home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
The World of Computer Software
/
World_Of_Computer_Software-02-385-Vol-1of3.iso
/
i
/
ifp1s156.zip
/
INFOPLUS.HIS
< prev
next >
Wrap
Text File
|
1992-12-30
|
30KB
|
543 lines
Changes by Andrew Rossmann v1.56 (12/30/92)
All : The AP command now accepts a filename for automatic output.
Added an NP command switch. When specified, INFOPLUS will not
attempt to read the partition table from the hard disk. This
a problem on some SCSI drives or controllers.
Added an F command switch. This tell Infoplus to leave 16550
FIFO's enabled.
Reduced potential conflict with CBIS PowerLAN redirector when
checking for the true DOS version (since DOS 5 can change it.)
Reduced chances of false data being returned by some pages when
running in an OS/2 DOS session.
Page 1 : Added detection of HP 95LX.
Page 3 : Unless Windows 3 enhanced mode is detected, will now check for
both VCPI and DPMI.
Page 6 : VESA will try the next mode even if no information is returned
from the current one. (Thanks to Joerg.Prust@gmd.de for finding
this one.)
Page 7 : Added extra information about the mouse, primarily if you have
Microsoft Mouse driver version 7.00 or newer, or compatible.
Page 8 : Changed the Serial Port detection method. Also, 16550 FIFO's
are disabled unless the 'F' command is given on the command
line.
Modified the Sound Blaster detection to restore modified
ports. Should fix 'lost mouse' problems.
Added Gravis UltraSound detection. (untested)
Page 9 : If DOS 5 is detected, INFOPLUS will display the revision letter,
and if DOS is using the HMA, or is based in ROM.
Page 10: CDROM detection should work, but higher risk of false detection.
Page 15: Added several partition type names.
Page 18: Added Advanced Power Managment Specification.
Corrected 386^Max Windows support info.
Added extra information about Hyperdisