be interesting as well.
VALIDATION:
Methods to validate that the file you have is not corrupt. Normally
this is a method to check the theoretical file size against the
real filesize. Some file formats allow no reliable validation.
--------!-FORMAT----------------------------
The block oriented files are organized in some other fashion, since the
order of blocks is at best marginally obligatory.
Each block type starts with the block ID (eg. RIFFblock for a RIFF file) and
in square brackets the character value of the ID field (eg. [WAVE] for RIFF
WAVe sound files). The block itself is descripted in the format description,
that means you will have to look after RIFF or FORM. In the record
description, the header information is omitted !
If a record is descripted, the record ends when the next offset is given.
Bitmapped values have a description for each bit. The value left of the
slash ("/") is for the bit not set (=0), the right sided value applies
if the bit is set.
A note on the tables section. The tables were added as they were introduced
into Ralf Browns interrupt list - so not everything was pressed into a table.
The tables (should) have unique numbers, but they sure are out of order !
--------!-MACHINES--------------------------
Machines that use Intel byte ordering
PC
Machines that use Motorola byte ordering
AMIGA, ATARI ST, MAC, SUN
--------M-669-------------------------------
The .669 format is a module format for digital music.
OFFSET Count TYPE Description
0000h 1 word ID=6669h
0002h 108 byte ASCII song message
006Eh 1 byte Number of saved samples (0-40h)
="NOS"
006Fh 1 byte Number of saved patterns (0-80h)
="NOP"
0070h 1 byte Loop order number
0071h 128 byte Order list
00F1h 128 byte Tempo list for patterns
0171h 128 byte Break location list for patterns
01F1h "NOS" rec Sample data
The sample data is in the file
for "NOS"
13 byte ASCIIZ filename of instrument
1 dword Length of instrument sample
1 dword Offset of beginning of loop
1 dword Offset of end of loop
01F1h+ "NOP"*600 rec The note patterns
"NOS"*19h Those patterns are repeated for each row,
and the array of these is repeated 64 times
for each pattern.
3 byte Note(see table 0000)
01F1h+ ? byte Sample data (unsigned)
"NOS"*0x19+
"NOP"*0x600
(Table 0000)
669 Note format
Each note looks like this :
BYTE[0]: BYTE[1]: BYTE[2]:
nnnnnnii iiiivvvv ccccdddd
n : note value
i : 6-bit instrument number
v : 4-bit volume
c : command data (Protracker format mapped) :
0 = a
1 = b
2 = c
3 = d
4 = e
5 = f
d : command value (Protracker format)
Special values for byte 0 :
0FEh : no note, only volume
0FFh : no note or no command, if byte 2 = 0FFh
EXTENSION:669
OCCURENCES:PC
SEE ALSO:MOD
PROGRAMS:669 Mod Composer, DMP
VALIDATION:
--------S-8SVX-MG---------------------------
The 8SVX files are IFF files used for digital audio data. The format of
the VHDR block is complete guesswork. These files use Motorola byte order.
The 8SVX file format is fixed to 8-bit mono sample data - at least GoldWave
does not support saving files in any other format than 8-bit mono.
FORMblock [VHDR]
This is the sample information block. The normal size is 20 bytes.
OFFSET Count TYPE Description
0000h 1 dword Sampling rate of digital data in Hz.
This count seems not to be too accurate,
at least GoldWave v2.0 creates different
rates for Wave and 8SVX files.
0004h 4 dword Other data, unknown
FORMblock [BODY]
This block contains the raw sample data, maybe the usual IFF compression was
used. The details of both the compression and the information about the IFF
format are unknown to me.
EXTENSION:IFF
OCCURENCES:PC,Amiga
PROGRAMS:GoldWave
SEE ALSO:IFF,WAVE
VALIDATION:
--------S-AIFC-MG---------------------------
The AIFC files seem to be a variation of the AIFF files - but I don't know
anything about them.
EXTENSION:IFF
SEE ALSO:AIFF
--------S-AIFF-MG---------------------------
The Audio Interchangeable File Format files are digital audio files
stored in the IFF format; the samples are stored in signed PCM. The header
block is [AIFF], different subblocks are :
[AUTH]
The authors information; optional
[COMM]
This record stores information about the sampled data :
OFFSET Count TYPE Description
0000h 1 word ??? number of channels ???
??? number of instrument samples ???
0002h 1 dword Sample length
0006h 1 dword lower frequency
000Ah 1 dword maximum frequency
000Dh 1 dword ???
[MARK]
[NAME]
The name of the instrument / sample
[SSND]
The stored sample data.
Further information wanted.
EXTENSION:AIF,IFF
--------E-AMIGA EXECUTABLE-MG---------------
All Amiga executables I've seen start with this signature. Of course the
bytes are in Motorola byte order, as you would exspect from a Motorola
based machine. This info here is based completely on my guesswork, maybe
somebody from the Amiga could help flesh out this part.
OFFSET Count TYPE Description
0000h 1 dword ID=03F3h
EXTENSION:EXE
OCCURENCES:AMIGA
SEE ALSO:
VALIDATION:
--------M-AMS-------------------------------
The AMS format is a multichannel module format created by the X-Tracker (not
to be mistaken for he tracker of the same name by D-Lusion).
The X-Tracker by Extreme PC is a multichannel tracker that features 32 digital
channels, 64 MIDI channels, 255 samples, 64K patterns and positions. The tracker
is currently in beta status and not enough information is yet available yet.
OFFSET Count TYPE Description
EXTENSION:
OCCURENCES:
PROGRAMS:
REFERENCE:
SEE ALSO:MOD
VALIDATION:
--------A-ARC-------------------------------
The ARC files are archive files created by the SEA ARC program. The compression
has been superceded by more recent compression programs. Similar archives can
be created by the PAK and PkPAK programs.
OFFSET Count TYPE Description
0000h 1 byte ID=1Ah
0001h 1 byte Compression method (see table 0001)
0002h 12 char File name
000Fh 1 dword Compressed file size
0013h 1 dword File date in MS-DOS format (see table 0009)
0017h 1 word 16-bit CRC
0019h 1 dword Original file size
="SIZ"
(Table 0001)
ARC compression types
1 - unpacked (obsolete)
2 - unpacked
3 - packed
4 - squeezed (after packing)
5 - crunched (obsolete)
6 - crunched (after packing) (obsolete)
7 - crunched (after packing, using faster hash algorithm)
8 - crunched (after packing, using dynamic LZW variations)
9 - Squashed c/o Phil Katz (no packing) (var. on crunching)
10 - crushed (PAK only)
11 - distilled (PAK only)
(Table 0009)
Format of the MS-DOS time stamp (32-bit)
The MS-DOS time stamp is limited to an even count of seconds, since the
count for seconds is only 5 bits wide.
31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16
|<---- year-1980 --->|<- month ->|<--- day ---->|
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
|<--- hour --->|<---- minute --->|<- second/2 ->|
EXTENSION:ARC,PAK
OCCURENCES:PC
PROGRAMS:SEA ARC,PAK,PkPAK
SEE ALSO:
VALIDATION:FileSize="SIZ"
--------A-ARJ-------------------------------
The ARJ program by Robert K. Jung is a "newcomer" which compares well to PKZip
and LhArc in both compression and speed. An ARJ archive contains two types of
header blocks, one archive main header at the head of the archive and local
file headers before each archived file.
OFFSET Count TYPE Description
0000h 1 word ID=0EA60h
0002h 1 word Basic header size (0 if end of archive)
0004h 1 byte Size of header including extra data
0005h 1 byte Archiver version number
0006h 1 byte Minimum version needed to extract
0007h 1 byte Host OS (see table 0002)
0008h 1 byte Internal flags, bitmapped :
0 - no password / password
1 - reserved
2 - file continues on next disk
3 - file start position field is available
4 - path translation ( "\" to "/" )
0009h 1 byte Compression method :
0 - stored
1 - compressed most
2 - compressed
3 - compressed faster
4 - compressed fastest
Methods 1 to 3 use Lempel-Ziv 77 sliding window
with static Huffman encoding, method 4 uses
Lempel-Ziv 77 sliding window with pointer/
length unary encoding.
000Ah 1 byte File type :
0 - binary
1 - 7-bit text
2 - comment header
3 - directory
4 - volume label
000Bh 1 byte reserved
000Ch 1 dword Date/Time of original file in MS-DOS format
0010h 1 dword Compressed size of file
0014h 1 dword Original size of file
0018h 1 dword Original file's CRC-32
001Ah 1 word Filespec position in filename
001Ch 1 word File attributes
001Eh 1 word Host data (currently not used)
? 1 dword Extended file starting position when used
(see above)
? char ASCIIZ file name
? char Comment
????h 1 dword Basic header CRC-32
????h 1 word Size of first extended header (0 if none)
="SIZ"
????h+"SIZ"+2 1 dword Extended header CRC-32
????h+"SIZ"+6 ? byte Compressed file
(Table 0002)
ARJ HOST-OS types
0 - MS-DOS
1 - PRIMOS
2 - UNIX
3 - AMIGA
4 - MAC-OS (System xx)
5 - OS/2
6 - APPLE GS
7 - ATARI ST
8 - NeXT
9 - VAX VMS
EXTENSION:ARJ
OCCURENCES:PC
PROGRAMS:ARJ.EXE
REFERENCE:
SEE ALSO:
VALIDATION:
--------S-AU-MG-----------------------------
The AU files are digital audio files used by the Sun and NeXT
workstations.
Further information wanted.
OFFSET Count TYPE Description
0000h 4 char ID='.snd'
0004h 1 dword Offset of start of sample
0008h 1 dword Length of stored sample
000Ch 1 dword Sound encoding :
1 - 8-bit ISDN u-law
2 - 8-bit linear PCM (REF-PCM)
3 - 16-bit linear PCM
4 - 24-bit linear PCM
5 - 32-bit linear PCM
6 - 32-bit IEEE floating point
7 - 64-bit IEEE floating point
23 - 8-bit ISDN u-law compressed(G.721 ADPCM)
0010h 1 dword Sampling rate
0014h 1 dword Number of sample channels
EXTENSION:AU
OCCURENCES:SunOS
--------B-BGI-G-----------------------------
The BGI files are graphic drivers used by the Borland compilers to
provide graphics output for different graphics cards. They are loaded
dynamically. The exact format is not known to me ...
OFFSET Count TYPE Description
0000h 4 char ID='FBGD'
0004h 1 dword ID=08080808h
used to backspace over ID if typing the file
0008h ? char Driver ID string, terminated with #26
EXTENSION:BGI
OCCURENCES:PC
PROGRAMS:Borland Pascal, Borland C, Turbo Pascal
--------I-BMP-------------------------------
The BMP files are the way, Windows stores bit mapped images. The BMP image
data is bit packed but every line must end on a dword boundary - if thats not
the case, it must be padded with zeroes. BMP files are stored bottom-up,
that means that the first scan line is the bottom line. The BMP format has four
incarnations, two under Windows (new and old) and two under OS/2, all are
described here.
OFFSET Count TYPE Description
0000h 2 char ID='BM' - BitMap
OS/2 also supports the following IDs :
ID='BA' - Bitmap Array
ID='CI' - Color Icon
ID='CP' - Color Pointer (mouse cursor)
ID='IC' - Icon
ID='PT' - Pointer (mouse cursor)
0002h 1 dword Filesize of whole file
0006h 4 byte reserved
000Ah 1 dword Offset of bitmap in file
="BOF"
000Eh 1 dword Length of BitMapInfoHeader
The BitMapInfoHeader starts directly after
this header.
12 - OS/2 1.x format
40 - Windows 3.x format
64 - OS/2 2.x format
0012h 1 dword Horizontal width of bitmap in pixels
0016h 1 dword Vertical width of bitmap in pixels
001Ah 1 word Number of planes
001Ch 1 word Bits per pixel ( thus the number of colors )
="BPP"
001Eh 1 dword Compression type, see ALGRTHMS.txt for descrip-
tion of the different types
0 - none
1 - RLE 8-bit/Pixel
2 - RLE 4-bit/Pixel
0022h 1 dword Size of picture in bytes
0026h 1 dword Horizontal resolution
002Ah 1 dword Vertical resolution
002Ah 1 dword Number of used colors
002Ah 1 dword Number of important colors
0036h ? rec Definition of N colors
N=1 shl "BPP"
1 byte Blue component
1 byte Green component
1 byte Red component
1 byte Filler
"BOF" ? byte Image data
EXTENSION:BMP,RLE,LGO
OCCURENCES:PC
PROGRAMS:Windows,Paintbrush
REFERENCE:DDJ0994
VALIDATION:
SEE ALSO:rDIB
--------I-CEG-------------------------------
The CEG (Continous Edge Graphic)-format is a raw picture format used by the
Edsun cards with CEG-chips which provide some better look through anti-aliasing
or something like that. The header before the data looks like this :
OFFSET Count TYPE Description
0000h 1 word Version number of the CEG-format
0002h 9 char ID='Edsun CEG'
000Bh 1 byte Number of pixels per byte
000Ch 9 byte Reserved
0015h 80 char ASCIIZ copyright notice for the image
0065h 1 byte CEG-revision number (1)
0066h 1 byte Used CEG-mode (0..15)
0067h 1 Word Number of pixels per line
0069h 1 word Number of lines
006Ah 1 byte Old VGA-mode
006Bh 1 byte VGA Data flag :
0 - VGA registers are invalid
1 - VGA registers are valid
006Ch 92 byte VGA register data
00C2h 256 rec VGA palette entries
1 byte Value for red
1 byte Value for green
1 byte Value for blue
03C2h 1 word Year of file creation
03C4h 1 byte Day of file creation
03C5h 1 byte Month of file creation
03C6h 1 byte Hour of file creation
03C7h 1 byte Minute of file creation
03C8h 1 byte Second of file creation
03C9h 24 byte Reserved for future use
EXTENSION:???
OCCURENCES:PC
PROGRAMS:???
--------a-CEL-------------------------------
CEL files contain one or more frames of image data used by the Autodesk
Animator and Animator Pro animation pakages. Both Animator Pro and the original
Animator produce CEL files, but each uses a different file format.
--- Animator Pro CEL Files
An Animator Pro CEL file is identical to a FLC file in all respects. A CEL file
should have a Celdata block in the file prefix block which describes the x,y
placement of the CEL. If the Celdata placement block is not present, assume a
placement of 0,0.
--- Original Animator CEL Files
The original Animator also produced CEL files. These were still-picture files,
not the multi-frame files Animator Pro now uses. A CEL file from the original
Animator is identical to a PIC file from the original Animator in all respects.
EXTENSION:CEL
OCCURENCES:PC
PROGRAMS:Autodesk Animator
SEE ALSO:FLIc,FLC,PIC
VALIDATION:
--------F-CHR-------------------------------
The CHR files are scalable fonts used by the Borland graphics interface
(BGI) to display fonts in graphics mode.
OFFSET Count TYPE Description
0000h 4 char ID='PK',08,08
0004h 4 char ID='BGI '
0008h ? char Font description, terminated with #26
0008h 1 word Headersize
+???? ="SIZ"
4 char Internal font name
1 word Font file size in bytes
1 byte Font driver major version
1 byte Font driver minor version
1 word 0100h
"SIZ" word Zeroes to pad out the header
0080h 1 char Signature byte, '+' means stroke font
0081h 1 word Number of chars in font file
="NUM"
0083h 1 byte undefined
0084h 1 byte ASCII value of first char in file
0085h 1 word Offset to stroke definitions
0087h 1 byte Scan flag ??
0088h 1 byte Distance from origin to top of capital
0089h 1 byte Distance from origin to baseline
008Ah 1 byte Distance from origin to bottom descender
008Bh 4 char Four character name of font
0090h "NUM" word Offsets to character definitions
0090h+ "NUM" byte Width table for the characters
"NUM"*2
0090h+ Start of character definitions
"NUM"*3
The individual character definitions consist of a variable number of words
describing the operations required to render a character. Each word
consists of an (x,y) coordinate pair and a two-bit opcode, encoded as shown
here:
Byte 1 7 6 5 4 3 2 1 0 bit #
op1 <seven bit signed X coord>
Byte 2 7 6 5 4 3 2 1 0 bit #
op2 <seven bit signed Y coord>
Opcodes
op1=0 op2=0 End of character definition.
op1=0 op2=1 Do scan
op1=1 op2=0 Move the pointer to (x,y)
op1=1 op2=1 Draw from current pointer to (x,y)
EXTENSION:CHR
OCCURENCES:PC
PROGRAMS:Borland Pascal, Borland C
REFERENCE:BGIKIT.ZIP
SEE ALSO:
VALIDATION:
--------M-CMF-G-----------------------------
The CMF files are music files used by the SoundBlaster sound card family. The
Creative Labs Music Format might be proprietary, the info is guesswork.
OFFSET Count TYPE Description
*********
EXTENSION:CMF
OCCURENCES:PC
PROGRAMS:PLAYCMF.EXE
--------I-COL-------------------------------
A COL file stores the rgb values for entries in the color palette. Both
Animator Pro and the original Animator produce COL files, but the formats
are different. To process a COL file for input, check the file size. If it is
exactly 768 bytes, the file is an original Animator COL file. If the file is
any other size, it is an Animator Pro COL file - which makes identification
almost impossible.
Animator Pro COL Files do have a 8-byte header :
OFFSET Count TYPE Description
0000h 1 dword File size, including this header
0004h 1 word ID=0B123h
0006h 1 word Version, currently 0
Following the file header are palette entries in rgbrgb... order. Each of
the r, g, and b components is a single byte in the range of 0-255. Generally,
there will be data for 256 palette entries, but this cannot be assumed.
The actual number of palette entries is ((size-8)/3); if this value is not
an even multiple of three, the file is corrupted.
Original Animator COL Files
A COL file created by the original Animator is exactly 768 bytes
long. There is no file header or other control information in
the file.
EXTENSION:COL
OCCURENCES:PC
PROGRAMS:Autodesk Animator, Autodesk Animator Pro
SEE ALSO:FLIc,FLT
--------E-COM-------------------------------
The COM files are raw binary executables and are a leftover from the old CP/M
machines with 64K RAM. A COM program can only have a size of less than one
segment (64K), including code and static data since no fixups for segment
relocation or anything else is included. One method to check for a COM file is
to check if the first byte in the file could be a valid jump or call opcode, but
this is a very weak test since a COM file is not required to start with a jump
or a call. In principle, a COM file is just loaded at offset 100h in the segment
and then executed.
OFFSET Count TYPE Description
0000h 1 byte ID=0E9h
ID=0EBh
Those are not safe ways to determine wether a
file is a COM file or not, but most COM files
start with a jump.
Further information not available.
EXTENSION:COM
OCCURENCES:PC
SEE ALSO:EXE,MZ EXE,NE EXE
--------E-CORE-G----------------------------
The core images are dumps of the system core from different unix machines (as
far as I gather). Info comes from a magic file - so this is only good for
identification. What you would do with a core image on a foreign machine, eludes
me anyway.
OFFSET Count TYPE Description
0000h 4 char ID='core'
0176h 1 word Executable type
0176h - 386 executable
************
>566 short 072401 386 executable
>564 short 0535 B370 executable
>564 short 056401 B370 executable
>564 short 0530 B370 executable
>564 short 054001 B370 executable
>564 short 0537 XA370 executable
>564 short 057401 XA370 executable
>564 short 055001 XA370 executable
>564 short 0532 XA370 executable
EXTENSION:???
OCCURENCES:Unix flavours
PROGRAMS:N/A
SEE ALSO:
--------D-CPI-G-----------------------------
The DOS CPI files are data files which are loaded by the country drivers of
MS-DOS. The information comes from a magic file, which makes it good for
identification only.
OFFSET Count TYPE Description
0000h 9 char ID=255,'FONT ',0
EXTENSION:CPI
OCCURENCES:PC
PROGRAMS:MS-DOS
--------X-DBase II-O------------------------
The DBase II file format.
The dBASE II file header has a fixed size of 521 bytes.
OFFSET Count TYPE Description
0000h 1 byte dBASE version, 02h = dBASE II
0001h 1 word Number of data records in file
="NDR"
0003h 1 byte Month of last update
0004h 1 byte Day of last update
0005h 1 byte Year of last update
0006h 1 word Size of each data record
="DRS"
0008h 64 rec Field descriptors
11 char ASCIIZ field name, 0Dh as first
char indicates end of list.
1 char Data type
'C' - Char
'N' - Numerical
'L' - Logical
1 byte Field length
1 word Field data address ( set in RAM )
1 byte Number of decimal places
0208h 1 byte If 0Dh, then all 32 field descriptors were used;
otherwise 00h
EXTENSION:DBF
OCCURENCES:PC
PROGRAMS:DBase
SEE ALSO:DBASE III,XBase
VALIDATION:FileSize="NDR"*"DRS"+0208h
--------X-DBase III-------------------------
DBASE - File header structure (DBASE III)
OFFSET Count TYPE Description
0000h 1 byte dBASE version,
03h = dBASE III w/o *.DBT
83h = dBASE III w *.DBT
0001h 1 byte Month of last update
0002h 1 byte Day of last update
0003h 1 byte Year of last update
0004h 1 dword Number of data records in file
="NDR"
0008h 1 word Header size
="HSZ"
000Ah 1 word Data record size
="DRS"
000Ch 12 byte reserved
0020h ? rec Field descriptors
The list of field descriptors is
terminated with a terminator
byte 0Dh.
11 char ASCIIZ field name
1 char Data type
'C' - Char
'D' - Date
'L' - Logical
'M' - Memo
'N' - Numerical
1 dword Field data address ( set in RAM )
1 byte Field length
1 byte Number of decimal places
14 byte reserved
EXTENSION:DBF
SEE ALSO:DBASE II,XBase
OCCURENCES:PC
PROGRAMS:DBase
VALIDATION:FileSize="NDR"*"DRS"+"HSZ"
--------X-DBASE IV--------------------------
**** Description missing ****
EXTENSION:DBF,DBT
OCCURENCES:PC
PROGRAMS:DBase 4.0, Clipper
REFERENCE:
SEE ALSO:DBASE II,DBASE III,XBase
VALIDATION:
--------M-DMF-------------------------------
The Digital Music Files are high quality MOD style files with up to
32 channels/1024 beats per track. The X-Tracker by the demo group
D-Lusion produces this format. In general, the format is well organised
due to the ID/Blocklength structure wich makes downward compatibility to
older version files easy, but the Version 4 (current version) of the file
format, produced by X-Tracker 0.30ß still requires some manual scanning for
the next ID which I regard as not so nice. Version 5 of the format has the
[SEQU] block length fixed, but the [SMPD] block has the length 0.
The file consists of several blocks, each with a 4 char (dwordint) ID tag
and a length of the record data. The main file header looks as follows :
OFFSET Count TYPE Description
0000h 4 char ID='DDMF'
0004h 1 byte Version id.
4 -> XTracker 0.30ß
0005h 8 char Tracker name, e.g. 'XTRACKER', 'HACKTRAK' :-)
000Dh 30 char Song name (ASCIIZ?)
002Bh 20 char Name of composer (ASCIIZ?)
0049h 1 byte Day of creation
004Ah 1 byte Month of creation
004Bh 1 byte Year of creation
The other headers have the standard skip record format, in this section
named DMFblock. The offsets start _after_ this header record :
OFFSET Count TYPE Description
0000h 4 char Record tag (see below)
0004h 1 dword Size of data bedwording to this tag
DMFblock [INFO]
Contains some message in ASCII. Length of the message is the size of
the record.
DMFblock [CMSG]
Contains the message the composer wants to bring to us. After the ID
record, another fill byte preceeds the real message !
OFFSET Count TYPE Description
0000h 1 byte Junk byte
0001h ? char Composer message
DMFblock [SEQU]
Contains the information necessary for sequencing the different tracks.
OFFSET Count TYPE Description
0000h 1 word Song loop start
0002h 1 word Song loop end
0004h ? word Sequencer data
DMFblock [PATT]
This block contains the information about the different patterns and tracks.
0000h 1 word Maximum pattern (=Songlength)
="MPT"
0004h 1 byte Number of channels of this song (<= 16)
0005h "MPT" rec Pattern data.
1 byte Track entries. (<=32)
="TET"
How many tracks this pattern has.
XTracker allows a different number of
tracks for each pattern.
1 byte Beat information
High nibble : Ticks per beat
Low nibble : Beats per measure
1 word Maximum number of ticks (<=512)
1 dword Number of bytes to skip for the
next pattern information.
? rec Track data stream
1 byte Global track effect
1 byte Global track data (only if global
effect >0 !!!)
"TET" rec
1 byte Information byte, bitmapped
For each bit set in the info byte, one
or two data byte(s) follow. This info byte
must not always be there, see below. For
effects, 2 bytes follow.
0 - reserved
1 - Volume effect
2 - Note effect
3 - Instrument effect
4 - Volume set
5 - Note set
6 - Instrument set
7 - Counter to next information byte.
Not set means, that next info byte
follows in 1 tick, unit is in
ticks.
The maximum number of effects is 3 at a time,
the maxximum size of a track information is
11 bytes (with info=0FEh).
? rec Effect bytes
1 byte Effect number
1 byte Effect data
? byte Set data
** Here follows the pattern data, but it's too late today **
DMFblock [INST]
This block contains the information about the instrument data.
If this block does not exists, then the instrument numbers in the patterns
point directly to the samples in the [SMPI] block.
OFFSET Count TYPE Description
0000h 1 byte Number of instruments
0001h ? rec Instrument information block
30 char The name of the instrument
1 byte Instrument type, bitmapped
0 - Instrument type
1 - Instrument type
00 - Sample in [SMPI] block
01 - MIDI device
10 - FM instrument
11 - reserved
2 - valid attack envelope
3 - sustain on
4 - reserved
5 - reserved
6 - reserved
7 - reserved
1 byte Range entries
="REN"
Like the GF1 patterns, an instrument can
consist of several samples.
"REN" rec Range definition
1 byte Sample to be played in this range
1 byte Length of this range in half tone steps up
6 byte Not yet defined 6-point envelope
DMFblock [SMPI]
This block contains the information about the samples stored in the file.
OFFSET Count TYPE Description
0000h 1 byte Number of samples (<= 250)
="NUM"
"NUM" rec Sample record
1 byte Length of sample name
? char Name of the sample
1 dword Length of sample in bytes
1 dword Start of sample loop
1 dword End of sample loop
1 word Frequency used for C-3
1 byte Volume for sample
0 - Don't change current volume
otherwise volume (linear scale)
1 byte Sample type, bitmapped
0 - not looped/looped
1 - 8/16-bit sample
(16-bit not supported with X-Tracker v0.30)
2,3 - Pack type :
00 - unpacked, signed sample
01 - pack type 0
10 - pack type 1
11 - pack type 2
4-6 - reserved, set to zero
7 - Sample stored in dmf/bib
1 word reserved, set to zero
1 dword crc32 of sample to identify samples
in BIB.
DMFblock [SMPD]
This block contains the sample data (raw or packed, see [SMPI] block) in the
following format :
<SampleLength> <SampleData> <SampleLength> <SampleData> etc.
OFFSET Count TYPE Description
0000h 1 dword Length of the following sample
? byte Sample data (might be packed)
DMFBlock [ENDE]
This block serves as a end of file marker and can be used for validation.
Note that the four ID characters are _not_ followed by a length dword ! Each DMF
file simply ends with the four characters 'ENDE'.
EXTENSION:DMF
OCCURENCES:PC
PROGRAMS:X-TRACKER,PLAY_DMF
SEE ALSO:
VALIDATION:
--------?-DMS-------------------------------
The DMS (Digital Music System??) are some other files I found on a
mixed system CD, so I include them in my listing. They are Amiga files,
so here's the call to the Amiga folks again.
OFFSET Count TYPE Description
0000h 4 char ID="DMS!"
EXTENSION:DMS
OCCURENCES:Amiga
--------A-DWC-?-----------------------------
The DWC archives seem to be a relict from ancient computing times - I've never
seen any program that dealt with them or could create them. They are yet
included in this compilation for reasons I don't know. But maybe one of you
stumbles over such a file, he might find this documentation helpful.
The DWC archives consist of single file entries with one archive trailer. The
archive entries seem to be at the start of the archive, but maybe they are
stored at the end of the archive, before the trailer. Each file header has the
following format :
OFFSET Count TYPE Description
0000h 13 char Name of the original file in ASCIIZ.
000Dh 1 dword Size of the original file
0011h 1 dword MS-DOS date and time of the original file
0015h 1 dword Size of the compressed file
0019h 1 dword Offset of compressed data in archive file
001Dh 3 byte reserved
0020h 1 byte Method :
1 - crunched
2 - stored
The trailer at the end of each archive has the following format :
OFFSET Count TYPE Description
0000h 1 word Length of trailer (=27)
0002h 1 word Size of the directory entries (=34)??
0004h 16 byte reserved
0014h 1 dword Count of the directory entries
0018h 3 char ID="DWC"
EXTENSION:DWC??
OCCURENCES:PC??
PROGRAMS:DWC.EXE??
--------S-EFE-------------------------------
The EFE files are instrument files for the Ensoniq sampler system.
Further information wanted.
EXTENSION:EFE
SEE ALSO:GKH,INS
--------E-EXE-X-----------------------------
Different types of executables have emerged on the Intel DOS related platforms -
but all contain at least a stub MZ Exe before their actual EXE body...
SEE ALSO:MZ EXE,NE EXE
--------M-FAR-------------------------------
The Fandarole composer is a 16 channel composer created by the group
Digital Infinity / Daniel Potter for digital music in module style.
The Fandarole modules have the following format :
OFFSET Count TYPE Description
0000h 4 char ID='FAR',254
0004h 40 char Song name
002Ch 3 char ID=13,10,26
This ID makes it possible to see the song name
by simply typing the .far file.
002Fh 1 word Remaining header size
0031h 1 byte Version number as BCD,
high nibble = major version
low nibble = minor version
0032h 16 byte Channel on/off map
<> 0 means that channel is used
0042h 1 rec Editing data.
This data is not necessary for playback,
but is stored by the composer for resume of
edit.
1 byte Current octave
1 byte Current voice
1 byte Current row
1 byte Current pattern
1 byte Current order
1 byte Current sample
1 byte Current volume
1 byte Current top of screen display
1 byte Current editing area
0=samples,
1=patterns,
2=orders
1 byte Current tempo (default tempo)
004Ch 16 byte Panning map for each channel, 0=left,15=right
005Ch 1 byte Marked block start
005Dh 1 byte Marked block end
005Eh 1 byte Grid granularity
005Fh 1 byte Edit mode
0060h 1 word Song text length
="STL"
0062h "STL" char Song text
0062h+ 256 byte Order bytes for pattern ordering
"STL"
0162h+ 1 byte Number of stored patterns
"STL"
0163h+ 1 byte Song length in patterns
"STL"
0164h+ 1 byte Loop position. This is the restart position
"STL" if the end of the song is reached.
0165h+ 256 word Length of each pattern. The number of rows in
"STL" each pattern is ( this word-2 )/(16*4)
After this block, there might be additional data in the future (see remaining
header size, above), after that, the pattern data follows.
The pattern data :
OFFSET Count TYPE Description
0000h 1 byte Length of pattern in rows
="LIR"
0001h 1 byte Tempo for this pattern - Unsupported,
use not recommended
0002h 4*"LIR" rec Note data for each pattern in 4 channels
1 byte Note value (Octave*12+Note)+1
0 means no note
1 byte Sample number
1 byte Volume byte. The volume is stored reversed,
the lower nibble is the major volume, the lower
nibble is the minor volume adjust.
1 byte Effect byte. Upper nibble is effect, lower
nibble is data. (see table 0004)
(Table 0004)
Note Effects in FAR-modules
01 - Pitch adjust
02 - Pitch adjust
03 - Portamento to note
04 - Retrigger note data times for one bar
05 - Set vibrato depth
06 - Vibrato
07-0C - ?Possibly undefined?
0D - Fine tune tempo down 128/Tempo
0E - Fine tune tempo up 128/Tempo
0F - Tempo, notes per second = 32/Tempo
After the pattern data, the sample map follows. This is an array of 64 bits
(eight bytes), each set bit corresponds to a sample record stored in the file,
each zero bit means that the corresponding record is not stored in the file.
OFFSET Count TYPE Description
0000h 8 byte Sample flags, see above
After the sample flags, the samples themselves are stored in the FSM format,
except for the ("FSM",254) header. They follow header-data-header-data-etc.,
see the FSM entry for further information.
EXTENSION:FAR
OCCURENCES:PC
PROGRAMS:Fandarole Composer
REFERENCE:
SEE ALSO:FSM
VALIDATION:
--------a-FLT-------------------------------
The FLC files are files created by the Autodesk Animator Pro and contain
animations. The FLC files are a superset of those created by the Autodesk
Animator (FLIc files). In some cases, new data fields or compression methods
were added. The FLC files use a hierarchical block oriented structure and blocks
are a combination of control information and data. The file consists of one
header followed by data blocks. It is possible that new types of blocks not
described in this document will be added to animation files in the future. It is
recommended that you quietly ignore unknown block types you encounter during
animation playback. The size fields in the block headers make it easy to skip
an entire unrecognized block.
The FLC files consist of one 128-byte header block and one or more of the
following blocks :
The prefix block, if present, contains Animator Pro settings information,
CEL placement information, and other auxiliary data.
A frame block exists for each frame in the animation. In addition, a ring frame
follows all the animation frames. Each frame block contains color palette
information and/or pixel data.
The ring frame contains delta-compressed information to loop from the last frame
of the flic back to the first. It can be helpful to think of the ring frame as a
copy of the first frame, compressed in a different way. All flic files will
contain a ring frame, including a single-frame flic.
The FLC file header
OFFSET Count TYPE Description
0000h 1 dword The size of the whole animation file, including
the size of this header.
0004h 1 word ID=0AF12h
0006h 1 word Number of frames in this animation, not
including the ring frame. FLC files have a
maximum length of 4000 frames.
0008h 1 word Screen width in pixels
000Ah 1 word Screen height in pixels
000Ch 1 word Bits per pixel (always 8)
000Eh 1 word Flags - bitmapped
0 - Ring frame not written / ring frame written
1 - Flic header not updated / updated
2-15 - reserved
0010h 1 dword Delay between frames in miliseconds.
0014h 1 word reserved
0016h 1 dword MS-DOS date and time of file creation (see table 0009)
001Ah 1 dword Serial number of the Animator Pro program used
to create the file. If the file was created
with the FlicLib development kit, this value
equals 0464c4942h ("FLIB").
001Eh 1 dword MS-DOS date and time of last modification (see table 0009)
0022h 1 dword Serial number of program that made the last
modification. See Serial Number.
0026h 1 word X-axis aspect ratio of the file
0028h 1 word Y-axis aspect ratio of the file
(320x200 = 6:5)
002Ah 38 byte reserved (0)
0050h 1 dword Offset from begin of file to the first
animation frame block.
0054h 1 dword Offset from begin of file to the second
animation frame block. This value is used
when looping the animation.
0058h 40 byte reserved (0)
Each subblock in the animation file has an identical header structure,
which is formatted like this :
0000h 1 dword The size of the whole block and all subordinate
blocks including the size of this header
0004h 1 word Block ID, varies depending on the block type.
0006h 1 word Number of subordinate blocks in this block.
including the ring frame. FLC files have a
maximum length of 4000 frames.
0008h 8 byte reserved(0)
Immediately after the header there may be an optional prefix block, which is
used to store additional data which is not directly involved in animation
playback.
The prefix block has the usual header with an ID of 0F100h.
The prefix block should only be created by the Animator Pro programs and never
by any other software, it is to be ignored by other software.
The FLC frame blocks contain the information to convert the current frame into
the next frame; they have an ID of 0F1FAh. Directly after the frame header,
there are the subordinate data blocks - if the subblock count is 0 this means,
that the current frame is identical to the previous frame, only the appropriate
delay has to be made.
The data blocks have a different header format :
OFFSET Count TYPE Description
0000h 1 dword Size of this block, including header size
0004h 1 word Data type identifier :
4 - 256-level color palette info
7 - Word-oriented delta compression
11 - 64-level color palette info
12 - Byte-oriented delta compression
13 - Entire frame is color index 0
15 - Byte run length compression
16 - No compression
18 - Postage stamp sized image
0006h ? byte Color or pixel data
The following sections describe each of these data encoding methods in detail.
--- Block Type 4 (FLI_COLOR256) - 256-Level Color
The data in this block is organized in packets. The first word following the
block header is a count of the number of packets in the blocks. Each packet
consists of a one-byte color index skip count, a one-byte color count and three
bytes of color information for each color defined.
At the start of the block, the color index is assumed to be zero. Before
processing any colors in a packet, the color index skip count is added to the
current color index. The number of colors defined in the packet is retrieved.
A zero in this byte indicates 256 colors follow. The three bytes for each color
define the red, green, and blue components of the color in that order. Each
component can range from 0 (off) to 255 (full on). The data to change colors
2,7,8, and 9 would appear as follows:
2 ; two packets
2,1,r,g,b ; skip 2, change 1
4,3,r,g,b,r,g,b,r,g,b ; skip 4, change 3
--- Block Type 11 (FLI_COLOR) - 64-Level Color
This block is identical to FLI_COLOR256 except that the values for the red,
green and blue components are in the range of 0-63 instead of 0-255, i.e. in
native VGA values which can be written to the VGA without modification.
--- Block Type 13 (FLI_BLACK) - No Data
This block has no data following the header. All pixels in the frame are set to
color index 0.
--- Block Type 16 (FLI_COPY) - No Compression
This block contains an uncompressed raw image of the frame, from upper left
to the lower right, storing each line sequentially. This type of block is
created when the preferred compression method (SS2 or BRUN) generates more
data than the uncompressed frame image; a relatively rare situation.
--- Block Type 15 (FLI_BRUN) - Byte Run Length Compression
This block contains the entire image in a compressed format. Usually this block
is used in the first frame of an animation, or within a postage stamp image
block.
The data is organized in lines. Each line contains packets of compressed pixels.
The first line is at the top of the animation, followed by subsequent lines
moving downward. The number of lines in this block is given by the height of the
animation.
The first byte of each line is a count of packets in the line. This value is
ignored, it is a holdover from the original Animator. It is possible to generate
more than 255 packets on a line. The width of the animation is now used to drive
the decoding of packets on a line; continue reading and processing packets until width
pixels have been processed, then proceed to the next line.
Each packet consist of a type/size byte, followed by one or more pixels. If the
high bit of the packet type is set, the remaining 7 bits are a count of pixels
to be copied from the packet to the animation image, otherwise the next byte
contains a single pixel which is to be replicated; The lower 7 bits are the
number of times the pixel is to be replicated.
--- Block Type 12 (FLI_LC) - Byte Aligned Delta Compression
This block contains the differences between the previous frame and this frame. This compression method was used by the original
Animator, but is not created by Animator Pro. This type of block can appear in
an Animator Pro file, however, if the file was originally created by Animator,
then some (but not all) frames were modified using Animator Pro.
The first word following the block header contains the position of the first
line in the block. This is a count of lines (down from the top of the image)
which are unchanged from the prior frame. The second word contains the number of
lines in the block. The data for the lines follows these two words.
Each line begins with two bytes. The first byte contains the starting x position
of the data on the line, and the second byte the number of packets for the line.
Unlike BRUN compression, the packet count is significant (because this
compression method is only used on 320x200 flics).
Each packet consists of a single byte column skip, followed by a packet type/
size byte, which has the reverse meaning of in the block type 15.
--- Block Type 7 (FLI_SS2) - Word Aligned Delta Compression
This format contains the differences between consecutive frames. This is the
format most often used by Animator Pro for frames other than the first frame of
an animation. It is similar to the line coded delta (LC) compression, but is
word oriented instead of byte oriented. The data is organized into lines and
each line is organized into packets.
The first word in the data following the block header contains the number of
lines in the block. Each line can begin with some optional words that are used
to skip lines and set the last byte in the line for animations with odd widths.
These optional words are followed by a count of the packets in the line. The
line count does not include skipped lines.
The high order two bits of the word is used to determine the contents of
the word :
Bit 15 Bit 14 Meaning
0 0 The word contains the packet count. The packets follow
this word. The packet count can be zero; this occurs
when only the last pixel on a line changes.
1 0 The low order byte is to be stored in the last byte of
the current line. The packet count always follows this
word.
1 1 The word contains a line skip count. The number of
lines skipped is given by the absolute value of the
word. This word can be followed by more skip counts,
by a last byte word, or by the packet count.
The packets in each line are similar to the packets for the line coded block.
The first byte of each packet is a column skip count. The second byte is a
packet type. If the packet type is positive, the packet type is a count of words
to be copied from the packet to the animation image. If the packet type is
negative, the packet contains one more word which is to be replicated. The
absolute value of the packet type gives the number of times the word is to be
replicated. The high and low order byte in the replicated word do not
necessarily have the same value.
--- Block Type 18 (FLI_PSTAMP) - Postage Stamp Image
This block type holds a postage stamp - a reduced-size image - of the frame. It
generally appears only in the first frame block within a flic file. When
creating a postage stamp, Animator Pro considers the ideal size to be 100x63
pixels. The actual size will vary as needed to maintain the same aspect ratio as
the original.
The pixels in a postage stamp image are mapped into a six-cube color space,
regardless of the color palette settings for the full frame image. A six-cube
color space is formed as follows:
start at palette entry 0
for red = 0 thru 5
for green = 0 thru 5
for blue = 0 thru 5
palette_red = (red * 256)/6
palette_green = (green * 256)/6
palette_blue = (blue * 256)/6
move to next palette entry
end for blue
end for green
end for red
Any arbitrary rgb value (where each component is in the range of 0-255) can be
mapped into the six-cube space using the formula:
((6*red)/256)*36 + ((6*green)/256)*6 + ((6*blue)/256)
The full postage stamp block header is defined as follows:
Offset Length Name Description
OFFSET Count TYPE Description
0000h 1 dword Size of this block, including header size
0004h 1 word ID=18
0006h 1 word Height of the postage stamp image
0008h 1 word Width of the image
000Ah 1 word Color translation type :
1 - six-cube color space
Immediately following this header is the postage stamp data. The data is
formatted as a block with standard size/type header. The type will be one of:
15 FPS_BRUN Byte run length compression
16 FPS_COPY No compression
18 FPS_XLAT256 Six-cube color xlate table
The FPS_BRUN and FPS_COPY types are identical to the FLI_BRUN and FLI_COPY
encoding methods described above.
The FPS_XLAT256 type indicates that the block contains a 256-byte color
translation table instead of pixel data. To process this type of postage stamp,
read the pixel data for the full-sized frame image, and translate its pixels
into six-cube space using a lookup in the 256-byte color translation table. This
type of postage stamp appears when the size of the animation frames is smaller
than the standard 100x63 postage stamp size.
*************
TWE - Tween Data Files
A TWE file holds information about a tweening operation set up
via the Tween menus. The information includes the starting and
ending shapes, and the optional userD specified links between the
shapes. Animator Pro creates tween files.
A TWE file begins with an 8-byte header defined as follows:
Offset Length Name Description
0 2 magic File format identifier. Always hex 1995.
2 2 version The file format version; always zero.
4 4 tcount The number of tween shapes in the file;
always 2.
8 8 reserved Unused space; set to zeroes.
16 4 linkcount The number of link entries in the file.
Immediately following the file header are the link entries. If
the linkcount value is zero there are no links. Each link entry
is a pair of 32-bit integers. The first value in each pair is the
index of the point in the first shape, and the second value is
the index of the point in the ending shape. (IE, a link value of
2,7 says to link the second starting-shape point to the seventh
ending-shape point.)
Following the link entries is the data block that describes the
starting shape, then the data block that describes the ending
shape. The format of these blocks is identical to that of the
polygon (PLY) file, including file header data. In other words,
they appear as if a pair of polygon files are embedded in the
tween file at this point.
**********
OPT - Optics Menu Settings Files
An OPT file holds information about an optics operation set up
via the Optics menus. Both Animator Pro and the original
Animator create OPT files. The file format is the same for both.
An OPT file starts with a 4-byte header, as follows:
Offset Length Name Description
0 2 magic File type identifier. Always hex 1A3F.
2 2 count Number of records in the file.
Following the file header are optics records of 50 bytes each. A
record is generated for each click on CONTINUE MOVE in the OPTICS
menu. The move records are formatted as follows:
Offset Length Name Description
0 4 link In the file, this field is always zero.
In memory, it's a pointer to the next
move record.
4 6 spincenter The x,y,z coordinates of the spin
center point; three 16-bit values.
10 6 spinaxis The x,y,z coordinates of the spin axis;
three 16-bit values.
16 6 spinturns The x,y,z coordinates of the spin turns;
three 16-bit values.
22 4 spininter Intermediate turns. Two 16-bit values.
These are values for a conjugation matrix
that corresponds to spin axis.
26 6 sizecenter The x,y,z coordinates of the size
center point; three 16-bit values.
32 2 xmultiplier Determines (along with xdivisor)
how to scale along x dimension.
34 2 xdivisor Determines (along with xmultiplier) how
to scale along x dimension.
36 2 ymultiplier Determines (along with ydivisor)
how to scale along y dimension.
38 2 ydivisor Determines (along with ymultiplier) how
to scale along y dimension.
40 2 bothmult Like xmultiplier, but applied to both
dimensions.
42 2 bothdiv Like xdivisor, but applied to both
dimensions.
44 6 linearmove The x,y,z offset for a linear move;
three 16-bit values.
EXTENSION:FLT
OCCURENCES:PC
PROGRAMS:Autodesk Animator Pro
REFERENCE:
SEE ALSO:FLIc
VALIDATION:
--------a-FLIc------------------------------
The Flic file format was one of the first graphic animation formats on the PC.
It was developed by <> and used by the Autodesk Animator. It provides relatively
fast animation in 320x200 resolution modes. The FLI use delta updating for
faster animation. The single block information and prefix blocks are missing for
the FLI files, see the FLT format for a discussion.
OFFSET Count TYPE Description
0000h 1 dword Size of the FLIc file
0004h 1 word ID=0AF11h
AF11h means the file is a FLI file.
0006h 1 word Number of frames
0008h 1 word Width of displayed animation
000Ah 1 word Height of displayed animation
000Ch 1 word Number of used colors ("Depth")
000Eh 1 word Flags (=0003h)
0010h 1 dword Frame speed in sec/1024 **
0014h 1 word reserved
0016h 1 dword Date/Time of creation in DOS format (see table 0009)
001Ah 1 dword Creator
001Eh 1 dword Date/Time of last change in DOS format (see table 0009)
0022h 1 dword Serial number? of changer
0026h 1 word X-Aspect ratio of animation
0028h 1 word Y-Aspect ratio of animation
002Ah 38 byte reserved
0052h 1 dword Offset of frame 1 in file
0056h 1 dword Offset of frame 2 in file
005Ah 40 byte reserved
EXTENSION:FLI,FLT
OCCURENCES:PC
REFERENCE:DDJ0693
PROGRAMS:Autodesk Animator
SEE ALSO:QuickTime,AVI,FLT
--------D-FON-?-----------------------------
The Telix .FON files are the telephone books Telix uses to store numbers in.
The format is for Telix 3.22
OFFSET Count TYPE Description
0000h 1 dword ID=2E2B291Ah
0004h 1 word Version info (=1)
0006h 1 word Number of entries in directory (count from 1)
0007h 1 char ?will be used for encryption?
Currently 0
0008h 55 byte reserved
0040h ? rec Actual phonebook entry
25 char Name (0 terminated)
17 char Phone number (0 terminated)
1 byte Baud rate (see table 0006)
1 byte Parity type (see table 0007)
1 byte Data bits (7 or 8)
1 byte Stop bits (1 or 2)
12 char Script file name
6 char Date of last call in ASCII
1 word Number of total calls
1 byte Terminal type (see table 0008)
1 byte Protocol
1 byte Flags, bitmapped
0 - Local echo on / off
1 - add linefeeds on / off
2 - backspace is destructive on / off
3 - backspace sends DEL / sends BS
4 - strip high bits on / off
5-7 - reserved
1 word unknown
1 byte Dial prefix index
14 char Password
(Table 0006)
Baud rate tables for Telix
0 = 300 baud
1 = 1200 baud
2 = 2400 baud
3 = 4800 baud
4 = 9600 baud
5 = 19200 baud
6 = 38400 baud
7 = 57600 baud
8 = 115200 baud
(Table 0007)
Parity types for Telix
0 = None
1 = Even
2 = Odd
3 = Mark
4 = Space
(Table 0008)
Terminal types for Telix
0 = TTY
1 = ANSI-BBS
2 = VT102
3 = VT52
4 = AVATAR
5 = ANSI
EXTENSION:FON
OCCURENCES:PC
PROGRAMS:Telix v3.22
REFERENCE:
SEE ALSO:
VALIDATION:
--------M-FPT-------------------------------
The Fandarole Pattern files are used by the Fandarole Composer to store
single patterns in a file.
OFFSET Count TYPE Description
0000h 4 char ID='FPT',254
0004h 32 char ASCII pattern name
0024h 3 char ID=10,13,26
0027h 1 word Remaining size of file (size of pattern)
0029h 1 byte Break location (length of pattern)
002Ah 1 byte reserved
002Bh ? byte Pattern in raw format like in the .FAR file
EXTENSION:FAR,FPT
OCCURENCES:PC
PROGRAMS:Fandarole Composer
SEE ALSO:FAR,FSM
VALIDATION:
--------S-FSM-------------------------------
The .FSM files are samples to be used for module style music with the
Fandarole Composer. Currently only samples of up to 64K length are supported,
altough the header reserves a dword for the sample size.
OFFSET Count TYPE Description
0000h 4 char ID='FSM',254
0004h 32 char ASCII name of sample
0024h 3 char ID=10,13,26
0027h 1 dword Length of sample (<=64K)
0028h 1 byte Fine tune value for sample
(currently unsupported)
0029h 1 byte Sample volume
(currently unsupported)
002Ah 1 dword Start of sample loop
002Dh 1 dword End of sample loop. If the sample is not set
to loop (see below) this should be set to the
end of the sample.
0032h 1 byte Sample type, bitmapped
0 - 8-bit/16-bit sample
1-7 - reserved
0033h 1 byte Loop mode, ?bit mapped?
0-2 - reserved
3 - loop off/loop on
4-7 - reserved
0034h ? byte Sample data in signed format
EXTENSION:FSM
OCCURENCES:PC
PROGRAMS:Fandarole Composer
REFERENCE:
SEE ALSO:FAR,USM
VALIDATION:
--------S-GF1 PATCH-------------------------
The GF1 Patch files are multipart sound files for the Gravis Ultrasound
sound card to emulate MIDI sounds in high quality. Each Patch can consist
of many samples (for example, a string ensemble consists of Violin, Viola,
Cello, Bass) which are played depending on the note to play. A patch can also
contain a part to be played before the loop and a part to be played after
the tone has been released.
OFFSET Count TYPE Description
0000h 12 char ID='GF1PATCH110'
000Ch 10 char Manufacturer ID
0018h 60 char Description of the contained Instruments
or copyright of manufacturer.
0054h 1 byte Number of instruments in this patch
0055h 1 byte Number of voices for sample
0056h 1 byte Number of output channels
(1=mono,2=stereo)
0057h 1 word Number of waveforms
0059h 1 word Master volume for all samples
005Bh 1 dword Size of the following data
0060h 36 byte reserved
Following this header, the instruments with their headers follow.
An instrument header contains the name and other data about one
instrument contained within the patch.
OFFSET Count TYPE Description
0000h 1 word Instrument number. ?Maybe the MIDI instrument
number?. In the Gravis patches, this is 0, in
other patches, I found random values.
0002h 16 char ASCII name of the instrument.
0012h 1 dword Size of the whole instrument in bytes.
0016h 1 byte Layers. Needed for whatever.
0017h 40 byte reserved
About the patch, I don't know anything. Maybe somebody could enlighten me.
Each patch record has the following format :
OFFSET Count TYPE Description
0000h 7 char Wave file name
0007h 1 byte Fractions
0008h 1 dword Wave size.
Size of the wave digital data
000Ch 1 dword Start of wave loop
0010h 1 dword End of wave loop
0012h 1 word Sample rate of the wave
0014h 1 word Minimum frequency to play the wave
0016h 1 word Maximum frequency to play the wave
0018h 1 dword Original sample rate of the wave data
001Ch 1 int Fine tune value for the wave
001Eh 1 byte Stereo balance, values unknown**
001Fh 6 byte Filter envelope rate
0025h 6 byte Filter envelope offse
002Bh 1 byte Tremolo sweep
002Ch 1 byte Tremolo rate
002Dh 1 byte Tremolo depth
002Fh 1 byte Vibrato sweep
0030h 1 byte Vibrato rate
0031h 1 byte Vibrato depth
0032h 1 byte Wave data, bitmapped
0 - 8/16 bit wave data
1 - signed/unsigned data
2 - de/enable looping
3 - no/has bidirectional looping
4 - loop forward/backward
5 - Turn envelope sustaining off/on
6 - Dis/Enable filter envelope
7 - reserved
0033h 1 int Frequency scale, whatever that means
0035h 1 word Frequency scale factor
0037h 36 byte Reserved
EXTENSION:PAT
OCCURENCES:PC
PROGRAMS:Patch Maker
SEE ALSO:VOC,WAVe
--------I-GIF-------------------------------
The Graphics Interchange Format (tm) was created by Compuserve Inc. as a
standard for the storage and transmission of raster-based graphics information,
i.e. images. A GIF file may contain several images, which are to be displayed
overlapping and without any delay betwenn the images. The image data itself is
compressed using a LZW scheme.
The GIF file consists of a global GIF header, one or more image blocks and
optionally some GIF extensions.
OFFSET Count TYPE Description
0000h 6 char ID='GIF87a', ID='GIF89a'
This ID may be viewed as a version number
0006h 1 word Image width
0008h 1 word Image height
000Ah 1 byte bit mapped
0-2 - bits per pixel -1
3 - reserved
4-6 - bits of color resolution
7 - Global color map follows image descriptor
000Bh 1 byte Color index of screen background
000Ch 1 byte reserved
The global color map immediately follows the screen descriptor and has the size
(2**BitsPerPixel), and has the RGB colors for each color index. 0 is none, 255
is full intensity. The bytes are stored in the following format :
OFFSET Count TYPE Description
0000h 1 byte Red component
0001h 1 byte Green component
0002h 1 byte Blue component
After the first picture, there may be more pictures attached in the file
whic overlay the first picture or parts of the first picture. The Image
Descriptor defines the actual placement and extents of the following image
within the space defined in the Screen Descriptor. Each Image Descriptor is
introduced by an image separator character. The role of the Image Separator
is simply to provide a synchronization character to introduce an Image
Descriptor, the image separator is defined as ",", 02Ch, Any characters
encountered between the end of a previous image and the image separator
character are to be ignored.
The format of the Image descriptor looks like this :
OFFSET Count TYPE Description
0000h 1 char Image separator
ID=','
0001h 1 word Left offset of image
0003h 1 word Upper offset of image
0005h 1 word Width of image
0007h 1 word Height of image
0009h 1 byte Palette description - bitmapped
0-2 - Number of bits per pixel-1
3-5 - reserved (0)
6 - Interlaced / sequential image
7 - local / global color map, ignore bits 0-2
To provide for some possibility of an extension of the GIF files, a special
extension block introducer can be added after the GIF data block. The block has
the following structure :
OFFSET Count TYPE Description
0000h 1 char ID='!'
0001h 1 byte Extension ID
0002h ? rec
1 word Byte count
? byte Extra data
????h 1 byte Zero byte count - terminates extension block.
EXTENSION:GIF
OCCURENCES:PC
PROGRAMS:CSHOW.EXE
SEE ALSO:
VALIDATION:
--------A-GZIP------------------------------
The GNU ZIP program is an archive program mostly for the UNIX machines developed
by the GNU project.
OFFSET Count TYPE Description
0000h 2 char ID='!',139
0002h 1 byte Method :
0-7 - reserved
8 - deflated
0003h 1 byte File flags :
0 - ASCII-text
1 - Multi-part file
2 - Name present
3 - Comment present
4 - Encrypted
5-8 - reserved
0004h 1 dword File date and time (see table 0009)
0008h 1 byte Extra flags
0009h 1 byte Target OS :
0 - DOS
1 - Amiga
2 - VMS
3 - Unix
4 - ????
5 - Atari
6 - OS/2
7 - MacOS
10 - TOPS-20
11 - Win/32
EXTENSION:ZIP
PROGRAMS:GNU gzip
--------S-GKH-------------------------------
The GKH files are disk images of the Ensoniq EPS sampler system.
Further information is missing.
EXTENSION:GKH
SEE ALSO:EFE,INS
--------a-GRASPRT GL-G----------------------
The .GL animation files are graphic animations, some just .GIF files, others
mini-movies, used mostly for x-rated adult animations. The format of the files
is plain guesswork by me. The analyzed file did not include any animations but
only .GIF files and two text files which seemed to be the animation script.
There is no safe way of identifying a file as a GL animation, maybe except for
adding the subfile sizes and the header size and then check if this matches the
file size.
OFFSET Count TYPE Description
0000h 1 word Length of header, excluding this word
="HLN"
0002h ? rec The directory entries for each file
1 dword Offset of the stored file
12 char DOS file name of the stored file
0002h+ 1 dword Length of the first stored file
"HLN"
? byte The first file
The other files follow in similar
manner, length->file->length->file
EXTENSION:GL
OCCURENCES:PC
PROGRAMS:GRASPRT
--------?-GRIB------------------------------
The GRIB weather product information files just might be some satellite images
or something else. I have only seen this signature in a magic file and further
informations about the format is not known to me.
OFFSET Count TYPE Description
0000h 4 char ID='GRIB'
EXTENSION:???
OCCURENCES:???
PROGRAMS:???
--------I-HSI1------------------------------
The HSI1 images are a JPEG derivative made by Handmade Software for their
Image Alchemy package.
OFFSET Count TYPE Description
0000h 4 char ID='HSI1'
EXTENSION:JPG
OCCURENCES:PC,SUN
PROGRAMS:Image Alchemy
REFERENCE:
SEE ALSO:JPEG
VALIDATION:
--------A-HYP-------------------------------
The Hyper archiver is a very fast compression program by P. Sawatzki and K.P.
Nischke, which uses LZW compression techniques for compression. It is not very
widespread - in fact, I've yet to see a package distributed in this format.
OFFSET Count TYPE Description
0000h 1 byte ID=1Ah
0001h 2 char Compression method
"HP" - compressed
"ST" - stored
0003h 1 byte Version file was compressed by in BCD
0004h 1 dword Compressed file size
0008h 1 dword Original file size
000Ch 1 dword MS-DOS date and time of file (see table 0009)
0010h 1 dword CRC-32 of file
0014h 1 byte MS-DOS file attribute
0015h 1 byte Length of filename
="LEN"
0016h "LEN" char Filename
EXTENSION:HYP
OCCURENCES:PC
PROGRAMS:HYPER.EXE
--------f-IFF-M-----------------------------
The IFF format is comparable to the RIFF file format, but it
uses Motorola byte ordering. After the FORM header, the different
records follow. Each record has a header ID of 4 bytes and then following
the size of the data (in Motorola byte ordering). Each IFF record starts on
an even byte boundary, that means if the record length is odd, you will have
to skip one more byte to get the next record.
OFFSET Count TYPE Description
0000h 4 char ID='FORM'
0004h 1 dword Size of the whole IFF block
0008h 4 char Type of the IFF file
Each IFF record has the following format :
OFFSET Count TYPE Description
0000h 4 char ID
0004h 1 dword Blocksize
0008h ? byte Block data, depends on block type.
OCCURENCES:Amiga,PC
SEE ALSO:8SVX,LBM,RIFF
--------S-INS-------------------------------
The INS files are instrument files for the Ensoniq sampler system.
Further information wanted.
EXTENSION:INS
SEE ALSO:EFE,GKH
--------I-JPEG-G----------------------------
The JPEG image standard is a standard for lossy (but efficient) image
compression made by the ???? Group. The endianness of the JPEG files is unknown
to me, there seem to exist both types of JPEG files.
OFFSET Count TYPE Description
0000h 1 dword ID=FFD9FFE0h
ID=FFD8FFE0h
Big endian JPEG file (Intel)
ID=E0FFD8FFh
Little endian JPEG file (Motorola)
EXTENSION:JPG
OCCURENCES:PC,Amiga,SUN
PROGRAMS:
REFERENCE:
SEE ALSO:HSI1
VALIDATION:
--------I-LBM-M-----------------------------
The LBM/ILBM format is used by Deluxe Paint to store bitmap images. It
uses the IFF file format and Motorola byte order.
FORMblock [BMHD]
This block contains the information about the image.
OFFSET Count TYPE Description
0000h 1 word The image width (x-axis)
0002h 1 word The image height (y-axis)
0004h 1 dword reserved
0008h 1 byte Bits per pixel
0009h 1 byte ??reserved??
FORMblock [BODY]
This block contains the (compressed) image data... ****
FORMblock [CRGN]
This block contains palette information for a range of palette entries.
OFFSET Count TYPE Description
FORMblock [TINY]
This block contains a small image used for previewing.
OFFSET Count TYPE Description
EXTENSION:IFF,LBM
OCCURENCES:AMIGA,PC
PROGRAMS:Deluxe Paint
REFERENCE:???
SEE ALSO:IFF
--------A-LBR-------------------------------
The LBR files consist of a direcotry and one or more "members". The directory
contains from 4 to 256 entries and each entry describes one member.
The first directory entry describes the directory itself. All space allocations
are in terms of sectors, where a sector is 128 bytes long. Four directory
entries fit in one sector thus the number of directory entries is always evenly
divisible by 4. Different types of LBR files exist, all versions are discussed
here, the directory entry looks like this :
OFFSET Count TYPE Description
0000h 1 byte File status :
0 - active
254 - deleted
255 - free
0001h 11 char File name in FCB format (8/3, blank padded),
directory name is blanks for old LU,
ID='********DIR'
for LUPC
000Ch 1 word Offset to file data in sectors
000Eh 1 word Length of stored data in sectors
For the LUPC program, the remaining 16 bytes are used like this :
OFFSET Count TYPE Description
0000h 8 char ASCII date of creation (MM/DD/YY)
0008h 8 char ASCII time of creation (HH:MM:SS)
For the LU86 program, the remaining 16 bytes are used like this :
OFFSET Count TYPE Description
0000h 1 word CRC-16 or 0
0002h 1 word Creation date in CP/M format
0004h 1 word Creation time in DOS format
0006h 1 word Date of last modification, CP/M format
0008h 1 word Time of last modification, DOS format
000Ah 1 byte Number of bytes in last sector
000Bh 5 byte reserved (0)
EXTENSION:LBR
OCCURENCES:PC,CP/M
PROGRAMS:LU.COM, LUU.COM, LU86.COM
SEE ALSO:
--------A-LZH-------------------------------
The LHArc/LHA archiver is a multi platform archiver made by Haruyasu Yoshizaki,
which has a relatively good compression. It uses more or less the same
technology like the ZIP programs by Phil Katz. There was a hack named "ICE",
which had only the graphic characters displayed on decompression changed.
OFFSET Count TYPE Description
0000h 1 byte Size of archived file header
0001h 1 byte Checksum of remaining bytes
0002h 3 char ID='-lh'
ID='-lz'
0005h 1 char Compression methods used (see table 0005)
0006h 1 char ID='-'
0007h 1 dword Compressed size
000Bh 1 dword Uncompressed size
000Fh 1 dword Original file date/time (see table 0009)
0013h 1 word File attribute
0015h 1 byte Filename / path length in bytes
="LEN"
0016h "LEN" char Filename / path
0018h 1 word CRC-16 of original file
+"LEN"
(Table 0005)
LHArc compression types
"0" - No compression
"1" - LZW, 4K buffer, Huffman for upper 6 bits of position
"2" - unknown
"3" - unknown
"4" - LZW, Arithmetic Encoding
"5" - LZW, Arithmetic Encoding
"s" - LHa 2.x archive?
"\" - LHa 2.x archive?
"d" - LHa 2.x archive?
EXTENSION:LZH,ICE
OCCURENCES:PC
PROGRAMS:LHArc.EXE, LHA.EXE
--------M-MIDI-M----------------------------
The MIDI file format is used to store MIDI song data on disk. The discussed
version of the MIDI file spec is the approved MIDI Manufacturers' Associations
format version 0.06 of (3/88). The contact address is listed in the adresses
file. Version 1.0 is technically identical but the description has been
rewritten. The description was made by Dave Oppenheim, most of the text was
taken right out of his document.
MIDI files contain one or more MIDI streams, with time information for
each event. Song, sequence, and track structures, tempo and time signature
information, are all supported. Track names and other descriptive information
may be stored with the MIDI data. This format supports multiple tracks and
multiple sequences so that if the user of a program which supports multiple
tracks intends to move a file to another one, this format can allow that to
happen.
The MIDI files are block oriented files, currently only 2 block types are
defined, header and track data. Opposed to the IFF and RIFF formats, no
global header is given, so that the validation must be done by adding the
different block sizes.
A MIDI file always starts with a header block, and is followed by one or
more track block.
The format of the header block :
OFFSET Count TYPE Description
0000h 4 char ID='MThd'
0004h 1 dword Length of header data (=6)
0008h 1 word Format specification
0 - one, single multi-channel track
1 - one or more simultaneous tracks
2 - one or more sequentially independent
single-track patterns
000Ah 1 word Number of track blocks in the file
000Ch 1 int Unit of delta-time values.
If negative :
Absolute of high byte :
Number of frames per second.
Low byte :
Resolution within one frame
If positive, division of a quarter-note.
The track data format :
The MTrk block type is where actual song data is stored. It is simply a
stream of MIDI events (and non-MIDI events), preceded by delta-time
values.
Some numbers in MTrk blocks are represented in a form called a variable-
length quantity. These numbers are represented 7 bits per byte, most
significant bits first. All bytes except the last have bit 7 set, and
the last byte has bit 7 clear. If the number is between 0 and 127, it
is thus represented exactly as one byte. Since this explanation might not be
too clear, some exapmles :
Number (hex) Representation (hex)
00000000 00
00000040 40
0000007F 7F
00000080 81 00
00002000 C0 00
00003FFF FF 7F
001FFFFF FF FF 7F
08000000 C0 80 80 00
0FFFFFFF FF FF FF 7F
The largest number which is allowed is 0FFFFFFF so that the variable-
length representation must fit in 32 bits in a routine to write
variable-length numbers.
Each track block contains one or more MIDI events, each event consists of
a delta-time and the number of the event. The delta-time is stored as a
variable-length quantity and represents the time to delay before the following
event. A delta-time of 0 means, that the event occurs simultaneous with the
previous event or occurs right at the start of a track. The delta-time unit is
specified in the header block.
Format of track information block :
OFFSET Count TYPE Description
0000h 4 char ID='MTrk'
0004h 1 dword Length of header data
0008h ? rec <delta-time>, <event>
Three types of events are defined, MIDI event, system exclusive event and
meta event. The first event in a file must specify status; delta-time itself
is not an event. Meta events are non-MIDI informations.
The format of the meta event :
OFFSET Count TYPE Description
0000h 1 byte ID=FFh
0001h 1 byte Type (<=128)
0002h ? ? Length of the data, 0 if no data
stored as variable length quantity
? byte Data
A few meta-events are defined. It is not required for every program to support
every meta-event. Meta-events initially defined include:
FF 00 02 ssss Sequence Number
This optional event, which must occur at the beginning of a track,
before any nonzero delta-times, and before any transmittable MIDI
events, specifies the number of a sequence.
FF 01 len text Text Event
Any amount of text describing anything. It is a good idea to put a text
event right at the beginning of a track, with the name of the track, a
description of its intended orchestration, and any other information
which the user wants to put there. Programs on a computer which does not
support non-ASCII characters should ignore those characters with the hi-bit
set. Meta event types 01 through 0F are reserved for various types of text
events, each of which meets the specification of text events(above) but is
used for a different purpose:
FF 02 len text Copyright Notice
Contains a copyright notice as printable ASCII text. The notice should
contain the characters (C), the year of the copyright, and the owner of
the copyright. If several pieces of music are in the same MIDI file,
all of the copyright notices should be placed together in this event so
that it will be at the beginning of the file. This event should be the
first event in the first track block, at time 0.
FF 03 len text Sequence/Track Name
If in a format 0 track, or the first track in a format 1 file, the name
of the sequence. Otherwise, the name of the track.
FF 04 len text Instrument Name
A description of the type of instrumentation to be used in that track.
FF 05 len text Lyric
A lyric to be sung. Generally, each syllable will be a separate lyric
event which begins at the event's time.
FF 06 len text Marker
Normally in a format 0 track, or the first track in a format 1 file.
The name of that point in the sequence, such as a rehearsal letter or
section name ("First Verse", etc.).
FF 07 len text Cue Point
A description of something happening on a film or video screen or stage
at that point in the musical score ("Car crashes into house", "curtain
opens", "she slaps his face", etc.)
FF 2F 00 End of Track
This event is not optional. It is included so that an exact ending
point may be specified for the track, so that it has an exact length,
which is necessary for tracks which are looped or concatenated.
FF 51 03 tttttt Set Tempo, in microseconds per MIDI quarter-note
This event indicates a tempo change. Another way of putting
"microseconds per quarter-note" is "24ths of a microsecond per MIDI
clock". Representing tempos as time per beat instead of beat per time
allows absolutely exact dword-term synchronization with a time-based sync
protocol such as SMPTE time code or MIDI time code. This amount of
accuracy provided by this tempo resolution allows a four-minute piece at
120 beats per minute to be accurate within 500 usec at the end of the
piece. Ideally, these events should only occur where MIDI clocks would
be located Q this convention is intended to guarantee, or at least
increase the likelihood, of compatibility with other synchronization
devices so that a time signature/tempo map stored in this format may
easily be transferred to another device.
FF 54 05 hr mn se fr ff SMPTE Offset
This event, if present, designates the SMPTE time at which the track
block is supposed to start. It should be present at the beginning of
the track, that is, before any nonzero delta-times, and before any
transmittable MIDI events. The hour must be encoded with the SMPTE
format, just as it is in MIDI Time Code. In a format 1 file, the SMPTE
Offset must be stored with the tempo map, and has no meaning in any of
the other tracks. The ff field contains fractional frames, in 100ths of
a frame, even in SMPTE-based tracks which specify a different frame
subdivision for delta-times.
FF 58 04 nn dd cc bb Time Signature
The time signature is expressed as four numbers. nn and dd represent
the numerator and denominator of the time signature as it would be
notated. The denominator is a negative power of two: 2 represents a
quarter-note, 3 represents an eighth-note, etc. The cc parameter
expresses the number of MIDI clocks in a metronome click. The bb
parameter expresses the number of notated 32nd-notes in a MIDI quarter-
note (24 MIDI Clocks).
FF 59 02 sf mi Key Signature
sf = -7: 7 flats
sf = -1: 1 flat
sf = 0: key of C
sf = 1: 1 sharp
sf = 7: 7 sharps
mi = 0: major key
mi = 1: minor key
FF 7F len data Sequencer-Specific Meta-Event
Special requirements for particular sequencers may use this
event type: the first byte or bytes of data is a manufacturer ID.
However, as this is an interchange format, growth of the spec proper is
preferred to use of this event type. This type of event may be used by
a sequencer which elects to use this as its only file format;
sequencers with their established feature-specific formats should
probably stick to the standard features when using this format.
The system exclusive event is used as an escape to specify arbitrary bytes
to be transmitted. The system exclusive event has two forms, to compensate
for some manufacturer-specific modes, the F7h event is used if a F0h is to
be transmitted. Each system exclusive event must end with an F7h event.
The format of a system exclusive event :
OFFSET Count TYPE Description
0000h 1 byte ID=F0h,ID=F7h
0001h ? ? Length as variable length qty.
? byte bytes to be transmitted
EXTENSION:MID,MIDI
OCCURENCES:PC,MAC
PROGRAMS:Cubase
VALIDATION:
--------M-MOD-M-----------------------------
The Protracker composer is a composer for digital music. The MOD files are a
quasi standard for digital music, all words are in Motorola byte order. The
original MOD format allowed only 4 digital channels and 15 instruments, the
specification became enlarged (maybe by Mahoney and Kaktus??) to 4 channels and
31 instruments. Check the file at offset 1080d for the signatures 'M.K', '4CHN',
'6CHN','8CHN','FLT4','FLT8. If you find any of them, the module uses 31
instruments. With rising sound quality on the PC and other platforms, the old
MODule format has been replaced by numerous other formats. The 4/15 format has
almost become extinct. Below, only the 4/31 format is descripted.
The digital sample data is signed (two's complement) as necessary for the Amiga,
the sample data immediately follows the pattern data.
Maybe this is not valid for some 8CHN files; One of the two I have, uses Intel
byte ordering and unsigned samples.
OFFSET Count TYPE Description
0000h 20 char Song title, padded with spaces
0014h 31 rec Sample description record
For original MOD files, the number of
instruments would be 15.
22 char Sample name, padded with zeroes to
full length.
2 word Sample length / 2. Needs to be multiplied
by 2 to get the actual length. If the sample
length is greater than 8000h, the sample
is bigger than 64k.
1 byte Sample finetune. Only the lower nibble is
valid. Fine tune table :
0 1 2 3 4 5 6 7 8 9 A B C D E F
0 +1 +2 +3 +4 +5 +6 +7 -8 -7 -6 -5 -4 -3 -2 -1
1 byte Sample volume (0-40h)
1 word Sample loop start / 2
1 word Sample loop length / 2
950d 1 byte Song length in patterns (0-80h)
951d 1 byte Restart byte for song looping (Noisetracker?)
952d 128 byte Pattern play sequences
1080d 4 char ID='M.K.', ID='4CHN',ID='6CHN',ID='8CHN'
ID='4FLT',ID='8FLT'
If this position contains 'M.K.','8CHN',
'4CHN','6CHN','FLT4' or 'FLT8' the module
has 31 instruments.
1084d ? rec Patterns
Each pattern has 64 rows.
Depending on the number of channels, each row
has from 4 to 8 notes. The channel count is
determined by the ID. (see table 0005)
The number of patterns is the highest pattern
number stored in the pattern list.
Each note has four bytes. Four notes make
up a track in a four channel MOD file. Each
track is saved sequentially :
byte 0-3 4-7 8-11 12-15
Chn #1 Chn #2 Chn #3 Chn #4
byte 16-19 20-23 24-27 28-31
Chn #1 Chn #2 Chn #3 Chn #4
1 word Instrument / period
The instrument number is in bits 12-15, the
12-bit period in bits 0-11.
1 byte Upper nibble : Lower 4 bits of the instrument,
Lower nibble : Special effect command.
1 byte Special effects data
(Table 0005)
Protracker 16 note conversion table / MOD Period table
+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+
PT16 : I 1I 2I 3I 4I 5I 6I 7I 8I 9I 10I 11I 12I
MOD : I 1712I 1616I 1524I 1440I 1356I 1280I 1208I 1140I 1076I 1016I 960I 906I
Note : I C-0I C#0I D-0I D#0I E-0I F-0I F#0I G-0I G#0I A-0I A#0I B-0I
+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+
I 13I 14I 15I 16I 17I 18I 19I 20I 21I 22I 23I 24I
I 856I 808I 762I 720I 678I 640I 604I 570I 538I 508I 480I 453I
I C-1I C#1I D-1I D#1I E-1I F-1I F#1I G-1I G#1I A-1I A#1I B-1I
+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+
I 25I 26I 27I 28I 29I 30I 31I 32I 33I 34I 35I 36I
I 428I 404I 381I 360I 339I 320I 302I 285I 269I 254I 240I 226I
I C-2I C#2I D-2I D#2I E-2I F-2I F#2I G-2I G#2I A-2I A#2I B-2I
+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+
I 37I 38I 39I 40I 41I 42I 43I 44I 45I 46I 47I 48I
I 214I 202I 190I 180I 170I 160I 151I 143I 135I 127I 120I 113I
I C-3I C#3I D-3I D#3I E-3I F-3I F#3I G-3I G#3I A-3I A#3I B-3I
+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+
I 49I 50I 51I 52I 53I 54I 55I 56I 57I 58I 59I 60I
I 107I 101I 95I 90I 85I 80I 75I 71I 67I 63I 60I 56I
I C-4I C#4I D-4I D#4I E-4I F-4I F#4I G-4I G#4I A-4I A#4I B-4I
+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+-----+
EXTENSION:MOD,module
OCCURENCES:AMIGA,PC
PROGRAMS:DMP,ModEdit
VALIDATION:NONE
--------I-MSK-------------------------------
The MSK files are mask files used by the Autodesk Animator and Animator Pro
packages. Two types of MSK files exist. The Animator Pro version is simply a PIC
file with the depth 1; A MSK file created by the original Animator is exactly
8000 bytes long. There is no file header or other control information in the
file. It contains the image bit map, 1 bit per pixel, with the leftmost pixels
packed into the high order bits of each byte. The size of the image is fixed at
320x200. The image is stored left-to-right, top-to-bottom.
EXTENSION:MSK
OCCURENCES:PC
PROGRAMS:Autodesk Animator
SEE ALSO:PIC,FLIc
--------M-MTM-------------------------------
The MTM format is generated by the Multi Track Module tracker by the demo group
Renaissance. The tracker features up to 32 channel digital music. Instead of
saving whole patterns, the tracker only saves the different tracks and the data
which tracks should be played together at which time, thus saving some pattern
space.
OFFSET Count TYPE Description
0000h 3 char ID='MTM'
0003h 1 byte Version data
upper nibble is major version number
lower nibble is minor version number
0004h 20 char ASCIIZ song name
0018h 1 word Number of saved tracks.
="NOT"
001Ah 1 byte Highest pattern number saved
="NOP"
001Bh 1 byte Last order number to play(=Songlength-1)
001Ch 1 word Length of extra comment field in bytes
="XSZ"
001Eh 1 byte Number of samples
="NOS"
001Fh 1 byte Attribute byte (currently defined as 0)
0020h 1 byte Beats per track
0021h 1 byte Number of tracks
0022h 32 byte Pan positions of the voices
(0=left, 15=right??)
0042h "NOS" rec Instrument data
22 char Sample name
1 dword Sample length in bytes
1 dword Start of sample loop in bytes
1 dword End of sample loop in bytes
1 byte Fine tune value for sample
1 byte Default volume for sample
1 byte Attribute byte, bit mapped
0 0=8 bit sample,1=16 bit sample
1-7 undefined (set to zero)
0042h+ 128 byte Pattern order data
"NOS"*37
01C2h+
"NOS"*37 "NOT" rec Track data
Each track is saved independently and has
the size of exactly 192 bytes. Each track
is arranged as 64 3-byte notes with the
following format :
64*3 byte BYTE 0 BYTE 1 BYTE 2
ppppppii iiiieeee aaaaaaaa
p = pitch value (0=no pitch stated)
i = instrument number (0=no instrument number)
e = effect number
a = effect argument
The effects are the standard Protracker
effects.
01C2h+ ("NOP"+1)*32 word Track sequencing data
"NOS"*37+ This is the list of which track is used
"NOT"*192 as which voice in each pattern. One track
can be part of many patterns, the drums
for example.
Track 0 is never saved but is always
considered as an empty track. That means
that counting really starts at one.
The data is organized in sets of 32 voices.
The first word contains the information which
track is used in pattern 0, voice 0. The next
word is for pattern 0, voice 1, etc., this
is repeated for each pattern, 32 words for each
saved pattern.
01C2h+ "XSZ" char Extra comment field. This contains some
"NOS"*37+ message or something.
"NOT"*192+
("NOP"+1)*32*2
01C2h+ ? byte Raw sample data(unsigned).
"NOS"*37+
"NOT"*192+
("NOP"+1)*32*2+
"XSZ"
EXTENSION:MTM
SEE ALSO:MOD
OCCURENCES:PC
PROGRAMS:MMEDIT,DMP
VALIDATION:
--------M-MTS-------------------------------
The Master Tracker program by the french demo group Arkham is a tracker
for AdLib, SB and speaker - the further limits of this tracker are unknowm
to me.
OFFSET Count TYPE Description
0000h 6 char ID="MTRAC "
0006h 20 char Song name, zero padded
EXTENSION:MST
OCCURENCES:PC
PROGRAMS:Master Tracler v1.0
SEE ALSO:MOD
--------E-MZ EXE----------------------------
The old EXE files are the EXE files executed directly by MS-DOS. They were a
major improvement over the old 64K COM files, since EXE files can span multiple
segments. An EXE file consists of three different parts, the header, the
relocation table and the binary code.
The header is expanded by a lot of programs to store their copyright information
in the executable, some extensions are documented below.
The format of the header is as follows :
OFFSET Count TYPE Description
0000h 2 char ID='MZ'
ID='ZM'
0002h 1 word Number of bytes in last 512-byte page
of executable
0004h 1 word Total number of 512-byte pages in executable
(including the last page)
0006h 1 word Number of relocation entries
0008h 1 word Header size in paragraphs
000Ah 1 word Minimum paragraphs of memory allocated in
addition to the code size
000Ch 1 word Maximum number of paragraphs allocated in
addition to the code size
000Eh 1 word Initial SS relative to start of executable
0010h 1 word Initial SP
0012h 1 word Checksum (or 0) of executable
0014h 1 dword CS:IP relative to start of executable
(entry point)
0018h 1 word Offset of relocation table;
40h for new-(NE,LE,LX,W3,PE etc.) executable
001Ah 1 word Overlay number (0h = main program)
Following are the header expansions by some other prorams like TLink, LZExe and
other linkers, encryptors and compressors; all offsets are relative to the start
of the whole header :
---new executable
OFFSET Count TYPE Description
001Ch 4 byte ????
0020h 1 word Behaviour bits ??
0022h 26 byte reserved (0)
003Ch 1 dword Offset of new executable header from start of
file (or 0 if plain MZ executable)
---Borland TLINK
OFFSET Count TYPE Description
001Ch 2 byte ?? (apparently always 01h 00h)
001Eh 1 byte ID=0FBh
001Fh 1 byte TLink version, major in high nybble
0020h 2 byte ??
---old ARJ self-extracting archive
OFFSET Count TYPE Description
001Ch 4 char ID='RJSX' (older versions)
new signature is 'aRJsf'" in the first 1000
bytes of the file)
---LZEXE compressed executable
OFFSET Count TYPE Description
001Ch 2 char ID='LZ'
001Eh 2 char Version number :
'09' - LZExe 0.90
'91' - LZExe 0.91
---PKLITE compressed executable
OFFSET Count TYPE Description
001Ch 1 byte Minor version number
001Dh 1 byte Bit mapped :
0-3 - major version
4 - Extra compression
5 - Multi-segment file
001Eh 6 char ID='PKLITE'
---LHarc 1.x self-extracting archive
OFFSET Count TYPE Description
001Ch 4 byte unused???
0020h 3 byte Jump to start of extraction code
0023h 2 byte ???
0025h 12 char ID='LHarc's SFX '
--LHA 2.x self-extracting archive
OFFSET Count TYPE Description
001Ch 8 byte ???
0024h 10 char ID='LHa's SFX '
For version 2.10
ID='LHA's SFX ' (v2.13)
For version 2.13
---LH self-extracting archive
OFFSET Count TYPE Description
001Ch 8 byte ???
0024h 8 byte ID='LH's SFX '
---TopSpeed C 3.0 CRUNCH compressed file
OFFSET Count TYPE Description
001Ch 1 dword ID=018A0001h
0020h 1 word ID=1565h
---PKARC 3.5 self-extracting archive
OFFSET Count TYPE Description
001Ch 1 dword ID=00020001h
0020h 1 word ID=0700h
---BSA (Soviet archiver) self-extracting archive
OFFSET Count TYPE Description
001Ch 1 word ID=000Fh
001Eh 1 byte ID=A7h
---LARC self-extracting archive
OFFSET Count TYPE Description
001Ch 4 byte ???
0020h 11 byte ID='SFX by LARC '
After the header, there follow the relocation items, which are used to span
multpile segments. The relocation items have the following format :
OFFSET Count TYPE Description
0000h 1 word Offset within segment
0002h 1 word Segment of relocation
To get the position of the relocation within the file, you have to compute the
physical adress from the segment:offset pair, which is done by multiplying the
segment by 16 and adding the offset and then adding the offset of the binary
start. Note that the raw binary code starts on a paragraph boundary within the
executable file. All segments are relative to the start of the executable in
memory, and this value must be added to every segment if relocation is done
manually.
EXTENSION:EXE,OVR,OVL
OCCURENCES:PC
PROGRAMS:MS-DOS
REFERENCE:Ralf Brown's Interrupt List
SEE ALSO:COM,EXE,NE EXE
--------E-NE EXE----------------------------
The NE EXE files are the new exe files used by windows and OS/2 executables.
They contain a small MZ EXE which prints "This program requires Microsoft
Windows" or something similar but Some files contain both DOS and Windows
versions of the executable. The position of the new EXE header can be found
in the old exe header - see the MZ EXE topic for further information. All
offsets within this header are from the start of the header if not noted
otherwise.
OFFSET Count TYPE Description
0000h 2 char ID='NE'
0002h 1 byte Linker major version
0003h 1 byte Linker minor version
0004h 1 word Offset of entry table (see below)
0006h 1 word Length of entry table in bytes
0008h 1 dword File load CRC (0 in Borland's TPW)
000Ch 1 byte Program flags, bitmapped :
0-1 - DGroup type :
0 - none
1 - single shared
2 - multiple
3 - (null)
2 - Global initialization
3 - Protected mode only
4 - 8086 instructions
5 - 80286 instructions
6 - 80386 instructions
7 - 80x87 instructions
000Dh 1 byte Application flags, bitmapped
0-2 - Application type
1 - Full screen (not aware of
Windows/P.M. API)
2 - Compatible with Windows/P.M. API
3 - Uses Windows/P.M. API
3 - OS/2 family application
4 - reserved?
5 - Errors in image/executable
6 - "non-conforming program" whatever
7 - DLL or driver (SS:SP info invalid, CS:IP
points at FAR init routine called with
AX=module handle which returns AX=0000h
on failure, AX nonzero on successful
initialization)
000Eh 1 byte Auto data segment index
0010h 1 word Initial local heap size
0012h 1 word Initial stack size
0014h 1 dword Entry point (CS:IP),
CS is index into segment table
0018h 1 dword Initial stack pointer (SS:SP)
SS is index into segment table
001Ch 1 word Segment count
001Eh 1 word Module reference count
0020h 1 word Size of nonresident names table in bytes
0022h 1 word Offset of segment table (see below)
0024h 1 word Offset of resource table
0026h 1 word Offset of resident names table
0028h 1 word Offset of module reference table
002Ah 1 word Offset of imported names table
(array of counted strings, terminated with a
string of length 00h)
002Ch 1 dword Offset from start of file to nonresident
names table
0030h 1 word Count of moveable entry point listed in
entry table
0032h 1 word File alignment size shift count
0 is equivalent to 9 (default 512-byte pages)
0034h 1 word Number of resource table entries
0036h 1 byte Target operating system
0 - unknown
1 - OS/2
2 - Windows
3 - European MS-DOS 4.x
4 - Windows 386
5 - BOSS (Borland Operating System Services)
0037h 1 byte Other OS/2 EXE flags, bitmapped
0 - Long filename support
1 - 2.x protected mode
2 - 2.x proportional fonts
3 - Executable has gangload area
*****
0038h WORD offset to return thunks or start of gangload area
003Ah WORD offset to segment reference thunks or length of gangload area
003Ch WORD minimum code swap area size
003Eh 2 BYTEs expected Windows version (minor version first)
EXTENSION:DLL,EXE,FOT
OCCURENCES:PC
PROGRAMS:
REFERENCE:Windows 3.1 SDK Programmer's Reference, Vol 4.
SEE ALSO:EXE,MZ EXE
--------D-NG-G------------------------------
Information about this format comes only from a magic file, thus is only good
for file identification. I did not test it, since I don't have any NG files.
The Norton Guides are a popup help program for the IBM PCs which provide instant
help anywhere...
OFFSET Count TYPE Description
0000h 2 char ID='NG'
0002h 1 dword ID=0
EXTENSION:NG
OCCURENCES:PC
PROGRAMS:NG.EXE
SEE ALSO:TPH,HLP
--------B-OBJ-------------------------------
Most of the description was taken from the Microsoft Product Support
Services Application Note SS0288. The .OBJ files are binary files used
by compilers to link in precompiled code. They contain symbol and relocation
information necessary to link the data and code contained in the files. The
.OBJ files have no common header which makes a validation or identification
guesswork at best. The .OBJ files consist of at least one record, each of the
following type :
OFFSET Count TYPE Description
0000h 1 byte Record type (see below)
0001h 1 word Record length
="LEN"
0003h "LEN" byte Record data
0003h 1 byte Checksum or 0
+"LEN" (that much for validation)
The maximum size of the entire record (unless otherwise noted for specific
record types) is 1024 bytes.
For LINK386, the format is determined by the least-significant bit
of the Record Type field. An odd Record Type indicates that certain
numeric fields within the record contain 32-bit values; an even
Record Type indicates that those fields contain 16-bit values. The
affected fields are described with each record. Note that this
principle does not govern the Use32/Use16 segment attribute (which
is set in the ACBP byte of SEGDEF records); it simply specifies the
size of certain numeric fields within the record. It is possible to
use 16-bit OMF records to generate 32-bit segments, or vice versa.
LINK ignores the value of the checksum byte, but some other utilities may
not. Microsoft's Quick languages write a 0 byte instead of computing a checksum.
The contents of each record are determined by the record type, but
certain subfields appear frequently enough to be explained separately.
The format of such fields is below.
Names :
A name string is encoded as an 8-bit unsigned count followed by a
string of count characters. The character set is usually some ASCII
subset. A null name is specified by a single byte of 0 (indicating a
string of length 0).
Indexed References :
Certain items are ordered by occurrence and are referenced by index.
The first occurrence of the item has index number 1. Index fields may
contain 0 (indicating that they are not present) or values from 1
through 7FFF. The index number field in an object record can be either
1 or 2 bytes long. If the number is in the range 0-7FH, the high-order
bit (bit 7) is 0 and the low-order bits contain the index number, so
the field is only 1 byte long. If the index number is in the range 80-
7FFFH, the field is 2 bytes long. The
Type Indexes :
Type Index fields occupy 1 or 2 bytes and occur in PUBDEF, LPUBDEF,
COMDEF, LCOMDEF, EXTDEF, and LEXTDEF records. They are encoded as
described above for indexed references, but the interpretation of the
values stored is governed by whether the module has the "new" or "old"
object module format.
"Old" versions of the OMF (indicated by lack of a COMENT record with
comment class A1), have Type Index fields that contain indexes into
previously seen TYPDEF records. This format is no longer produced by
Microsoft products and is ignored by LINK if it is present. See the
section of this document on TYPDEF records for details on how this was
used.
"New" versions of the OMF (indicated by the presence of a COMENT
record with comment class A1), have Type Index fields that contain
proprietary CodeView information. For more information on CodeView,
see Appendix 1.
Ordered Collections :
Certain records and record groups are ordered so that the records may
be referred to with indexes (the format of indexes is described in the
"Indexed References" section of this document). The same format is
used whether an index refers to names, logical segments, or other
items.
The overall ordering is obtained from the order of the records within
the file together with the ordering of repeated fields within these
records. Such ordered collections are referenced by index, counting
from 1 (index 0 indicates unknown or not specified).
For example, there may be many LNAMES records within a module, and
each of those records may contain many names. The names are indexed
starting at 1 for the first name in the first LNAMES record
encountered while reading the file, 2 for the second name in the first
record, and so forth, with the highest index for the last name in the
last LNAMES record encountered.
The ordered collections are:
Names Ordered by occurrence of LNAMES records and
names within each. Referenced as a name
index.
Logical Ordered by occurrence of SEGDEF records in
Segments file. Referenced as a segment index.
Groups Ordered by occurrence of GRPDEF records in
file. Referenced as a group index.
External Ordered by occurrence of EXTDEF, COMDEF,
Symbols LEXTDEF, and LCOMDEF records and symbols
within each. Referenced as an external name
index (in FIXUP subrecords).
Numeric 2- and 4-Byte Fields :
Certain records, notably SEGDEF, PUBDEF, LPUBDEF, LINNUM, LEDATA,
LIDATA, FIXUPP, and MODEND, contain size, offset, and displacement
values that may be 32-bit quantities for Use32 segments. The encoding
is as follows:
- When the least-significant bit of the record type byte is set (that
is, the record type is an odd number), the numeric fields are 4
bytes.
- When the least-significant bit of the record type byte is clear,
the fields occupy 2 bytes. The values are zero-extended when
applied to Use32 segments.
NOTE: See the description of SEGDEF records in this document for an
explanation of Use16/Use32 segments.
The general record ordering is not mandatory, but should be (for link speed)
like this :
THEADR or LHEADR record :
Records Processed by LINK Pass 1 :
All records may occur in any order but must stand before the link pass
separator, if it is present.
COMENT records identifying object format and extensions
COMENT records other than Link Pass Separator comment
LNAMES or LLNAMES records providing ordered name list
SEGDEF records providing ordered list of program segments
GRPDEF records providing ordered list of logical segments
TYPDEF records (obsolete)
ALIAS records
PUBDEF records locating and naming public symbols
LPUBDEF records locating and naming private symbols
COMDEF, LCOMDEF, EXTDEF, LEXTDEF, and CEXTDEF records
Link Pass Separator (Optional) :
COMENT class A2 record indicating that Pass 1 of the linker is
complete. When this record is encountered, LINK stops reading the
object file in Pass 1; no records after this comment are read in Pass
1. All the records listed above must come before this COMENT record.
For greater linking speed, all LIDATA, LEDATA, FIXUPP, BAKPAT, INCDEF,
and LINNUM records should come after the A2 COMENT record, but this is
not required. In LINK, Pass 2 begins again at the start of the object
module, so these records are processed in Pass 2 no matter where they
are placed in the object module.
Records Ignored by LINK Pass 1 and Processed by LINK Pass 2 :
The following records may come before or after the Link Pass
Separator:
LIDATA, LEDATA, or COMDAT records followed by applicable FIXUPP
records
FIXUPP records containing only THREAD subrecords
BAKPAT and NBKPAT FIXUPP records
COMENT class A0, subrecord type 03 (INCDEF) records containing
incremental compilation information for FIXUPP and LINNUM records
LINNUM and LINSYM records providing line number and program code or
data association
Terminator :
MODEND record indicating end of module with optional start address
Details of each record (form and content) follow below.
Conflicts between various OMFs that overlap in their use of record
types or fields are marked.
Below is a combined list of record types defined by the Intel 8086 OMF
specification and record types added after that specification was
finished. Titles in square brackets ([]) indicate record types that
have been implemented and that are described in this document. Titles
not in square brackets indicate record types that have not been
implemented and are followed by a paragraph of description from the
Intel specification.
For unimplemented record types, a subtle distinction is made between
records that LINK ignores and those for which LINK generates an
"illegal object format" error condition.
Records Currently Defined
6EH RHEADR R-Module Header Record
This record serves to identify a module that has
been processed (output) by LINK-86/LOCATE-86. It
also specifies the module attributes and gives
information on memory usage and need. This record
type is ignored by Microsoft LINK.
70H REGINT Register Initialization Record
This record provides information about the 8086
register/register-pairs: CS and IP, SS and SP, DS
and ES. The purpose of this information is for a
loader to set the necessary registers for
initiation of execution. This record type is
ignored by Microsoft LINK.
72H REDATA Relocatable Enumerated Data Record
This record provides contiguous data from which a
portion of an 8086 memory image may eventually be
constructed. The data may be loaded directly by
an 8086 loader, with perhaps some base fixups.
The record may also be called a Load-Time
Locatable (LTL) Enumerated Data Record. This
record type is ignored by Microsoft LINK.
74H RIDATA Relocatable Iterated Data Record
This record provides contiguous data from which a
portion of an 8086 memory image may eventually be
constructed. The data may be loaded directly by
an 8086 loader, but data bytes within the record
may require expansion. The record may also be
called a Load-Time Locatable (LTL) Iterated Data
Record. This record type is ignored by Microsoft
LINK.
76H OVLDEF Overlay Definition Record
This record provides the overlay's name, its
location in the object file, and its attributes.
A loader may use this record to locate the data
records of the overlay in the object file. This
record type is ignored by Microsoft LINK.
78H ENDREC End Record
This record is used to denote the end of a set of
records, such as a block or an overlay. This
record type is ignored by Microsoft LINK.
7AH BLKDEF Block Definition Record
This record provides information about blocks
that were defined in the source program input to
the translator that produced the module. A BLKDEF
record will be generated for every procedure and
for every block that contains variables. This
information is used to aid debugging programs.
This record type is ignored by Microsoft LINK.
7CH BLKEND Block End Record
This record, together with the BLKDEF record,
provides information about the scope of variables
in the source program. Each BLKDEF record must be
followed by a BLKEND record. The order of the
BLKDEF, debug symbol records, and BLKEND records
should reflect the order of declaration in the
source module. This record type is ignored by
Microsoft LINK.
7EH DEBSYM Debug Symbols Record
This record provides information about all
local symbols, including stack and based symbols.
The purpose of this information is to aid debug-
ging programs. This record type is ignored by
Microsoft LINK.
[80H] [THEADR] [Translator Header Record]
[82H] [LHEADR] [Library Module Header Record]
84H PEDATA Physical Enumerated Data Record
This record provides contiguous data,
from which a portion of an 8086 memory
image may be constructed. The data
belongs to the "unnamed absolute segment"
in that it has been assigned absolute
8086 memory addresses and has been
divorced from all logical segment
information. This record type is ignored
by Microsoft LINK.
86H PIDATA Physical Iterated Data Record
This record provides contiguous data,
from which a portion of an 8086 memory
image may be constructed. It allows
initialization of data segments and
provides a mechanism to reduce the size
of object modules when there is repeated
data to be used to initialize a memory
image. The data belongs to the "unnamed
absolute segment." This record type is
ignored by Microsoft LINK.
[88H] [COMENT] [Comment Record]
[8AH/8BH] [MODEND] [Module End Record]
[8CH] [EXTDEF] [External Names Definition Record]
[8EH] [TYPDEF] [Type Definition Record]
[90H/91H] [PUBDEF] [Public Names Definition Record]
92H LOCSYM Local Symbols Record
This record provides information about
symbols that were used in the source
program input to the translator that
produced the module. This information is
used to aid debugging programs. This
record has a format identical to the
PUBDEF record. This record type is
ignored by Microsoft LINK.
[94H/95H] [LINNUM] [Line Numbers Record]
[96H] [LNAMES] [List of Names Record]
[98H/99H] [SEGDEF] [Segment Definition Record]
[9AH] [GRPDEF] [Group Definition Record]
[9CH/9DH] [FIXUPP] [Fixup Record]
9EH (none) Unnamed record
This record number was the only even
number not defined by the original Intel
specification. Apparently it was never
used. This record type is ignored by
Microsoft LINK.
[A0H/A1H] [LEDATA] [Logical Enumerated Data Record]
[A2H/A3H] [LIDATA] [Logical Iterated Data Record]
A4H LIBHED Library Header Record
This record is the first record in a library
file. It immediately precedes the modules
(if any) in the library. Following the
modules are three more records in the
following order: LIBNAM, LIBLOC, and LIBDIC.
This record type is ignored by Microsoft
LINK.
A6H LIBNAM Library Module Names Record
This record lists the names of all the
modules in the library. The names are listed
in the same sequence as the modules appear
in the library. This record type is ignored
by Microsoft LINK.
A8H LIBLOC Library Module Locations Record
This record provides the relative location,
within the library file, of the first byte
of the first record (either a THEADR or
LHEADR or RHEADR record) of each module in
the library. The order of the locations
corresponds to the order of the modules in
the library. This record type is ignored by
Microsoft LINK.
AAH LIBDIC Library Dictionary Record
This record gives all the names of public
symbols within the library. The public names
are separated into groups; all names in the
nth group are defined in the nth module of
the library. This record type is ignored by
Microsoft LINK.
[B0H] [COMDEF] [Communal Names Definition Record]
[B2H/B3H] [BAKPAT] [Backpatch Record]
[B4H] [LEXTDEF] [Local External Names Definition Record]
[B6H/B7H] [LPUBDEF] [Local Public Names Definition Record]
[B8H] [LCOMDEF] [Local Communal Names Definition Record]
BAH/BBH COMFIX Communal Fixup Record
Microsoft doesn't support this never-
implemented IBM extension. This record type
generates an error when it is encountered by
Microsoft LINK.
BCH CEXTDEF COMDAT External Names Definition Record
C0H SELDEF Selector Definition Record
Microsoft doesn't support this never-
implemented IBM extension. This record type
generates an error when it is encountered by
Microsoft LINK.
[C2H/C3] [COMDAT] [Initialized Communal Data Record]
[C4H/C5H] [LINSYM] [Symbol Line Numbers Record]
[C6H] [ALIAS] [Alias Definition Record]
[C8H/C9H] [NBKPAT] [Named Backpatch Record]
[CAH] [LLNAMES] [Local Logical Names Definition Record]
[F0H] [Library Header Record]
Although this is not actually an OMF record
type, the presence of a record with F0H as
the first byte indicates that the module is
a Microsoft library. The format of a library
file is given in Appendix 2.
[F1H] [Library End Record]
80H THEADR--TRANSLATOR HEADER RECORD
The THEADR record contains the name of the object module. This name
identifies an object module within an object library or in messages
produced by the linker.
OFFSET Count TYPE Description
0000h 1 byte ID=80h
0001h 1 byte Record length
="LEN"
0002h "LEN" char Name
0002h 1 byte Checksum
+"LEN"
82H LHEADR--LIBRARY MODULE HEADER RECORD
This record is very similar to the THEADR record. It is used to
indicate the name of a module within a library file (which has an
internal organization different from that of an object module).
This record type was defined in the original Intel specification with
the same format but with a different purpose, so its use for libraries
should be considered a Microsoft extension.
OFFSET Count TYPE Description
0000h 1 byte ID=82h
0001h 1 byte Record length
="LEN"
0002h "LEN" char Name
0002h 1 byte Checksum
+"LEN"
EXTENSION:OBJ,OBP,OBW,LIB
OCCURENCES:PC
PROGRAMS:MS Link, TLink
REFERENCE:****
--------H-OS/2 HELP-------------------------
The OS/2 help files are different from the WinHelp help files,since the WinHelp
format is proprietary to MicroSoft because of the patented LZ-packing they
implemented.
OFFSET Count TYPE Description
0000h 3 char ID='HSP'
0003h 1 byte Flags :
0 - INF style file
1-3 - unknown
4 - HLP style file
Patching this file allows reading HLP files
using the VIEW command, while HLP files seem to
work with INF settings as well.
0005h 1 word Total size of header
0007h 1 word Unknown
????h other data
0047h ? char ASCIIZ name of the HLP/INF file
EXTENSION:HLP,INF
OCCURENCES:OS/2
REFERENCE:INF02A.DOC
SEE ALSO:WinHelp HLP
--------I-PBM-G-----------------------------
The PBM files are image files, which were used at least by DMGraph, an utility
to insert new graphics into a DOOM WAD file. The image dimensions seem to be
stored in ASCII format delimited with CR/LF, after that follows the raw binary
image data.
OFFSET Count TYPE Description
0000h 1 char ID='P'
0001h 1 char Bitmap type :
'1' - PBM bitmap
'2' - PGM greymap
'3' - PPM pixmap
'4' - PBM raw bitmap
'5' - PGM raw greymap
'6' - PPM raw pixmap
EXTENSION:PBM,PGM,PPM
OCCURENCES:PC
PROGRAMS:DMGraph.EXE
--------I-PCX-------------------------------
The PCX files are created by the programs of the ZSoft Paintbrush family
and the FRIEZE package by the same manufacturer. A PCX file contains only
one image, the data for this image and possibly palette information for
this image. The encoding scheme used for PCX encoding is a simple RLE
mechanism, see ALGRTHMS.txt for further information. A PCX image is stored
from the upper scan line to the lower scan line.
The size of a decoded scan line is ******
The header has a fixed size of 128 bytes and looks like this :
OFFSET Count TYPE Description
0000h 1 byte Manufacturer.
10=ZSoft
0001h 1 byte Version information
0=PC Paintbrush v2.5
2=PC Paintbrush v2.8 w palette information
3=PC Paintbrush v2.8 w/o palette information
4=PC Paintbrush/Windows
5=PC Paintbrush v3.0+
0002h 1 byte Encoding scheme, 1 = RLE, none other known
0003h 1 byte Bits per pixel
0004h 1 word left margin of image
0006h 1 word upper margin of image
0008h 1 word right margin of image
000Ah 1 word lower margin of image
000Ch 1 word Horizontal DPI resolution
000Eh 1 word Vertical DPI resolution
0010h 48 byte Color palette setting for 16-color images
16 RGB triplets
0040h 1 byte reserved
0041h 1 byte Number of color planes
="NCP"
0042h 1 word Number of bytes per scanline (always even,
use instead of right margin-left margin).
="NBS"
0044h 1 word Palette information
1=color/bw palette
2=grayscale image
0046h 1 word Horizontal screen size
0048h 1 word Vertical screen size
004Ah 54 byte reserved, set to 0
The space needed to decode a single scan line is "NCP"*"NBS" bytes, the last
byte may be a junk byte which is not displayed.
After the image data, if the version number is 5 (or greater?) there possibly
is a VGA color palette. The color ranges from 0 to 255, 0 is zero intensity,
255 is full intensity. The palette has the following format :
OFFSET Count TYPE Description
0000h 1 byte VGA palette ID (=0Ch)
0001h 768 byte RGB triplets with palette information
EXTENSION:PCX
OCCURENCES:PC
PROGRAMS:PC Paintbrush
SEE ALSO:
--------I-PIC-------------------------------
PIC files contain images in an uncompressed format. Both the original Animator
and Animator Pro from Autodesk produce PIC files. The file formats are
different; Animator Pro produces a hierarchial block oriented file, while the
original Animator file is a simpler fixed format. See PIC(Pro) for further
information on the Animator Pro PIC format.
The original Animator uses this format to store a single-frame picture image.
This format description applies to both PIC and original Animator CEL files. The
file begins with a 32 byte header, as follows:
OFFSET Count TYPE Description
0000h 1 word ID=9119h
0002h 1 word Width of image; PIC files have always a width
of 320, CEL images may have any value.
0004h 1 word Height of image, 200 for a PIC, any value for
a CEL file.
0006h 1 word X offset of image, always 0 for a PIC image,
may be nonzero in a CEL image.
0008h 1 word Y offset of image. Zero for a PIC file.
000Ah 1 byte Bits per pixel (8)
000Bh 1 byte Compresion flag, always zero
000Ch 1 dword Size of the image data in bytes
0010h 16 byte reserved(0)
Immediately following the header is the color map. It contains all 256 palette
entries in rgb order. Each of the r, g, and b components is a single byte in the
range of 0-63. Following the color palette is the image data, one byte per
pixel, from left to right, top to bottom.
EXTENSION:PIC,CEL
OCCURENCES:PC
PROGRAMS:Autodesk Animator
SEE ALSO:CEL,FLIc,PIC(PRO)
--------I-PIC(PRO)--------------------------
This format description applies to both PIC and MSK files created with the
Autodesk Animator Pro package. The file begins with a 64-byte header defined
as follows:
Offset Length Name Description
0000h 1 dword The size of the whole file including the size
of this header.
0004h 1 word ID=9500h
0006h 1 word Width of the image
0008h 1 word Height of the image
000Ah 1 word X offset of image
000Ch 1 word Y offset of image
000Eh 1 dword User ID, set to zero
0012h 1 byte Bits per pixel (8 for PIC, 1 for MSK)
0013h 45 byte reserved (0)
Following the file header are the data blocks for the image. Each data block
within a PIC or MSK file is formatted as follows:
OFFSET Count TYPE Description
0000h 1 dword The size of the block, including this header.
0004h 1 word Data type ID :
0 - Color palette info
1 - Byte-per-pixel image data
2 - Bit-per-pixel mask data
0006h ? byte Data
The type values in the block headers indicate what type of graphics data the
block contains.
In a PIC_CMAP block, the first 2-byte word is a version code;
currently this is set to zero. Following the version word are all 256 palette
entries in rgb order. Each of the r, g, and b components is a single byte in the
range of 0-255. This type of block appears in PIC files; there will generally be
no color map block in a MSK file.
In a PIC_BYTEPIXELS block, the image data appears immediately following the
6-byte block header. The data is stored as one byte per pixel, in left-to-right,
topD to-bottom sequence.
In a PIC_BITPIXELS block, the bitmap data appears immediately following the
6-byte block header. The data is stored as bits packed into bytes such that the
leftmost bits appear in the high-order positions of each byte. The bits are
stored in left-to-right, top-to bottom sequence. When the width of the bitmap is
not a multiple of 8, there will be unused bits in the low order positions of the
last byte on each line. The number of bytes per line is ((width+7)/8). This type
of block appears in MSK files.
EXTENSION:PIC,MSK
OCCURENCES:PC
PROGRAMS:Autodesk Animator Pro
REFERENCE:
SEE ALSO:PIC,FLT
--------I-PLY-------------------------------
The PoLYgon files created by the Autodesk Animator packages contain a set of
points that describe a polygon.
OFFSET Count TYPE Description
0000h 1 word Number of points in the file
0002h 1 dword reserved (0)
0006h 1 byte Closed shape flag. If nonzero there is an
implied connection between the last and the
first point. If it is zero, the shape is open.
0007h 1 byte ID=99h
After the header, there follows the point data, organized in records like this :
OFFSET Count TYPE Description
0000h 1 word X coordinate
0002h 1 word Y coordinate
0006h 1 word Z coordinate, always zero
EXTENSION:PLY
OCCURENCES:PC
PROGRAMS:Autodesk Animator
--------M-PS16------------------------------
The Protracker Studio 16 Modules are yet another digital music format.
The Protracker modules can have up to 255 different patterns and a length of
up to 255 patterns with 31 instruments. The samples can only have a size of
up to 64K, there is a maximum of 16 tracks supported.
The header of each MOD file looks like this :
OFFSET Count TYPE Description
0000h 5 char Header string
ID='PS16',254
0005h 75 char Song name, ending with ^Z so that if typing
the file will result in PS16 <SongName>
0050h 1 byte File type :
0 - Module with patterns and samples
1 - Song with patterns but without samples
0051h 1 dword Offset of comment field from start of file.
Zero if no commend is stored.
0055h 1 byte Format version byte (0)
0056h 1 byte Number of patterns in the file
="PAT"
0057h 1 dword Total size of all patterns in bytes, stored
for quick disk reads.
005Bh 1 byte Songlength, number of sequences.
005Ch 128 byte Sequencing information for file
00CCh 31 rec Sample information
1 byte Sample flags, bitmapped :
0 - synthesized / digital
1 - Waveform / FM
(only if bit 0 is set)
2 - 16 bit / 8 bit
1 byte Default volume for the sample (0-64)
1 byte Sample fine tuning (signed nibble)
1 dword Sample length - Protracker does only support
samples with a size less than 64K.
1 dword Sample loop start
1 dword Sample loop length
1 word Default playback frequency for C2
Can be used to fine tune a sample.
00CCh+ "PAT" rec The pattern information
31*17 The tracks are stored sequentially after each
other, first all rows of track 1, then all rows
of track 2 and so on.
1 word Pattern size+3, rounded up to a paragraph
boundary.
1 byte Number of rows in pattern
="ROW"
"ROW" rec
2 byte Note information, bitmapped :
0-5 - Note (see table 0005)
6 - Bit 4 of instrument
7 - Compression bit
If this bit not set, there is another
byte following the note record
specifying the row where the next
event takes place - if it is set,
the next note follows immediately.
A track is terminated by a 0FFh byte.
8-11 - Effect bits
12-15 - Bits 0-3 of instrument
1 byte Effect data
00CCh+ ? byte Sample data in delta format
31*17+ See algorthm.txt for details.
????
The comment block contains information about the sample names as well
as some comments to the module. It is formatted like this :
OFFSET Count TYPE Description
0000h 4 char ID='INST'
1 byte Instrument name length
="LEN"
1 byte Sample name count
="CNT"
"LEN"*"CNT" char Sample names
0000h+ 4 char ID='TEXT'
"LEN"*"CNT"+4
1 word Length of following text
EXTENSION:MOD
OCCURENCES:PC
PROGRAMS:Protracker
REFERENCE:
SEE ALSO:DMF,MOD,S3M,STM
VALIDATION:
--------I-QFX-------------------------------
QFX files are yet another graphic file format used to store received
fax images. The .QFX file format is proprietary to Smith Micro Software, Inc.
and is used by the Quick Link II fax software.
The QFX file header is exactly 1536 bytes long. The fax pages themselves
are stored in byte aligned, bit reversed T4 format terminated with 6 EOL's.
See CCITT Recommendation T.4 for full documentation on this coding scheme.
OFFSET Count TYPE Description
0000h 8 char ID='QLIIFAX',0
0008h 1 word Number of pages in the QFX file
000Ah 1 word Number of scan lines on last page
000Ch 1 dword Number of scan lines for all pages
0010h 1 word Horizontal scaling
1 - High res (200x200),
2 - Normal res (200x100)
0012h 1 word Vertical scaling (always = 1).
0014h 12 byte reserved
0020h 375 dword Offsets of the single pages in the document.
Page 1 always starts at offset 1536. The last
non-zero dword points to the end of the last
page, the first zero dword marks the end of
the pages.
0600h ? byte Start of fax page images
EXTENSION:QFX
OCCURENCES:PC
PROGRAMS:Quick Link II
--------S-RAW-------------------------------
The RAW files are raw signed PCM sound files. PCM means Pulse Code
Modulation - which can be played through most sound devices without
further manipulation. There is no header or whatsoever.
The properties include 8/16-bit samples in INTEL order,
stereo or mono format. No identification is possible.
EXTENSION:RAW
SEE ALSO:SND
--------I-RDIB------------------------------
The RDIB files are Device Independent Bitmaps used by Windows. They are RIFF
format files. The blocks are unknown to me.
SEE ALSO:RIFF
--------f-RIFF------------------------------
The RIFF (Resource interchange file format) format was created by Microsoft and
is used by many applications like Windows, Corel Draw etc.. It is block
structured, each block has a header ID and a size, so that even a program that
works with an old version of the file format can skip the unknown parts of the
file and work on the known parts of the file. All RIFF blocks begin on a word
boundary so it might be necessary to skip an additional byte. In the present
specification, only one RIFF block per file is allowed, and only the RIFF and
LIST blocks may contain subblocks.
The order of blocks in a RIFF file is not mandatory, so you should always scan
the whole file for the block ID you seek. Throughout this file, the RIFF block
IDs are given in square brackets []. Each ID is always 4 characters dword.
OFFSET Count TYPE Description
0000h 4 char ID='RIFF'
Each RIFF format file has a header with the
signature and the size of the following
blocks.
0004h 1 dword Block size. This size is the size of the block
controlled by the RIFF header. Normally this
equals the file size.
="BSZ"
0008h 4 char Format name. This is the format name of the RIFF
file.
After this RIFF header comes the first RIFF record. Each RIFF record has the
following format :
OFFSET Count TYPE Description
0000h 4 char Signature. This is the description of what is
contained in this block.
0004h 1 dword Block size. This is the size of the following
data block. To get the offset of the next RIFF
block record, you have to add this value + 8 to
the offset of the current record.
---RiffBLOCK [LIST]
This block contains a string list, again in the RIFF subblock format. This list
is used for messages and/or copyright messages. All strings in the LIST block
share the same format, each block contains one ASCIIZ string - the most common
LIST block is the [INFO] block, which can contain the following subblocks :
[INAM]
The name of the data stored in the file
[ICRD]
Creation date of the file
SEE ALSO:BMP,rDIB,IFF,WAVe,RIFX
OCCURENCES:PC
PROGRAMS:Windows,Corel Draw
REFERENCE:DDJ0994
VALIDATION:FileSize="BSZ"+8
--------f-RIFX-M----------------------------
The RIFX file format is identical to the RIFF file format except that all values
are in Motorola byte order.
OFFSET Count TYPE Description
0000h 4 char ID='RIFX'
0004h 1 dword Block size. This size is the size of the block
controlled by the RIFX header.
="BSZ"
0008h 4 char Format name.
REFERENCE:DDJ0994
SEE ALSO:RIFF
--------S-S3I-------------------------------
This is the Digiplayer/ST3.0 digital sample file format. The sample
files include information about the loop of the instrument. The AdLib
instruments have another format listed below.
OFFSET Count TYPE Description
0000h 1 byte ID=01h
0001h 12 char DOS filename
000Dh 1 byte reserved (0)
000Eh 1 word Paragraph offset of the raw sample data
from beginning of file.
0010h 1 dword Sample length in bytes
0014h 1 dword Start of sample loop
0018h 1 dword End of sample loop
001Ch 1 byte Playback volumne of sample
001Dh 1 byte ??? "DSK" what ever that means
001Eh 1 byte Pack type
0 - unpacked
1 - DP30ADPCM 1
001Fh 1 byte Flags (bitmapped)
0 - loop on/off
1 - stereo sample (length bytes for left channel,
then another length bytes for right channel!)
2 - 16-Bit samples (in Intel byte order)
0020h 1 dword C2 frequency
0024h 1 dword reserved
0028h 1 word reserved
002Ah 1 word ID=512
002Ch 1 dword ?? Date of last modification ?? (see table 0009)
0030h 28 char ASCIIZ Sample name
003Ch 4 char ID='SCRS'
0040h ? byte Raw sample data
Here follows the AdLib instrument format for which I don't know the
extension (yet) :
OFFSET Count TYPE Description
0000h 1 byte Instrument type
2 - melodic instrument
3 - bass drum
4 - snare drum
5 - tom tom
6 - cymbal
7 - hihat
0001h 12 char DOS file name
000Dh 3 byte reserved
0010h 1 byte Modulator description (bitmapped)
0-3 - frequency multiplier
4 - scale envelope
5 - sustain
6 - pitch vibrato
7 - volume vibrato
0011h 1 byte Carrier description (same as modulator)
0012h 1 byte Modulator miscellaneous (bitmapped)
0-5 - 63-volume
6 - MSB of levelscale
7 - LSB of levelscale
0013h 1 byte Carrier description (same as modulator)
0014h 1 byte Modulator attack / decay byte (bitmapped)
0-3 - Decay
4-7 - Attack
0015h 1 byte Carrier description (same as modulator)
0016h 1 byte Modulator sustain / release byte (bitmapped)
0-3 - Release count
4-7 - 15-Sustain
0017h 1 byte Carrier description (same as modulator)
0018h 1 byte Modulator wave select
0019h 1 byte Carrier wave select
001Ah 1 byte Modulator feedback byte (bitmapped)
0 - additive synthesis on/off
1-7 - modulation feedback
001Bh 1 byte reserved
001Ch 1 byte Instrument playback volume
001Dh 1 byte ??? "DSK"
001Eh 1 word reserved
0020h 1 dword C2 frequency
0024h 12 byte reserved
0030h 28 char ASCIIZ Instrument name
004Ch 4 char ID='SCRI'
EXTENSION:S3I,SMP
OCCURENCES:PC
PROGRAMS:ScreamTracker 3.0
SEE ALSO:MTM,S3M,STM
--------M-S3M-------------------------------
The ScreamTracker composer and the ScreamTracker Music Interface Kit (STMIK)
were written by the demo group Future Crew for their demonstrations and
released. S3M files are the files of the version 3 of the ScreamTracker.
OFFSET Count TYPE Description
0000h 20 char Song name, ASCII, 0 padded
001Ch 1 byte ID=1Ah
001Dh 1 byte Filetype :
16=Module
17=Song
? What is this supposed to mean ?
001Eh 1 word Reserved
0020h 1 word Number of orders in song
="ORD"
0022h 1 word Number of instruments in song
="INS"
0024h 1 word Number of patterns in song
="PAT"
0026h 1 word Song flags, bitmapped
0 - ScreamTracker 2.0 type vibrato
1 - ScreamTracker 2.0 type tempo
2 - Amiga type slides
3 - Zero volume optimizations
4 - Amiga limits
5 - enable filters / sfx
0028h 1 word Tracker version
002Ah 1 word File format version
1=Original format
2=Original format, unsigned samples
002Ch 4 char ID='SCRM'
0032h 1 byte Maximum volume
0033h 1 byte Initial speed
0034h 1 byte Initial tempo
0035h 1 byte Master multiplier
Whats this ????
0036h 12 byte reserved
0040h 32 byte Channel balance settings
0=left
127=right
+128=disabled
255=unused
0060h "ORD" byte Ordering sequence of the patterns
0060h "INS" word Offset of the instruments in paragraphs from
+"ORD" begin of header (for binary offset, multiply with 16)
0060h "PAT" word Offset of the pattern data from begin of header
+"ORD" in paragraphs
+"INS"*2
EXTENSION:S3M
OCCURENCES:PC
PROGRAMS:ScreamTracker 3.0
SEE ALSO:S3I,STM,S2M
--------S-SND-------------------------------
The SND files are raw unsigned PCM sound files. PCM means Pulse Code
Modulation - which can be played through most sound devices without
further manipulation. There is no header or whatsoever.
The properties include 8/16-bit samples in INTEL order,
stereo or mono format. No identification is possible.
EXTENSION:SND
SEE ALSO:RAW
--------A-SQZ-------------------------------
The SQZ files are yet another archive format. The SQZ archives consist of one
archive header and several file headers. The archive header has the
following format :
OFFSET Count TYPE Description
0000h 5 char ID='HLSQZ'
0005h 1 char Version in ASCII
ID='1'
0006h 1 byte OS byte,
0 - PC-DOS / MS-DOS
1 - OS/2
2 - MVS
3 - HPFS (OS/2)
4 - Amiga
5 - Macintosh
6 - Unix
0007h 1 byte Misc. flags, bitmapped :
0 - Intel byte order / Motorola byte order
1 - Filetime in ?? / File time in DOS format
2 - No security envelope / security envelope
3-7 - reserved
After the header and each block, there is one byte denoting the type/size of the
next block :
OFFSET Count TYPE Description
0000h 1 byte Block/size specifier :
0 - End of archive
1 - Comment
2 - Password
3 - Security envelope
4-18 - reserved
>18 - normal file header,
byte value is size of header
The normal file header then has the following format :
OFFSET Count TYPE Description
0000h 1 byte Checksum of header
0001h 1 byte Flags, bitmapped :
0-3 : Compression method :
0 -
1 -
2 -
3 -
4 -
4 - Security envelope should follow
5-7 - reserved
0002h 1 dword Compressed size of file
0006h 1 dword Original file size
000Ah 1 dword Date and time of file (see table 0009)
000Eh 1 byte File attributes
000Fh 1 dword CRC-32 of file
0013h ? char Filename (see above for length)
The comment block :
OFFSET Count TYPE Description
0000h 1 word Size of uncompressed comment
0002h 1 word Size of compressed comment data
="LEN"
0004h 1 byte Flags, bitmapped, see above
0005h 1 dword CRC-32
0009h "LEN" byte Compressed comment data
The password block :
OFFSET Count TYPE Description
0000h 1 word Size of password block (=4)
0004h 1 dword CRC-32 of password
Other blocks :
OFFSET Count TYPE Description
0000h 1 word Size of this block
="LEN"
0002h "LEN" byte Block data
EXTENSION:SQZ
OCCURENCES:PC
PROGRAMS:??
REFERENCE:
SEE ALSO:
--------S-SDK-------------------------------
The SDK files are disk images from disks used by the Roland
S-550/S-50/S-330 sampler devices.
Further information wanted.
EXTENSION:SDK
--------S-SDS-------------------------------
The SDS files are MIDI Sample Dump Standart files and are used
to transfer samples between MIDI devices.
Further information wanted.
EXTENSION:SDS
SEE ALSO:MID,SDX
--------S-SDX-------------------------------
The SDX file are like the SDS files sample dump files used
for transfer of data between MIDI devices.
EXTENSION:SDX
SEE ALSO:MID,SDS
--------S-SMP-------------------------------
The SMP files are digital sample files used by Samplevision software.
Further information wanted.
EXTENSION:SMP
--------M-STM-------------------------------
The ScreamTracker 1.0 format was the module format used by the
ScreamTracker before version 2.0.
OFFSET Count TYPE Description
0000h 20 char ASCIIZ song name
0014h 8 char Tracker name
001Ch 1 byte ID=1Ah
001Dh 1 byte File type
1 - song (contains no samples)
2 - module (contains samples)
001Eh 1 byte Major version number
001Fh 1 byte Minor version number
0020h 1 byte Playback tempo
0021h 1 byte Number of patterns
="PAT"
0022h 1 byte Global playback volume
0023h 13 byte reserved
0030h 31 rec Instrument data
12 char ASCIIZ instrument name
1 byte ID=0
1 byte Instrument disk
1 word reserved
1 word Sample length in bytes
1 word Sample loop start
1 word Sample loop end
1 byte Sample playback volume
1 byte reserved
1 word C3 frequency in Hz
1 dword reserved
1 word length in paragraphs
(only for modules,in songs:reserved)
03D0h 64 byte Pattern orders
0410h 4*64*"PAT" rec Pattern data. Each pattern consists of
64 rows, each 4 channels. The channels
are stored from left ro right, row by row.
1 byte Note byte :
251 - last 3 bytes not stored, all bytes 0
252 - last 3 bytes not stored, note -0-,
whatever that means.
253 - last 3 bytes not stored, note ...
254 - undefined (reserved for run-time)
255 - undefined (reserved for run-time)
otherwise bit mapped :
0-3 : note (c=0,c#=1...)
4-7 : octave
1 byte Only valid if above byte < 251, bit mapped
0-2 ; lower bit of note volume
3-7 : instrument number
1 byte bit mapped
0-3 : Effect command in ProTracker format
seems to be overlapped by volume
bits...
4-6 : upper bits of volume
1 byte command data in ProTracker format
0410h+ ? byte Raw sample data padded to 16 byte boundaries.
4*64*4*"PAT"
EXTENSION:STM
OCCURENCES:PC
PROGRAMS:ScreamTracker 1.0
REFERENCE:
SEE ALSO:S3M,MOD
--------A-TAR-G-----------------------------
The Unix Tape ARchives mostly have the extension TAR. The info about this
comes from a magic file, thus useful only for identification.
--------A-TAR-------------------------------
The Unix TAR program is an archiver program which stores files in a single
archive without compression.
OFFSET Count TYPE Description
@section The Standard Format
A @dfn{tar tape} or file contains a series of records. Each record
contains @code{RECORDSIZE} bytes. Although this format may be
thought of as being on magnetic tape, other media are often used.
Each file archived is represented by a header record which describes
the file, followed by zero or more records which give the contents
of the file. At the end of the archive file there may be a record
filled with binary zeros as an end-of-file marker. A reasonable
system should write a record of zeros at the end, but must not
assume that such a record exists when reading an archive.
The records may be @dfn{blocked} for physical I/O operations. Each
block of @var{N} records (where @var{N} is set by the @samp{-b}
option to @code{tar}) is written with a single @code{write()}
operation. On magnetic tapes, the result of such a write is a
single tape record. When writing an archive, the last block of
records should be written at the full size, with records after the
zero record containing all zeroes. When reading an archive, a
reasonable system should properly handle an archive whose last block
is shorter than the rest, or which contains garbage records after a
zero record.
The header record is defined in C as follows:
@example
/*
* Standard Archive Format - Standard TAR - USTAR
*/
#define RECORDSIZE 512
#define NAMSIZ 100
#define TUNMLEN 32
#define TGNMLEN 32
union record @{
char charptr[RECORDSIZE];
struct header @{
char name[NAMSIZ];
char mode[8];
char uid[8];
char gid[8];
char size[12];
char mtime[12];
char chksum[8];
char linkflag;
char linkname[NAMSIZ];
char magic[8];
char uname[TUNMLEN];
char gname[TGNMLEN];
char devmajor[8];
char devminor[8];
@} header;
@};
/* The checksum field is filled with this while the checksum is computed. */
#define CHKBLANKS " " /* 8 blanks, no null */
/* The magic field is filled with this if uname and gname are valid. */
#define TMAGIC "ustar " /* 7 chars and a null */
/* The magic field is filled with this if this is a GNU format dump entry */
#define GNUMAGIC "GNUtar " /* 7 chars and a null */
/* The linkflag defines the type of file */
#define LF_OLDNORMAL '\0' /* Normal disk file, Unix compatible */
#define LF_NORMAL '0' /* Normal disk file */
#define LF_LINK '1' /* Link to previously dumped file */
#define LF_SYMLINK '2' /* Symbolic link */
#define LF_CHR '3' /* Character special file */
#define LF_BLK '4' /* Block special file */
#define LF_DIR '5' /* Directory */
#define LF_FIFO '6' /* FIFO special file */
#define LF_CONTIG '7' /* Contiguous file */
/* Further link types may be defined later. */
/* Bits used in the mode field - values in octal */
#define TSUID 04000 /* Set UID on execution */
#define TSGID 02000 /* Set GID on execution */
#define TSVTX 01000 /* Save text (sticky bit) */
/* File permissions */
#define TUREAD 00400 /* read by owner */
#define TUWRITE 00200 /* write by owner */
#define TUEXEC 00100 /* execute/search by owner */
#define TGREAD 00040 /* read by group */
#define TGWRITE 00020 /* write by group */
#define TGEXEC 00010 /* execute/search by group */
#define TOREAD 00004 /* read by other */
#define TOWRITE 00002 /* write by other */
#define TOEXEC 00001 /* execute/search by other */
@end example
All characters in header records are represented by using 8-bit
characters in the local variant of ASCII. Each field within the
structure is contiguous; that is, there is no padding used within
the structure. Each character on the archive medium is stored
contiguously.
Bytes representing the contents of files (after the header record of
each file) are not translated in any way and are not constrained to
represent characters in any character set. The @code{tar} format
does not distinguish text files from binary files, and no
translation of file contents is performed.
The @code{name}, @code{linkname}, @code{magic}, @code{uname}, and
@code{gname} are null-terminated character strings. All other
fileds are zero-filled octal numbers in ASCII. Each numeric field
of width @var{w} contains @var{w}@minus{} 2 digits, a space, and a null,
except @code{size}, and @code{mtime}, which do not contain the
trailing null.
The @code{name} field is the pathname of the file, with directory
names (if any) preceding the file name, separated by slashes.
The @code{mode} field provides nine bits specifying file permissions
and three bits to specify the Set UID, Set GID, and Save Text
(``stick'') modes. Values for these bits are defined above. When
special permissions are required to create a file with a given mode,
and the user restoring files from the archive does not hold such
permissions, the mode bit(s) specifying those special permissions
are ignored. Modes which are not supported by the operating system
restoring files from the archive will be ignored. Unsupported modes
should be faked up when creating or updating an archive; e.g. the
group permission could be copied from the @code{other} permission.
The @code{uid} and @code{gid} fields are the numeric user and group
ID of the file owners, respectively. If the operating system does
not support numeric user or group IDs, these fields should be
ignored.
The @code{size} field is the size of the file in bytes; linked files
are archived with this field specified as zero.
@xref{Extraction Options}; in particular the @samp{-G} option.@refill
The @code{mtime} field is the modification time of the file at the
time it was archived. It is the ASCII representation of the octal
value of the last time the file was modified, represented as an
integer number of seconds since January 1, 1970, 00:00 Coordinated
Universal Time.
The @code{chksum} field is the ASCII representation of the octal
value of the simple sum of all bytes in the header record. Each
8-bit byte in the header is added to an unsigned integer,
initialized to zero, the precision of which shall be no less than
seventeen bits. When calculating the checksum, the @code{chksum}
field is treated as if it were all blanks.
The @code{typeflag} field specifies the type of file archived. If a
particular implementation does not recognize or permit the specified
type, the file will be extracted as if it were a regular file. As
this action occurs, @code{tar} issues a warning to the standard
error.
@table @code
@item LF_NORMAL
@itemx LF_OLDNORMAL
These represent a regular file. In order to be compatible with
older versions of @code{tar}, a @code{typeflag} value of
@code{LF_OLDNORMAL} should be silently recognized as a regular
file. New archives should be created using @code{LF_NORMAL}. Also,
for backward compatibility, @code{tar} treats a regular file whose
name ends with a slash as a directory.
@item LF_LINK
This represents a file linked to another file, of any type,
previously archived. Such files are identified in Unix by each file
having the same device and inode number. The linked-to
name is specified in the @code{linkname} field with a trailing null.
@item LF_SYMLINK
This represents a symbolic link to another file. The linked-to
name is specified in the @code{linkname} field with a trailing null.
@item LF_CHR
@itemx LF_BLK
These represent character special files and block special files
respectively. In this case the @code{devmajor} and @code{devminor}
fields will contain the major and minor device numbers
respectively. Operating systems may map the device specifications
to their own local specification, or may ignore the entry.
@item LF_DIR
This specifies a directory or sub-directory. The directory name in
the @code{name} field should end with a slash. On systems where
disk allocation is performed on a directory basis the @code{size}
field will contain the maximum number of bytes (which may be rounded
to the nearest disk block allocation unit) which the directory may
hold. A @code{size} field of zero indicates no such limiting.
Systems which do not support limiting in this manner should ignore
the @code{size} field.
@item LF_FIFO
This specifies a FIFO special file. Note that the archiving of a
FIFO file archives the existence of this file and not its contents.
@item LF_CONTIG
This specifies a contiguous file, which is the same as a normal
file except that, in operating systems which support it,
all its space is allocated contiguously on the disk. Operating
systems which do not allow contiguous allocation should silently treat
this type as a normal file.
@item 'A' @dots{}
@itemx 'Z'
These are reserved for custom implementations. Some of these are
used in the GNU modified format, as described below.
@end table
Other values are reserved for specification in future revisions of
the P1003 standard, and should not be used by any @code{tar} program.
The @code{magic} field indicates that this archive was output in the
P1003 archive format. If this field contains @code{TMAGIC}, the
@code{uname} and @code{gname} fields will contain the ASCII
representation of the owner and group of the file respectively. If
found, the user and group ID represented by these names will be used
rather than the values within the @code{uid} and @code{gid} fields.
@section GNU Extensions to the Archive Format
The GNU format uses additional file types to describe new types of
files in an archive. These are listed below.
@table @code
@item LF_DUMPDIR
@itemx 'D'
This represents a directory and a list of files created by the
@samp{-G} option. The @code{size} field gives the total size of the
associated list of files. Each filename is preceded by either a @code{'Y'}
(the file should be in this archive) or an @code{'N'} (The file is a
directory, or is not stored in the archive). Each filename is
terminated by a null. There is an additional null after the last
filename.
@item LF_MULTIVOL
@itemx 'M'
This represents a file continued from another volume of a
multi-volume archive created with the @samp{-M} option. The original
type of the file is not given here. The @code{size} field gives the
maximum size of this piece of the file (assuming the volume does not
end before the file is written out). The @code{offset} field gives
the offset from the beginning of the file where this part of the
file begins. Thus @code{size} plus @code{offset} should equal the
original size of the file.
@item LF_VOLHDR
@itemx 'V'
This file type is used to mark the volume header that was given with
the @samp{-V} option when the archive was created. The @code{name}
field contains the @code{name} given after the @samp{-V} option.
The @code{size} field is zero. Only the first file in each volume
of an archive should have this type.
@end table
EXTENSION:
OCCURENCES:
PROGRAMS:
REFERENCE:
SEE ALSO:
VALIDATION:
OFFSET Count TYPE Description
0000h 256 byte Other header info ?
0100h 6 char ID='ustar',0
EXTENSION:TAR
OCCURENCES:PC, Unix
PROGRAMS:TAR
--------G-TDDD------------------------------
This format is used by the Imagine rendering package. The names of the blocks
are unknown to me.
OFFSET Count TYPE Description
EXTENSION:IFF
OCCURENCES:Amiga,PC
PROGRAMS:Imagine package
REFERENCE:DDJ0794
SEE ALSO:IFF
--------I-TIFF------------------------------
The TIFF file format was designed jointly by Aldus and Microsoft with leading
scanner vendors to faciliate incorporating scanned images into publishing.
The described TIFF specification is TIFF 5.0. A TIFF file consists of several
different blocks which define the palette data or the LZW-compressed body
among other things. TIFF files can be in Motorola _or_ Intel byte order,
depending on the first word. If it is 'II', the byte order is in Intel order,
if it is 'MM', then you have Motorola byte ordering.
Each TIFF file begins with a image file header which points to one or more image
file directories, which contain the image data and image information.
The format of the image header :
OFFSET Count TYPE Description
0000h 2 char ID='II', ID='MM'
This is the identification, 'II' stands for
Intel byte order, 'MM' for Motorola byte
order. The following data must be interpreted
accordingly !
0002h 1 word TIFF "version number". This version number
never changed and the value (42) was choosen
for its deep philosophical value. In fact, if
the version number ever changes, this means
that radical changes to the TIFF format have
been made, and a TIFF reader should give up
immediately.
You can consider this word to be a part of the
header ID.
0004h 1 dword Offset of first image directory in file form
start of file.
The first image directory must begin on an
even byte boundary. The image directory may
follow the image data it describes. The
image directory is described below.
An organization may wish to store information that is meaningful to only that
organization in a TIFF file. Tags numbered 32768 or higher are reserved for
that purpose. Upon request, the administrator will allocate and register a block
of private tags for an organization.
Private enumerated values can be accommodated in a similar fashion.
The format of the image file directory (IFD) :
All entries are sorted in ascending order by the tag field.
OFFSET Count TYPE Description
0000h 1 word Number of entries
="NUM"
0002h "NUM" rec Field descriptor
1 word Field tag, see below
1 word Field type
1 - byte
2 - ASCII string, counted in length.
Most often an ASCIIZ string, the
trailing zero is counted with the
data length.
3 - word
4 - dword / uword
5 - rational (2 dwords, numerator and denominator)
1 dword Length of the field in units of the data type.
A single 16-bit word has the length 1.
1 dword Data offset of the field. The data starts
on a word boundary, thus the dword should
be even. The data for the field may be
anywhere in the file, even after the image
data. If the data size is less or equal to
4 bytes (determined by the field type and
length), then this offset is not a offset
but instead the data itself, to save space.
If the data size is less than 4 bytes, the
data is stored left-justified within the 4
bytes of the offset field.
0002h+
"NUM"*12 1 dword Offset of next IFD in file, 0 if none follow
If a certain field in the IFD does not exist, you have to presume the default
values. The different fields are :
--- BitsPerSample
Tag = 258 (102)
Type = word
N = SamplesPerPixel
Default = 1
Number of bits per sample. Note that this tag allows a different number of
bits per sample for each sample corresponding to a pixel. For example, RGB
color data could use a different number of bits per sample for each of the
three color planes.
--- ColorMap
Tag = 320 (140)
Type = word
N = 3 * (2**BitsPerSample)
No default.ColorMap must be included in all palette color images.
This tag defines a Red-Green-Blue color map for palette color images.
The palette color pixel value is used to index into all 3 subcurves.
The subcurves are stored sequentially. The Red entries come first, followed
by the Green entries, followed by the Blue entries.
The width of each entry is 16 bits, as implied by the type of word.
0 represents the minimum intensity, and 65535 represents the maximum intensity.
--- ColorResponseCurves
Tag = 301 (12D)
Type = word
N = 3 * (2**BitsPerSample)
Default: curves based on the NTSC recommended gamma of 2.2.
This tag defines three color response curves, one each for Red, Green and Blue
color information. The Red entries come first, followed by the Green entries,
followed by the Blue entries. The length of each subcurve is 2**BitsPerSample,
using the BitsPerSample value corresponding to the respective primary. The width
of each entry is 16 bits, as implied by the type of word.
The purpose of the color response curves is to refine the content of RGB color images.
--- Compression
Tag = 259 (103)
Type = word
N = 1
Default = 1.
1 = No compression, but pack data into bytes as tightly as possible, with no
unused bits except at the end of a row. The bytes are stored as an array
of bytes, for BitsPerSample <= 8, word if BitsPerSample > 8 and <= 16, and
dword if BitsPerSample > 16 and <= 32. The byte ordering of data >8 bits
must be consistent with that specified in the TIFF file header (bytes 0
and 1). Rows are required to begin on byte boundaries.
2 = CCITT Group 3 1-Dimensional Modified Huffman run length encoding.
See ALGRTHMS.txt BitsPerSample must be 1, since this type of compression
is defined only for bilevel images (like FAX images...)
3 = Facsimile-compatible CCITT Group 3, exactly as specified in
"Standardization of Group 3 facsimile apparatus for document
transmission," Recommendation T.4, Volume VII, Fascicle VII.3,
Terminal Equipment and Protocols for Telematic Services, The
International Telegraph and Telephone Consultative Committee
(CCITT), Geneva, 1985, pages 16 through 31. Each strip must
begin on a byte boundary. (But recall that an image can be a
single strip.) Rows that are not the first row of a strip are
not required to begin on a byte boundary. The data is stored as
bytes, not words - byte-reversal is not allowed. See the
Group3Options field for Group 3 options such as 1D vs 2D coding.
4 = Facsimile-compatible CCITT Group 4, exactly as specified in
"Facsimile Coding Schemes and Coding Control Functions for Group
4 Facsimile Apparatus," Recommendation T.6, Volume VII, Fascicle
VII.3, Terminal Equipment and Protocols for Telematic Services,
The International Telegraph and Telephone Consultative Committee
(CCITT), Geneva, 1985, pages 40 through 48. Each strip must
begin on a byte boundary. Rows that are not the first row of a
strip are not required to begin on a byte boundary. The data is
stored as bytes, not words. See the Group4Options field for
Group 4 options.
5 = LZW Compression, for grayscale, mapped color, and full color images.
See ALGRTHMS.txt
32773 = PackBits compression, a simple byte oriented run length scheme for
1-bit images. See Appendix C.
Data compression only applies to raster image data, as pointed to
by StripOffsets.
--- GrayResponseCurve
Tag = 291 (123)
Type = word
N = 2**BitsPerSample
The purpose of the gray response curve and the gray units is to
provide more exact photometric interpretation information for
gray scale image data, in terms of optical density.
--- GrayResponseUnit
Tag = 290 (122)
Type = word
N = 1
For historical reasons, the default is 2. However, for greater
accuracy, 3 is recommended.
1 = Number represents tenths of a unit.
2 = Number represents hundredths of a unit.
3 = Number represents thousandths of a unit.
4 = Number represents ten-thousandths of a unit.
5 = Number represents hundred-thousandths of a unit.
--- ImageLength
Tag = 257 (101)
Type = word or dword
N = 1
No default.
The image's length (height) in pixels (Y:vertical). The number of rows
(sometimes described as "scan lines") in the image.
--- ImageWidth
Tag = 256 (100)
Type = word or dword
N = 1
No default.
The image's width, in pixels (X:horizontal). The number of columns in the image.
--- NewSubfileType
Tag = 254 (FE)
Type = dword
N = 1
Default is 0.
A general indication of the kind of data that is contained in
this subfile. This field is made up of a set of 32 flag bits.
Unused bits are expected to be 0. Bit 0 is the low-order bit.
Currently defined values for the bitmap are:
0 - Image is reduced of another TIFF image in this file
1 - Image is a single page of a multi-page
2 - Image is a transparency mask for another image in this file
--- PhotometricInterpretation
Tag = 262 (106)
Type = word
N = 1
No default.
0 = For bilevel and grayscale images: 0 is imaged as white.
2**BitsPerSample-1 is imaged as black. If GrayResponseCurve
exists, it overrides the PhotometricInterpretation value.
1 = For bilevel and grayscale images: 0 is imaged as black.
2**BitsPerSample-1 is imaged as white. If GrayResponseCurve
exists, it overrides the PhotometricInterpretation value.
2 = RGB. In the RGB model, a color is described as a combination
of the three primary colors of light (red, green, and blue) in
particular concentrations. For each of the three samples, 0
represents minimum intensity, and 2**BitsPerSample - 1 represents
maximum intensity. For PlanarConfiguration = 1, the samples are stored in
the indicated order: first Red, then Green, then Blue. For
PlanarConfiguration = 2, the StripOffsets for the sample planes
are stored in the indicated order: first the Red sample plane
StripOffsets, then the Green plane StripOffsets, then the Blue
plane StripOffsets.
3 = "Palette color." In this mode, a color is described with a
single sample. The sample is used as an index into ColorMap.
The sample is used to index into each of the red, green and blue
curve tables to retrieve an RGB triplet defining an actual color.
When this PhotometricInterpretation value is used, the color
response curves must also be supplied. SamplesPerPixel must be
1.
4 = Transparency Mask. This means that the image is used to
define an irregularly shaped region of another image in the same
TIFF file. SamplesPerPixel and BitsPerSample must be 1.
PackBits compression is recommended. The 1-bits define the
interior of the region; the 0-bits define the exterior of the
region. The Transparency Mask must have the same ImageLength and
ImageWidth as the main image.
PlanarConfiguration
Tag = 284 (11C)
Type = word
N = 1
Default is 1.
1 = The sample values for each pixel are stored contiguously, so
that there is a single image plane. See PhotometricInterpretation
to determine the order of the samples within the pixel data. So, for
RGB data, the data is stored RGBRGBRGB...and so on.
2 = The samples are stored in separate "sample planes." The
values in StripOffsets and StripByteCounts are then arranged as a
2-dimensional array, with SamplesPerPixel rows and StripsPerImage
columns. (All of the columns for row 0 are stored first,
followed by the columns of row 1, and so on.)
PhotometricInterpretation describes the type of data that is
stored in each sample plane. For example, RGB data is stored
with the Red samples in one sample plane, the Green in another,
and the Blue in another.
If SamplesPerPixel is 1, PlanarConfiguration is irrelevant, and
should not be included.
Predictor
Tag = 317 (13D)
Type = word
N = 1
Default is 1.
To be used when Compression=5 (LZW).
1 = No prediction scheme used before coding.
2 = Horizontal differencing. See Appendix I.
ResolutionUnit
Tag = 296 (128)
Type = word
N = 1
Default is 2.
To be used with XResolution and YResolution.
1 = No absolute unit of measurement. Used for images that may
have a non-square aspect ratio, but no meaningful absolute
dimensions. The drawback of ResolutionUnit=1 is that different
applications will import the image at different sizes. Even if
the decision is quite arbitrary, it might be better to use dots
per inch or dots per centimeter, and pick XResolution and
YResolution such that the aspect ratio is correct and the maximum
dimension of the image is about four inches (the "four" is quite
arbitrary.)
2 = Inch.
3 = Centimeter.
RowsPerStrip
Tag = 278 (116)
Type = word or dword
N = 1
Default is 2**32 - 1, which is effectively infinity. That is,
the entire image is one strip. Recomended is a strip size of 8K.
The number of rows per strip. The image data is organized into
strips for fast access to individual rows when the data is
compressed - though this field is valid even if the data is not
compressed.
--- SamplesPerPixel
Tag = 277 (115)
Type = word
N = 1
Default = 1.
The number of samples per pixel. SamplesPerPixel is 1 for
bilevel, grayscale, and palette color images. SamplesPerPixel is
3 for RGB images.
--- StripByteCounts
Tag = 279 (117)
Type = word or dword
N = StripsPerImage for PlanarConfiguration equal to 1.
= SamplesPerPixel * StripsPerImage for PlanarConfiguration equal to 2
No default.
For each strip, the number of bytes in that strip. The existence
of this field greatly simplifies the chore of buffering
compressed data, if the strip size is reasonable.
--- StripOffsets
Tag = 273 (111)
Type = word or dword
N = StripsPerImage for PlanarConfiguration equal to 1.
= SamplesPerPixel * StripsPerImage for PlanarConfiguration equal to 2
No default.
For each strip, the byte offset of that strip. The offset is
specified with respect to the beginning of the TIFF file. Note
that this implies that each strip has a location independent of
the locations of other strips. This feature may be useful for
editing applications. This field is the only way for a reader to
find the image data, and hence must exist.
--- XResolution
Tag = 282 (11A)
Type = RATIONAL
N = 1
No default.
The number of pixels per ResolutionUnit in the X direction, i.e.,
in the ImageWidth direction.
--- YResolution
Tag = 283 (11B)
Type = RATIONAL
N = 1
No default.
The number of pixels per ResolutionUnit in the Y direction, i.e.,
in the ImageLength direction.
--- Artist
Tag = 315 (13B)
Type = ASCII
Person who created the image. Copyright notice.
--- DateTime
Tag = 306 (132)
Type = ASCII
N = 20
Date and time of image creation. Uses the format "YYYY:MM:DD HH:MM:SS", with
hours on a 24-hour clock, and one space character between the date and the time.
The length of the string, including the null, is 20 bytes.
--- HostComputer
Tag = 316 (13C)
Type = ASCII
"ENIAC", or whatever.
--- ImageDescription
Tag = 270 (10E)
Type = ASCII
For example, a user may wish to attach a comment such as "1988 company
picnic" to an image.
--- Make
Tag = 271 (10F)
Type = ASCII
Manufacturer of the scanner, video digitizer, or whatever.
--- Model
Tag = 272 (110)
Type = ASCII
The model name/number of the scanner, video digitizer, or whatever.
This tag is intended for user information only so format is arbitrary.
--- Software
Tag = 305 (131)
Type = ASCII
Name and release number of the software package that created the image.
User information only.
--- Group3Options
Tag = 292 (124)
Type = dword
N = 1
Those options are for fax-images stored in TIFF format.
This field is made up of a set of 32 flag bits. Unused bits are expected
to be 0. It is probably not safe to try to read the file if any bit of this
field is set that you don't know the meaning of.
Bit map :
0 - 2-dimensional coding used.
1 - Image is uncompressed
2 - Fill bits have been added before EOL codes, so that EOL always ends on a
byte boundary.
--- Group4Options
Tag = 293 (125)
Type = dword
N = 1
This field is made up of a set of 32 flag bits and is used for the images
with fax group 4 compression. Unused bits are expected to be 0. It is
probably not safe to try to read the file if any bit of this field is set
that you don't know the meaning of. Gray scale and color coding schemes are
under study, and will be added when finalized.
For 2-D coding, each strip is encoded as if it were a separate image. In
particular, each strip begins on a byte boundary; and the coding for the
first row of a strip is encoded independently of the previous row, using
horizontal codes, as if the previous row is entirely white. Each strip ends
with the 24-bit end-of-facsimile block (EOFB).
Bit map :
0 - reserved (unused)
1 - uncompressed mode is used
2-31 - reserved
--- DocumentName
Tag = 269 (10D)
Type = ASCII
The name of the document from which this image was scanned.
--- PageName
Tag = 285 (11D)
Type = ASCII
The name of the page from which this image was scanned.
--- PageNumber
Tag = 297 (129)
Type = word
N = 2
This tag is used to specify page numbers of a multiple page (e.g. facsimile)
document. Two word values are specified. The first value is the page number;
the second value is the total number of pages in the document. Note that pages
need not appear in numerical order. The first page is 0 (zero).
--- XPosition
Tag = 286 (11E)
Type = RATIONAL
The X offset of the left side of the image, with respect to the left side of the
page, in ResolutionUnits.
--- YPosition
Tag = 287 (11F)
Type = RATIONAL
The Y offset of the top of the image, with respect to the top of the page, in
ResolutionUnits. In the TIFF coordinate scheme, the positive Y direction is down,
so that YPosition is always positive.
--- White Point
Tag = 318 (13E)
Type = RATIONAL
N = 2
Default is the SMPTE white point, D65: x = 0.313, y = 0.329.
The white point of the image. Note that this value is described using the 1931
CIE xyY chromaticity diagram and only the chromaticity is specified. The
luminance component is arbitrary and not specified. This can correspond to the
white point of a monitor that the image was painted on, the filter set/light
source combination of a scanner, or to the white point of the illumination model
of a rendering package. The ordering is x, y.
--- PrimaryChromaticities
Tag = 319 (13F)
Type = RATIONAL
N = 6
Default is the SMPTE primary color chromaticities:
Red: x = 0.635 y = 0.340
Green: x = 0.305 y = 0.595
Blue: x = 0.155 y = 0.070
The primary color chromaticities. Note that these values are described using
the 1931 CIE xyY chromaticity diagram and only the chromaticities are
specified.For paint images, these represent the chromaticities of the monitor
and for scanned images they are derived from the filter set/light source
combination of a scanner.
The ordering is red x, red y, green x, green y, blue x, blue y.
** The following fields are not recommended except perhaps for local use **
They have either been superseded by other fields, have been found to have serious
drawbacks, or are simply not as useful as once thought.
CellLength
Tag = 265 (109)
Type = word
N = 1
The length, in 1-bit samples, of the dithering/halftoning matrix. Assumes that
Threshholding = 2.
This field, plus CellWidth and Threshholding, are problematic because they cannot
safely be used to reverse-engineer grayscale image data out of dithered/halftoned
black-and-white data, which is their only plausible purpose. The only "right" way
to do it is to not bother with anything like these fields, and instead write some
sophisticated pattern-matching software that can handle screen angles that are not
multiples of 45 degrees, and other such challenging dithered/halftoned data.
FillOrder
Tag = 266 (10A)
Type = word
N = 1
The order of data values within a byte.
1 = most significant bits of the byte are filled first.
2 = least significant bits are filled first. Since little interest has been
expressed in least-significant fill order to date, and since it is easy
and inexpensive for writers to reverse bit order (use a 256-byte lookup
table), FillOrder=2 is for private (non-interchange) use only.
FreeByteCounts
Tag = 289 (121)
Type = dword
For each "free block" in the file, the number of bytes in the block.
FreeOffsets
Tag = 288 (120)
Type = dword
For each "free block" in the file, its byte offset.
MaxSampleValue
Tag = 281 (119)
Type = word
N = SamplesPerPixel
Default is 2**(BitsPerSample) - 1.
The maximum used sample value. Statistical use only.
MinSampleValue
Tag = 280 (118)
Type = word
N = SamplesPerPixel
Default is 0.
The minimum used sample value. Statistical use only.
SubfileType
Tag = 255 (FF)
Type = word
N = 1
A general indication of the kind of data that is contained in this subfile.
Currently defined values are:
1 = full resolution image data - ImageWidth, ImageLength, and StripOffsets are
required fields
2 = reduced resolution image data - ImageWidth, ImageLength, and StripOffsets are
required fields. It is further assumed that a reduced resolution image is a
reduced version of the entire extent of the corresponding full resolution
data.
3 = single page of a multi-page image (see the PageNumber tag description).
Continued use of this field is not recommended. Writers should instead use the
new and more general NewSubfileType field.
Orientation
Tag = 274 (112)
Type = word
N = 1
Default is 1.
1 = The 0th row represents the visual top of the image, and the 0th column
represents the visual left hand side.
2 = The 0th row represents the visual top of the image, and the 0th column
represents the visual right hand side.
3 = The 0th row represents the visual bottom of the image, and the 0th column
represents the visual right hand side.
4 = The 0th row represents the visual bottom of the image, and the 0th column
represents the visual left hand side.
5 = The 0th row represents the visual left hand side of the image, and the 0th
column represents the visual top.
6 = The 0th row represents the visual right hand side of the image, and the 0th
column represents the visual top.
7 = The 0th row represents the visual right hand side of the image, and the 0th
column represents the visual bottom.
8 = The 0th row represents the visual left hand side of the image, and the 0th
column represents the visual bottom.
It is extremely costly for most readers to perform image rotation "on the fly",
i.e., when importing and printing; and users of most desktop publishing
applications do not expect a file imported by the application to be altered
permanently in any way.
Threshholding
Tag = 263 (107)
Type = word
N = 1
1 = a bilevel "line art" scan. BitsPerSample must be 1.
2 = a "dithered" scan, usually of continuous tone data such as photographs.
BitsPerSample must be 1.
3 = Error Diffused.
ColorImageType
Tag = 318 (13E)
Type = word
N = 1
Default is 1.
Gives TIFF color image readers a better idea of what kind of color image it is. There will be borderline cases.
1 = Continuous tone, natural image.
2 = Synthetic image, using a greatly restricted range of colors.
Such images are produced by most color paint programs. See ColorList for
a list of colors used in this image.
ColorList
Tag = 319 (13F)
Type = BYTE or word
N = the number of colors that are used in this image, times SamplesPerPixel
A list of colors that are used in this image. Use of this field is only
practical for images containing a greatly restricted (usually less than or equal
to 256) range of colors. ColorImageType should be 2. See ColorImageType.
The list is organized as an array of RGB triplets, with no pad. The RGB triplets
are not guaranteed to be in any particular order. Note that the red, green, and
blue components can either be a BYTE or a word in length. BYTE should be
sufficient for most applications.
EXTENSION:TIF,TIFF
OCCURENCES:PC,MAC,UNIX
PROGRAMS:Aldus Pagemaker, Paintbrush
REFERENCE:
SEE ALSO:
VALIDATION:
--------S-TXW-------------------------------
The TXW files are disk images used by the Yamaha TX-16W.
Further information wanted.
EXTENSION:TXW
--------S-UWF-G-----------------------------
The UWF files are sample files used by the UltraTracker.
Further information wanted.
OFFSET Count TYPE Description
0000h 32 char ASCIIZ sample name
0020h 1 char ID=1Ah
0021h 1 char ID=10h
0022h 5 char ID='MUWFB'
0027h 1 char ID=0
0028h 6 char Length of sample as ASCII long integer
002Eh 1 word Length of sample
?????
EXTENSION:UWF
SEE ALSO:ULT
--------M-ULT-------------------------------
The ULT files are modules used by the UltraTracker. UltraTracker is a
module editor for the Gravis UltraSound soundcard. The version of the
file format used now is 6.
OFFSET Count TYPE Description
0000h 11 char ID="MAS_UTrack_V"
000Ch 4 char Version number in 4-digit ASCII :
1 - ULT version 1.0
2 - ULT version 2.0
3 - ULT version 2.1
4 - ULT version 2.2
000Fh 32 char Song title
002Fh 1 byte Number of song text lines
="NTL"
0030h "NTL"*32 char Song text
0030h+"NTL"*32 1 byte Number of samples
="NOS"
0031h+"NTL"*32 "NOS" rec Sample structure
************
IALL I 49+NTL*32+NOS*SSIbyte[256] IPattern Sequence Table I
IALL I305+NTL*32+NOS*SSIbyte INumber of tracks (NOT) Base 0 I
I I306+NTL*32+NOS*SSIbyte INumber of patterns (NOP) Base 0 I
I2.1-2.2I307+NTL*32+NOS*SSIbyte[NOT] IPAN-position table I
I I I I[0 left]-[F right] I
IALL I307+NTL*32+NOS*SSI IEvent structure (see event struct) I
I I+NOT Ivaries I I
├-------+-----------------+---------------+-----------------------------------┤
IThe remainder of the file is the raw sample data. (signed) I
└-----------------------------------------------------------------------------┘
┌-----------------------------------------------------------------------------┐
ISample Structure (length: ULT 1.0 - ULT 2.1 = 64bytes I ULT 2.2 = 66bytes) I
├-----------------------------------------------------------------------------┤
ISamplename : 32 bytes Sample name I
IDosName : 12 bytes when you load a sample into UT, I
I it records the file name here I
ILoopStart : dbl word loop start point I
ILoopEnd : dbl word loop end point I
ISizeStart : dbl word see below I
ISizeEnd : dbl word see below I
Ivolume : byte UT uses a logarithmic volume setting, ranging I
I from 0-255 (ULT 1.0) I
I from ULT 2.0: uses linear Volume ranging from 0-255 I
IBidi Loop : byte see below I
IFineTune : word Fine tune setting, uses full word value I
I Linear Finetune I
├---Additional-in-ULT-2.2-----------------------------------------------------┤
IC2-Freqency: word This is the frequency, UT uses to play a middle C, I
I all other notes are calculated relatively to this I
I value. I
└-----------------------------------------------------------------------------┘
8 Bit Samples:
--------------
SizeStart:
The SizeStart is the starting offset of the sample.
This seems to tell UT how to load the sample into the Gus's onboard memory.
All the files I have worked with start with a value of 32 for the first sample,
and the previous SizeEnd value for all sample after that. (See Example below)
If the previous sample was 16bit, then SizeStart = (Last SizeEnd * 2)
SizeEnd :
Like the SizeStart, SizeEnd seems to tell UT where to load the sample into the
Gus's onboard memory. SizeEnd equal SizeStart + the length of the sample.
Example:
--------
If a UT file had 3 samples, 1st 12000 bytes, 2nd 5600 bytes, 3rd 8000 byte.
The SizeStart and SizeEnd would look like this:
Sample SizeStart SizeEnd
1st 32 12032
2nd 12032 17632
3rd 17632 25632
***Note***
Samples may NOT cross 256k boundaries. If a sample is too large to fit into the
remaining space, its Sizestart will equal the start of the next 256k boundary.
UT does keep track of the free space at the top of the 256k boundaries, and
will load a sample in there if it will fit.
Example : EndSize = 252144
If the next sample was 12000 bytes, its SizeStart would be 262144, not 252144.
Note that this leaves 10000 bytes unused. If any of the following sample could
fit between 252144 and 262144, its Sizestart would be 252144.
Say that 2 samples after the 12000 byte sample we had a sample that was only
5000 bytes long. Its SizeStart would be 252144 and its SizeEnd would be 257144.
This also applies to 16 Bit Samples.
16 Bit Samples:
---------------
16 bit samples are handled a little different then 8 bit samples.
The SizeStart variable is calculated by dividing offset (last SizeEnd)
by 2. The SizeEnd variable equals SizeStart + (SampleLength / 2).
If the first sample is 16bit, then SizeStart = 16.
Example :
sample1 = 8bit, 1000 bytes
sample2 = 16bit, 5000 bytes
sample1 SizeStart = 32
SizeEnd = 1032 (32 + 1000)
sample2 SizeStart = 516 (offset (1032) / 2)
SizeEnd = 3016 (516 + (5000/2))
***Note***
If a 16bit sample is loaded into banks 2,3, or 4
the SizeStart variable will be
(offset / 2) + 262144 (bank 2)
(offset / 2) + 524288 (bank 3)
(offset / 2) + 786432 (bank 4)
The SizeEnd variable will be
SizeStart + (SampleLength / 2) + 262144 (bank 2)
SizeStart + (SampleLength / 2) + 524288 (bank 3)
SizeStart + (SampleLength / 2) + 786432 (bank 4)
BiDi Loop : (Bidirectional Loop)
--------------------------------
UT takes advantage of the Gus's ability to loop a sample in several different
ways. By setting the Bidi Loop, the sample can be played forward or backwards,
looped or not looped. The Bidi variable also tracks the sample
resolution (8 or 16 bit).
The following table shows the possible values of the Bidi Loop.
Bidi = 0 : No looping, forward playback, 8bit sample
Bidi = 4 : No Looping, forward playback, 16bit sample
Bidi = 8 : Loop Sample, forward playback, 8bit sample
Bidi = 12 : Loop Sample, forward playback, 16bit sample
Bidi = 24 : Loop Sample, reverse playback 8bit sample
Bidi = 28 : Loop Sample, reverse playback, 16bit sample
┌-----------------------------------------------------------------------------┐
IEvent Structure I
├-----------------------------------------------------------------------------┤
INote : byte (See note table below) I
ISampleNumber : byte (Sample Number) I
IEffect1 : nib (Effect1) I
IEffect2 : nib (Effect2) I
IEffectVar : word (Effect variables) I
I I
I The High order byte of EffectVar is the Effect variable for Effect1. I
I The Low order byte of EffectVar is the Effect variable for Effect2. I
I***(Note)*** I
I UT uses a form of compression on repetitive events. Say we read in the firstI
I byte, if it = $FC then this signifies a repeat block. The next byte is the I
I repeat count. followed by the event structure to repeat. I
I If the first byte read does NOT = $FC then this is the note of the event. I
I So repeat blocks will be 7 bytes long : RepFlag : byte ($FC) I
I RepCount : byte I
I note : byte I
I samplenumber : byte I
I effect1 : nib I
I effect2 : nib I
I effectVar : word I
I I
I ! Repeat blocks do NOT bridge patterns. ! I
├-----------------------------------------------------------------------------┤
INote Table I
├-----------------------------------------------------------------------------┤
Inote value of 0 = pause I
IC-0 to B-0 1 to 12 I
IC-3 to B-3 37 to 48 I
IC-4 to B-4 49 to 60 I
IC-5 to B-5 61 to 72 I
IC-6 to B-6 73 to 84 I
IC-7 to B-7 85 to 96 I
└-----------------------------------------------------------------------------┘
That should about cover it. If you have any questions , feel free
to e-mail me at
freejack@shell.portal.com
I can also be contacted on The UltraSound Connection (813) 787-8644
The UltraSound Connection is a BBS dedicated to the Gravis Ultrasound Card.
Also I'm the author of Ripper and Gvoc. If anyone has any questions or
problems, please contact me.
EXTENSION:ULT
SEE ALSO:UWF
--------S-WAVE------------------------------
The Windows .WAV files are RIFF format files. Some programs expect the fmt block
right behind the RIFF header itself, so your programs should write out this
block as the first block in the RIFF file.
The subblocks for the wave files are
RiffBLOCK [data]
This block contains the raw sample data. The necessary information
for playback is contained in the [fmt ] block.
RiffBLOCK [fmt ]
This block contains the data necessary for playback of the sound
files. Note the blank after fmt !
OFFSET Count TYPE Description
0000h 1 word Format tag
1 = PCM (raw sample data)
2 etc. for APCDM, a-Law, u-Law ...
0002h 1 word Channels (1=mono,2=stereo,...)
0004h 1 dword Sampling rate
0008h 1 dword Average bytes per second (=sampling rate*channels)
000Ch 1 word Block alignment / reserved ??
000Eh 1 word Bits per sample (8/12/16-bit samples)
RiffBLOCK [loop]
This block is for looped samples. Very few programs support this block,
but if your program changes the wave file, it should preserve any unknown
blocks.
OFFSET Count TYPE Description
0000h 1 dword Start of sample loop
0004h 1 dword End of sample loop
EXTENSION:WAV
SEE ALSO:RIFF,VOC
OCCURENCES:PC
PROGRAMS:Windows,GUSWAV,WAV2VOC
VALIDATION:NONE
--------W-WKS-------------------------------
The WKS files are worksheets/spreadsheets used by the Lotus 1-2-3 and Lotus
Symphony packages. More information has yet to be found since this information
origins from a magic file.
OFFSET Count TYPE Description
0000h 5 byte ID=0,0,2,0,4
0005h 1 byte WKS type :
4 - Lotus 1-2-3 v1.A WKS
5 - Symphony 1.0 WKS
other - ?WK1 file? (Lotus 2.01+, Symphony 1.1+)
EXTENSION:WKS
OCCURENCES:PC
PROGRAMS:Lotus 1-2-3,Lotus Symphony
SEE ALSO:WKS
--------T-WORD-G----------------------------
The Microsoft Word programs store their documents in files. The info comes
from a magic file and my own (not working) sources, so it is very unreliable
except for identification.
OFFSET Count TYPE Description
0000h 1 dword ID=31BE00
0002h 1 byte Document type :
0 - MS Word text
1 - MS Text building block
2 - Printer description file(maybe wrong topic)
0003h 1 byte ID=00
0004h 1 word ID=AB00h
ToolID, different for the different versions ?
0006h 6 word reserved(0)
0008h 1 dword Textbytes??? Whatever
000Ch 1 word Paragraph information
000Eh 1 word Foot note table
0010h 1 word Section property
0012h 1 word Section table
0014h 1 word Page table
0016h 64 char Style sheet path
0056h 1 word Windows Write page count
Can be used to identify Windows Write files,
because it is 0 for MS Word and nonzero for
Windows Write documents.
0058h 8 char Printer name
Used under MS Word / WinWord only
0060h 1 word MS Word page count
0062h 8 byte Document properties
006Ah 1 byte Word version this file was made by
006Bh 1 bool Autosave flag
006Ch 1 word Word 5 page table
006Eh 1 word Mac bkmk (whatever)
0070h 1 word ?Offset of file name for autosave?
0072h 1 word Running head table
0074h 1 word Code page used making this document
EXTENSION:DOC
OCCURENCES:PC
PROGRAMS:MS Word,Windows Write, WinWord
SEE ALSO:
VALIDATION:
--------T-WORDPERFERCT FILES----------------
The WordPerfect files all have a common header - even tough I don't know
anything else about them.
OFFSET Count TYPE Description
0000h 4 char ID=255,"WPC"
0004h 4 byte unknown
0008h 1 byte ID=1
0009h 1 byte Filetype (see table 0003)
(Table 0003)
File types of WordPerfect files
01h - macro file
02h - WordPerfect help file
03h - keyboard definition file
0Ah - document file
0Bh - dictionary file
0Ch - thesaurus file
0Dh - block
0Eh - rectangular block
0Fh - column block
10h - printer resource file (PRS)
11h - setup file
12h - prefix information file
13h - printer resource file (ALL)
14h - display resource file (DRS)
15h - overlay file (WP.FIL)
16h - graphics file (WPG)
17h - hyphenation code module
18h - hyphenation data module
19h - macro resource file (MRS)
1Ah - graphics driver (WPD)
1Bh - hyphenation lex module
EXTENSION:various
OCCURENCES:PC
--------W-WQ1-------------------------------
Similar to the WKS spreadsheet files, the Quattro Pro spreadsheet files exist,
and their header is somewhat similar. Info again from a magic file which
makes only identification possible.
OFFSET Count TYPE Description
0000h 1 dword ID=00000200h
0004h 1 char ID='Q'
EXTENSION:WQ1
OCCURENCES:PC
PROGRAMS:Borland Quattro Pro
REFERENCE:
SEE ALSO:WKS
VALIDATION:
--------M-XM--------------------------------
The .XM files (Extended Module) are multichannel MOD files created by Triton's
FastTracker ][. They feature up to 32 channels and different effects. FT 2 is
a shareware program. After the initial .XM header follows the pattern data,
after the patterns follow the instruments.
OFFSET Count TYPE Description
0000h 17 char ID="Extended module: "
0011h 20 char Module name, padded with zeroes
0025h 1 char ID=01Ah
0026h 20 char Tracker name
003Ah 1 word Tracker revision number, hi-byte is major version
003Ch 1 dword Header size
0040h 1 word Song length in patterns
0042h 1 word Restart position
0044h 1 word Number of channels
0046h 1 word Number of patterns (< 256)
="PAT"
0048h 1 word Number of instruments (<128)
004Ah 1 word Flags :
0 - Linear frequency table / Amiga freq. table
004Ch 1 word Default tempo
004Eh 1 word Default BPM
0050h 256 byte Pattern order table
--- Pattern header
The patterns are stored as ordinary MOD patterns, except that each note is
stored as 5 bytes:
? 1 (byte) Note (0-71, 0 = C-0)
+1 1 (byte) Instrument (0-128)
+2 1 (byte) Volume column byte (see below)
+3 1 (byte) Effect type
+4 1 (byte) Effect parameter
A simle packing scheme is also adopted, so that the patterns do not become TOO
large: Since the MSB in the note value is never used, it is used for the
compression.If the bit is set, then the other bits are interpreted as follows:
bit 0 set: Note byte ollows
1 set: Instrument byte follows
2 set: Volume column byte follows
3 set: Effect byte follows
4 set: Effect data byte follows
OFFSET Count TYPE Description
0000h 1 dword Length of pattern block/header ??
0004h 1 byte Pattern pack type
0005h 1 word Number of rows in pattern (1..256)
0007h 1 word Size of pattern data
="PSZ"
"PSZ" byte Pattern data
--- Instrument header
Each instrument has one or more sample headers following it.
OFFSET Count TYPE Description
0000h 1 dword Instrument block/header size
0004h 22 char ASCII Instrument name, 0 padded ?
001Ah 1 byte Instrument type (always 0)
001Bh 1 word Number of samples in instrument
001Dh 1 dword Sample header size
0021h 96 byte Sample numbers for all notes
0081h 48 byte Points of volume envelope
00C1h 48 byte Points of panning envelope
0101h 1 byte Number of volume points
0102h 1 byte Number of panning points
0103h 1 byte Volume sustain point
0104h 1 byte Volume loop start point
0105h 1 byte Volume loop end point
0106h 1 byte Panning sustain point
0107h 1 byte Panning loop start point
0108h 1 byte Panning loop end point
0109h 1 byte Volume type, bitmapped
0 - Volume on
1 - Sustain on
2 - Loop on
010Ah 1 byte Panning type, bitmapped
0 - Panning on
1 - Sustain on
2 - Loop on
010Bh 1 byte Vibrato type
010Ch 1 byte Vibrato sweep
010Dh 1 byte Vibrato depth
010Eh 1 byte Vibrato rate
010Fh 1 word Volume fadeout
0111h 1 word Reserved
--- Sample headers
OFFSET Count TYPE Description
0000h 1 dword Sample length
="LEN"
0004h 1 dword Sample loop start
0008h 1 dword Sample loop length
000Ch 1 byte Volume
000Dh 1 byte Finetune for sample (-128..+127)
+-127 is one half tone
000Eh 1 byte Sample type, bitmapped
0,1 : Loop type :
0 - no loop
1 - forward loop
2 - ping-pong loop
3 - reserved
4?: sample is 16-bit
000Fh 1 byte Sample pan
0010h 1 byte Relative note number (signed byte)
(-96..+95), 0 -> C-4 sounds as C-4
0011h 1 byte Reserved
0012h 22 char ASCII name of sample, 0 padded
0013h "LEN" byte Sample data. The sample data is stored
as delta compressed data like the ProTracker.
EXTENSION:XM,MOD
OCCURENCES:
PROGRAMS:
REFERENCE:
SEE ALSO:MOD,S3M
VALIDATION:
--------A-ZIP-------------------------------
The ZIP archives are created by the PkZIP/PkUnZIP combo produced
by the PkWare company. The PkZIP programs have with LHArc and ARJ
the best compression.
The directory information is stored at the end of the archive, each local
file in the archive begins with the following header; This header can be used
to identify a ZIP file as such :
OFFSET Count TYPE Description
0000h 4 char ID='PK',03,04
0004h 1 word Version needed to extract archive
0006h 1 word General purpose bit field (bit mapped)
0 - file is encrypted
1 - 8K/4K sliding dictionary used
2 - 3/2 Shannon-Fano trees were used
3-4 - unused
5-15 - used internally by ZIP
Note: Bits 1 and 2 are undefined if the
compression method is other than
type 6 (Imploding).
0008h 1 word Compression method (see table 0010)
000Ah 1 dword Original DOS file date/time (see table 0009)
000Eh 1 dword 32-bit CRC of file (inverse??)
0012h 1 dword Compressed file size
0016h 1 dword Uncompressed file size
001Ah 1 word Length of filename
="LEN"
001Ch 1 word Length of extra field
="XLN"
001Eh "LEN" char path/filename
001Eh "XLN" char extra field
+"LEN"
After all the files, there comes the central directory structure.
(Table 0010)
PkZip compression types
0 - Stored / No compression
1 - Shrunk / LZW, 8K buffer, 9-13 bits with partial clearing
2 - Reduced-1 / Probalistic compression, lower 7 bits
3 - Reduced-2 / Probalistic compression, lower 6 bits
4 - Reduced-3 / Probalistic compression, lower 5 bits
5 - Reduced-4 / Probalistic compression, lower 4 bits
6 - Imploded / 2/3 Shanno-Fano trees, 4K/8K sliding dictionary
--- Central directory structure
The CDS is at the end of the archive and contains additional information
about the files stored within the archive.
OFFSET Count TYPE Description
0000h 4 char ID='PK',01,02
0004h 1 byte Version made by
0005h 1 byte Host OS (see table 0011)
0006h 1 byte Minimum version needed to extract
0007h 1 byte Target OS
see above "Host OS"
0008h 1 word General purpose bit flag
see above "General purpose bit flag"
000Ah 1 word Compression method
see above "Compression method"
000Ch 1 dword DOS date / time of file
0010h 1 dword 32-bit CRC of file (see table 0009)
0014h 1 dword Compressed size of file
0018h 1 dword Uncompressed size of file
001Ch 1 word Length of filename
="LEN"
001Eh 1 word Length of extra field
="XLN"
0020h 1 word Length of file comment
="CMT"
0022h 1 word Disk number ??
0024h 1 word Internal file attributes (bit mapped)
0 - file is apparently an ASCII/binary file
1-15 - unused
0026h 1 dword External file attributes (OS dependent)
002Ah 1 dword Relative offset of local header from the
start of the first disk this file appears on
002Eh "LEN" char Filename / path; should not contain a drive
or device letter, all slashes should be forward
slashes '/'.
002Eh+ "XLN" char Extra field
+"LEN"
002Eh "CMT" char File comment
+"LEN"
+"XLN"
(Table 0011)
PkZip Host OS table
0 - MS-DOS and OS/2 (FAT)
1 - Amiga
2 - VMS
3 - *nix
4 - VM/CMS
5 - Atari ST
6 - OS/2 1.2 extended file sys
7 - Macintosh
8-255 - unused
--- End of central directory structure
The End of Central Directory Structure header has following format :
OFFSET Count TYPE Description
0000h 4 char ID='PK',05,06
0004h 1 word Number of this disk
0006h 1 word Number of disk with start of central directory
0008h 1 word Total number of file/path entries on this disk
000Ah 1 word Total number of entries in central dir
000Ch 1 dword Size of central directory
0010h 1 dword Offset of start of central directory relative
to starting disk number
0014h 1 word Archive comment length
="CML"
0016h "CML" char Zip file comment
EXTENSION:ZIP
OCCURENCES:PC,Amiga,ST
PROGRAMS:PkZIP,WinZIP
REFERENCE:Technote.APP
--------A-ZOO-------------------------------
The ZOO archive program by Raoul Dhesi is a file compression program now
superceeded in both compression and speed by most other compression programs.
The archive header looks like this :
OFFSET Count TYPE Description
0000h 20 char Archive header text, ^Z terminated, null padded
0014h 1 dword ID=0FDC4A7DCh
0018h 1 dword Offset of first file in archive
001Ch 1 dword Offset of ????
0020h 1 byte Version archive was made by
0021h 1 byte Minimum version needed to extract
Each stored file has its own header, which looks like this :
OFFSET Count TYPE Description
0000h 1 dword ID=0FDC4A7DCh
0004h 1 byte Type of directory entry
0005h 1 byte Compression method :
0 - stored
1 - Crunched : LZW, 4K buffer,
var len (9-13 bits)
0006h 1 dword Offset of next directory entry
000Ah 1 dword Offset of next header
000Dh 1 word Original date / time of file (see table 0009)
0012h 1 word CRC-16 of file
0014h 1 dword Uncompressed size of file
0018h 1 dword Compressed size of file
001Ch 1 byte Version this file was compressed by
001Dh 1 byte Minimum version needed to extract
001Eh 1 byte Deleted flag
0 - file in archive
1 - file is considered deleted
001Fh 1 dword Offset of comment field, 0 if none
0023h 1 word Length of comment field
0025h ? char ASCIIZ path / filename
EXTENSION:ZOO
OCCURENCES:PC
PROGRAMS:ZOO.EXE
REFERENCE:
VALIDATION:
--------S-ZyXEL-----------------------------
The ZyXEL Modems are capable of digitizing speech, the ZFAX software and
answering machine software like VoiceConnect store the sampled data in those
files. The Modems are capable of compressing the data down to 19.2k CPS (ADPCM)
and 9.6k CPS (CELP), the algorithms for the compression may be found in the
ZyxelVoc package by N. Igl, but as the firmware on the modems changes, so might
the compression algorithm. Playback on the modem is always possible.
OFFSET Count TYPE Description
0000h 5 char ID='ZyXEL'
0005h 1 byte 02h, ??? format tag
0006h 4 byte reserved
000Ah 1 word Compression scheme
0 - CELP
1 - 2 bit ADPCM
2 - 3 bit ADPCM
000Ch 4 byte reserved
0010h ? ???? Raw Data
The voice data is just
the data received from U1496
Modem/Fax.
EXTENSION:ZVD,ZYX
OCCURENCES:PC
PROGRAMS:Voice Connect,ZFAX
REFERENCE:ZYXELVOC.*
VALIDATION:NONE
--------!-ALGORITHMS------------------------
Some algorithms used for encoding images etc...
--- TIFF PackBits algorithm
Abstract
This document describes a simple compression scheme for bilevel
scanned and paint type files.
Motivation
The TIFF specification defines a number of compression schemes.
Compression type 1 is really no compression, other than basic
pixel packing. Compression type 2, based on CCITT 1D
compression, is powerful, but not trivial to implement.
Compression type 5 is typically very effective for most bilevel
images, as well as many deeper images such as palette color and
grayscale images, but is also not trivial to implement. PackBits
is a simple but often effective alternative.
Description
Several good schemes were already in use in various settings. We
somewhat arbitrarily picked the Macintosh PackBits scheme. It is
byte oriented, so there is no problem with word alignment. And
it has a good worst case behavior (at most 1 extra byte for every
128 input bytes). For Macintosh users, there are toolbox
utilities PackBits and UnPackBits that will do the work for you,
but it is easy to implement your own routines.
A pseudo code fragment to unpack might look like this:
Loop until you get the number of unpacked bytes you are
expecting:
Read the next source byte into n.
If n is between 0 and 127 inclusive, copy the next n+1 bytes
literally.
Else if n is between -127 and -1 inclusive, copy the next
byte -n+1 times.
Else if n is 128, noop.
Endloop
In the inverse routine, it's best to encode a 2-byte repeat run
as a replicate run except when preceded and followed by a literal
run, in which case it's best to merge the three into one literal
run. Always encode 3-byte repeats as replicate runs.
So that's the algorithm. Here are some other rules:
o Each row must be packed separately. Do not compress across
row boundaries.
o The number of uncompressed bytes per row is defined to be
(ImageWidth + 7) / 8. If the uncompressed bitmap is required to
have an even number of bytes per row, decompress into word-
aligned buffers.
o If a run is larger than 128 bytes, simply encode the
remainder of the run as one or more additional replicate runs.
When PackBits data is uncompressed, the result should be
interpreted as per compression type 1 (no compression).
--- TIFF LZW Compression
Abstract
This document describes an adaptive compression scheme for raster
images.
Reference
Terry A. Welch, "A Technique for High Performance Data
Compression", IEEE Computer, vol. 17 no. 6 (June 1984).
Describes the basic Lempel-Ziv & Welch (LZW) algorithm. The
author's goal in the article is to describe a hardware-based
compressor that could be built into a disk controller or database
engine, and used on all types of data. There is no specific
discussion of raster images. We intend to give sufficient
information in this Appendix so that the article is not required
reading.
Requirements
A compression scheme with the following characteristics should
work well in a desktop publishing environment:
o Must work well for images of any bit depth, including images
deeper than 8 bits per sample.
o Must be effective: an average compression ratio of at least
2:1 or better. And it must have a reasonable worst-case
behavior, in case something really strange is thrown at it.
o Should not depend on small variations between pixels.
Palette color images tend to contain abrupt changes in index
values, due to common patterning and dithering techniques. These
abrupt changes do tend to be repetitive, however, and the scheme
should make use of this fact.
o For images generated by paint programs, the scheme should
not depend on a particular pattern width. 8x8 pixel patterns are
common now, but we should not assume that this situation will not
change.
o Must be fast. It should not take more than 5 seconds to
decompress a 100K byte grayscale image on a 68020- or 386-based
computer. Compression can be slower, but probably not by more
than a factor of 2 or 3.
o The level of implementation complexity must be reasonable.
We would like something that can be implemented in no more than a
couple of weeks by a competent software engineer with some
experience in image processing. The compiled code for
compression and decompression combined should be no more than
about 10K.
o Does not require floating point software or hardware.
The following sections describe an algorithm based on the "LZW"
(Lempel-Ziv & Welch) technique that meets the above requirements.
In addition meeting our requirements, LZW has the following
characteristics:
o LZW is fully reversible. All information is preserved. But
if noise or information is removed from an image, perhaps by
smoothing or zeroing some low-order bitplanes, LZW compresses
images to a smaller size. Thus, 5-bit, 6-bit, or 7-bit data
masquerading as 8-bit data compresses better than true 8-bit
data. Smooth images also compress better than noisy images, and
simple images compress better than complex images.
o On a 68082- or 386-based computer, LZW software can be
written to compress at between 30K and 80K bytes per second,
depending on image characteristics. LZW decompression speeds are
typically about 50K bytes per second.
o LZW works well on bilevel images, too. It always beats
PackBits, and generally ties CCITT 1D (Modified Huffman)
compression, on our test images. Tying CCITT 1D is impressive in
that LZW seems to be considerably faster than CCITT 1D, at least
in our implementation.
o Our implementation is written in C, and compiles to about 2K
bytes of object code each for the compressor and decompressor.
o One of the nice things about LZW is that it is used quite
widely in other applications such as archival programs, and is
therefore more of a known quantity.
The Algorithm
Each strip is compressed independently. We strongly recommend
that RowsPerStrip be chosen such that each strip contains about
8K bytes before compression. We want to keep the strips small
enough so that the compressed and uncompressed versions of the
strip can be kept entirely in memory even on small machines, but
large enough to maintain nearly optimal compression ratios.
The LZW algorithm is based on a translation table, or string
table, that maps strings of input characters into codes. The
TIFF implementation uses variable-length codes, with a maximum
code length of 12 bits. This string table is different for every
strip, and, remarkably, does not need to be kept around for the
decompressor. The trick is to make the decompressor
automatically build the same table as is built when compressing
the data. We use a C-like pseudocode to describe the coding
scheme:
InitializeStringTable();
WriteCode(ClearCode);
Omega = the empty string;
for each character in the strip {
K = GetNextCharacter();
if Omega+K is in the string table {
Omega = Omega+K; /* string concatenation */
} else {
WriteCode (CodeFromString(Omega));
AddTableEntry(Omega+K);
Omega = K;
}
} /* end of for loop */
WriteCode (CodeFromString(Omega));
WriteCode (EndOfInformation);
That's it. The scheme is simple, although it is fairly
challenging to implement efficiently. But we need a few
explanations before we go on to decompression.
The "characters" that make up the LZW strings are bytes
containing TIFF uncompressed (Compression=1) image data, in our
implementation. For example, if BitsPerSample is 4, each 8-bit
LZW character will contain two 4-bit pixels. If BitsPerSample is
16, each 16-bit pixel will span two 8-bit LZW characters.
(It is also possible to implement a version of LZW where the LZW
character depth equals BitsPerSample, as was described by Draft 2
of Revision 5.0. But there is a major problem with this
approach. If BitsPerSample is greater than 11, we can not use
12-bit-maximum codes, so that the resulting LZW table is
unacceptably large. Fortunately, due to the adaptive nature of
LZW, we do not pay a significant compression ratio penalty for
combining several pixels into one byte before compressing. For
example, our 4-bit sample images compressed about 3 percent
worse, and our 1-bit images compressed about 5 percent better.
And it is easier to write an LZW compressor that always uses the
same character depth than it is to write one which can handle
varying depths.)
We can now describe some of the routine and variable references
in our pseudocode:
InitializeStringTable() initializes the string table to contain
all possible single-character strings. There are 256 of them,
numbered 0 through 255, since our characters are bytes.
WriteCode() writes a code to the output stream. The first code
written is a Clear code, which is defined to be code #256.
Omega is our "prefix string."
GetNextCharacter() retrieves the next character value from the
input stream. This will be number between 0 and 255, since our
characters are bytes.
The "+" signs indicate string concatenation.
AddTableEntry() adds a table entry. (InitializeStringTable() has
already put 256 entries in our table. Each entry consists of a
single-character string, and its associated code value, which is,
in our application, identical to the character itself. That is,
the 0th entry in our table consists of the string <0>, with
corresponding code value of <0>, the 1st entry in the table
consists of the string <1>, with corresponding code value of <1>,
..., and the 255th entry in our table consists of the string
<255>, with corresponding code value of <255>.) So the first
entry that we add to our string table will be at position 256,
right? Well, not quite, since we will reserve code #256 for a
special "Clear" code, and code #257 for a special
"EndOfInformation" code that we will write out at the end of the
strip. So the first multiple-character entry added to the string
table will be at position 258.
Let's try an example. Suppose we have input data that looks
like:
Pixel 0: <7>
Pixel 1: <7>
Pixel 2: <7>
Pixel 3: <8>
Pixel 4: <8>
Pixel 5: <7>
Pixel 6: <7>
Pixel 7: <6>
Pixel 8: <6>
First, we read Pixel 0 into K. OmegaK is then simply <7>, since Omega is
the empty string at this point. Is the string <7> already in the
string table? Of course, since all single character strings were
put in the table by InitializeStringTable(). So set Omega equal to
<7>, and go to the top of the loop.
Read Pixel 1 into K. Does OmegaK (<7><7>) exist in the string table?
No, so we get to do some real work. We write the code associated
with Omega to output (write <7> to output), and add OmegaK (<7><7>) to
the table as entry 258. Store K (<7>) into Omega. Note that
although we have added the string consisting of Pixel 0 and Pixel
1 to the table, we "re-use" Pixel 1 as the beginning of the next
string.
Back at the top of the loop. We read Pixel 2 into K. Does OmegaK
(<7><7>) exist in the string table? Yes, the entry we just
added, entry 258, contains exactly <7><7>. So we just add K onto
the end of Omega, so that Omega is now <7><7>.
Back at the top of the loop. We read Pixel 3 into K. Does OmegaK
(<7><7><8>) exist in the string table? No, so write the code
associated with Omega (<258>) to output, and add OmegaK to the table as
entry 259. Store K (<8>) into Omega.
Back at the top of the loop. We read Pixel 4 into K. Does OmegaK
(<8><8>) exist in the string table? No, so write the code
associated with Omega (<8>) to output, and add OmegaK to the table as
entry 260. Store K (<8>) into Omega.
Continuing, we get the following results:
After reading: We write to output: And add table entry:
Pixel 0
Pixel 1 <7> 258: <7><7>
Pixel 2
Pixel 3 <258> 259: <7><7><8>
Pixel 4 <8> 260: <8><8>
Pixel 5 <8> 261: <8><7>
Pixel 6
Pixel 7 <258> 262: <7><7><6>
Pixel 8 <6> 263: <6><6>
WriteCode() also requires some explanation. The output code
stream, <7><258><8><8><258><6>... in our example, should be
written using as few bits as possible. When we are just starting
out, we can use 9-bit codes, since our new string table entries
are greater than 255 but less than 512. But when we add table
entry 512, we must switch to 10-bit codes. Likewise, we switch
to 11-bit codes at 1024, and 12-bit codes at 2048. We will
somewhat arbitrarily limit ourselves to 12-bit codes, so that our
table can have at most 4096 entries. If we push it any farther,
tables tend to get too large.
What happens if we run out of room in our string table? This is
where the afore-mentioned Clear code comes in. As soon as we use
entry 4094, we write out a (12-bit) Clear code. (If we wait any
dworder to write the Clear code, the decompressor might try to
interpret the Clear code as a 13-bit code.) At this point, the
compressor re-initializes the string table and starts writing out
9-bit codes again.
Note that whenever you write a code and add a table entry, Omega is
not left empty. It contains exactly one character. Be careful
not to lose it when you write an end-of-table Clear code. You
can either write it out as a 12-bit code before writing the Clear
code, in which case you will want to do it right after adding
table entry 4093, or after the clear code as a 9-bit code.
Decompression gives the same result in either case.
To make things a little simpler for the decompressor, we will
require that each strip begins with a Clear code, and ends with
an EndOfInformation code.
Every LZW-compressed strip must begin on a byte boundary. It
need not begin on a word boundary. LZW compression codes are
stored into bytes in high-to-low-order fashion, i.e., FillOrder
is assumed to be 1. The compressed codes are written as bytes,
not words, so that the compressed data will be identical
regardless of whether it is an "II" or "MM" file.
Note that the LZW string table is a continuously updated history
of the strings that have been encountered in the data. It thus
reflects the characteristics of the data, providing a high degree
of adaptability.
LZW Decoding
The procedure for decompression is a little more complicated, but
still not too bad:
while ((Code = GetNextCode()) != EoiCode) {
if (Code == ClearCode) {
InitializeTable();
Code = GetNextCode();
if (Code == EoiCode)
break;
WriteString(StringFromCode(Code));
OldCode = Code;
} /* end of ClearCode case */
else {
if (IsInTable(Code)) {
WriteString(StringFromCode(Code));
AddStringToTable(StringFromCode(OldCode)+
FirstChar(StringFromCode(Code)));
OldCode = Code;
} else {
OutString = StringFromCode(OldCode) +
FirstChar(StringFromCode(OldCode));
WriteString(OutString);
AddStringToTable(OutString);
OldCode = Code;
}
} /* end of not-ClearCode case */
} /* end of while loop */
The function GetNextCode() retrieves the next code from the LZW-
coded data. It must keep track of bit boundaries. It knows that
the first code that it gets will be a 9-bit code. We add a table
entry each time we get a code, so GetNextCode() must switch over
to 10-bit codes as soon as string #511 is stored into the table.
The function StringFromCode() gets the string associated with a
particular code from the string table.
The function AddStringToTable() adds a string to the string
table. The "+" sign joining the two parts of the argument to
AddStringToTable indicate string concatenation.
StringFromCode() looks up the string associated with a given
code.
WriteString() adds a string to the output stream.
When SamplesPerPixel Is Greater Than 1
We have so far described the compression scheme as if
SamplesPerPixel were always 1, as will be be the case with
palette color and grayscale images. But what do we do with RGB
image data?
Tests on our sample images indicate that the LZW compression
ratio is nearly identical regardless of whether
PlanarConfiguration=1 or PlanarConfiguration=2, for RGB images.
So use whichever configuration you prefer, and simply compress
the bytes in the strip.
It is worth cautioning that compression ratios on our test RGB
images were disappointing low: somewhere between 1.1 to 1 and 1.5
to 1, depending on the image. Vendors are urged to do what they
can to remove as much noise from their images as possible.
Preliminary tests indicate that significantly better compression
ratios are possible with less noisy images. Even something as
simple as zeroing out one or two least-significant bitplanes may
be quite effective, with little or no perceptible image
degradation.
Implementation
The exact structure of the string table and the method used to
determine if a string is already in the table are probably the
most significant design decisions in the implementation of a LZW
compressor and decompressor. Hashing has been suggested as a
useful technique for the compressor. We have chosen a tree based
approach, with good results. The decompressor is actually more
straightforward, as well as faster, since no search is
involved - strings can be accessed directly by code value.
Performance
Many people do not realize that the performance of any
compression scheme depends greatly on the type of data to which
it is applied. A scheme that works well on one data set may do
poorly on the next.
But since we do not want to burden the world with too many
compression schemes, an adaptive scheme such as LZW that performs
quite well on a wide range of images is very desirable. LZW may
not always give optimal compression ratios, but its adaptive
nature and relative simplicity seem to make it a good choice.
Experiments thus far indicate that we can expect compression
ratios of between 1.5 and 3.0 to 1 from LZW, with no loss of
data, on continuous tone grayscale scanned images. If we zero
the least significant one or two bitplanes of 8-bit data, higher
ratios can be achieved. These bitplanes often consist chiefly of
noise, in which case little or no loss in image quality will be
perceived. Palette color images created in a paint program
generally compress much better than continuous tone scanned
images, since paint images tend to be more repetitive. It is not
unusual to achieve compression ratios of 10 to 1 or better when
using LZW on palette color paint images.
By way of comparison, PackBits, used in TIFF for black and white
bilevel images, does not do well on color paint images, much less
continuous tone grayscale and color images. 1.2 to 1 seemed to
be about average for 4-bit images, and 8-bit images are worse.
It has been suggested that the CCITT 1D scheme could be used for
continuous tone images, by compressing each bitplane separately.
No doubt some compression could be achieved, but it seems
unlikely that a scheme based on a fixed table that is optimized
for word black runs separated by dworder white runs would be a
very good choice on any of the bitplanes. It would do quite well
on the high-order bitplanes (but so would a simpler scheme like
PackBits), and would do quite poorly on the low-order bitplanes.
We believe that the compression ratios would generally not be
very impressive, and the process would in addition be quite slow.
Splitting the pixels into bitplanes and putting them back
together is somewhat expensive, and the coding is also fairly
slow when implemented in software.
Another approach that has been suggested uses uses a 2D
differencing step following by coding the differences using a
fixed table of variable-length codes. This type of scheme works
quite well on many 8-bit grayscale images, and is probably
simpler to implement than LZW. But it has a number of
disadvantages when used on a wide variety of images. First, it
is not adaptive. This makes a big difference when compressing
data such as 8-bit images that have been "sharpened" using one of
the standard techniques. Such images tend to get larger instead
of smaller when compressed. Another disadvantage of these
schemes is that they do not do well with a wide range of bit
depths. The built-in code table has to be optimized for a
particular bit depth in order to be effective.
Finally, we should mention "lossy" compression schemes.
Extensive research has been done in the area of lossy, or non-
information-preserving image compression. These techniques
generally yield much higher compression ratios than can be
achieved by fully-reversible, information-preserving image
compression techniques such as PackBits and LZW. Some
disadvantages: many of the lossy techniques are so
computationally expensive that hardware assists are required.
Others are so complicated that most microcomputer software
vendors could not afford either the expense of implementation or
the increase in application object code size. Yet others
sacrifice enough image quality to make them unsuitable for
publishing use.
In spite of these difficulties, we believe that there will one
day be a standardized lossy compression scheme for full color
images that will be usable for publishing applications on
microcomputers. An International Standards Organization group,
ISO/IEC/JTC1/SC2/WG8, in cooperation with CCITT Study Group VIII,
is hard at work on a scheme that might be appropriate. We expect
that a future revision of TIFF will incorporate this scheme once
it is finalized, if it turns out to satisfy the needs of desktop
publishers and others in the microcomputer community. This will
augment, not replace, LZW as an approved TIFF compression scheme.
LZW will very likely remain the scheme of choice for Palette
color images, and perhaps 4-bit grayscale images, and may well
overtake CCITT 1D and PackBits for bilevel images.
Future LZW Extensions
Some images compress better using LZW coding if they are first
subjected to a process wherein each pixel value is replaced by
the difference between the pixel and the preceding pixel.
Performing this differencing in two dimensions helps some images
even more. However, many images do not compress better with this
extra preprocessing, and for a significant number of images, the
compression ratio is actually worse. We are therefore not making
differencing an integral part of the TIFF LZW compression scheme.
However, it is possible that a "prediction" stage like
differencing may exist which is effective over a broad range of
images. If such a scheme is found, it may be incorporated in the
next major TIFF revision. If so, a new value will be defined for
the new "Predictor" TIFF tag. Therefore, all TIFF readers that
read LZW files must pay attention to the Predictor tag. If it is
1, which is the default case, LZW decompression may proceed
safely. If it is not 1, and the reader does not recognize the
specified prediction scheme, the reader should give up.
Acknowledgements
The original LZW reference has already been given. The use of
ClearCode as a technique to handle overflow was borrowed from the
compression scheme used by the Graphics Interchange Format (GIF),
a small-color-paint-image-file format used by CompuServe that
also is an adaptation of the LZW technique. Joff Morgan and Eric
Robinson of Aldus were each instrumental in their own way in
getting LZW off the ground.
The TIFF predictor algorithm
The idea is to make use of the fact that many continuous tone
images rarely vary much in pixel value from one pixel to the
next. In such images, if we replace the pixel values by
differences between consecutive pixels, many of the differences
should be 0, plus or minus 1, and so on. This reduces the
apparent information content, and thus allows LZW to encode the
data more compactly.
Assuming 8-bit grayscale pixels for the moment, a basic C
implementation might look something like this:
char image[ ][ ];
int row, col;
/* take horizontal differences:
*/
for (row = 0; row < nrows; row++)
for (col = ncols - 1; col >= 1; col--)
image[row][col] -= image[row][col-1];
If we don't have 8-bit samples, we need to work a little harder,
so that we can make better use of the architecture of most CPUs.
Suppose we have 4-bit samples, packed two to a byte, in normal
TIFF uncompressed (i.e., Compression=1) fashion. In order to
find differences, we want to first expand each 4-bit sample into
an 8-bit byte, so that we have one sample per byte, low-order
justified. We then perform the above horizontal differencing.
Once the differencing has been completed, we then repack the 4-
bit differences two to a byte, in normal TIFF uncompressed
fashion.
If the samples are greater than 8 bits deep, expanding the
samples into 16-bit words instead of 8-bit bytes seems like the
best way to perform the subtraction on most computers.
Note that we have not lost any data up to this point, nor will we
lose any data later on. It might at first seem that our
differencing might turn 8-bit samples into 9-bit differences, 4-
bit samples into 5-bit differences, and so on. But it turns out
that we can completely ignore the "overflow" bits caused by
subtracting a larger number from a smaller number and still
reverse the process without error. Normal twos complement
arithmetic does just what we want. Try an example by hand if you
need more convincing.
Up to this point we have implicitly assumed that we are
compressing bilevel or grayscale images. An additional
consideration arises in the case of color images.
If PlanarConfiguration is 2, there is no problem. Differencing
proceeds the same way as it would for grayscale data.
If PlanarConfiguration is 1, however, things get a little
trickier. If we didnt do anything special, we would be
subtracting red sample values from green sample values, green
sample values from blue sample values, and blue sample values
from red sample values, which would not give the LZW coding stage
much redundancy to work with. So we will do our horizontal
differences with an offset of SamplesPerPixel (3, in the RGB
case). In other words, we will subtract red from red, green from
green, and blue from blue. The LZW coding stage is identical to
the SamplesPerPixel=1 case. We require that BitsPerSample be the
same for all 3 samples.
Results and guidelines
LZW without differencing works well for 1-bit images, 4-bit
grayscale images, and synthetic color images. But natural 24-bit
color images and some 8-bit grayscale images do much better with
differencing. For example, our 24-bit natural test images hardly
compressed at all using "plain" LZW: the average compression
ratio was 1.04 to 1. The average compression ratio with
horizontal differencing was 1.40 to 1. (A compression ratio of
1.40 to 1 means that if the uncompressed image is 1.40MB in size,
the compressed version is 1MB in size.)
Although the combination of LZW coding with horizontal
differencing does not result in any loss of data, it may be
worthwhile in some situations to give up some information by
removing as much noise as possible from the image data before
doing the differencing, especially with 8-bit samples. The
simplest way to get rid of noise is to mask off one or two low-
order bits of each 8-bit sample. On our 24-bit test images, LZW
with horizontal differencing yielded an average compression ratio
of 1.4 to 1. When the low-orde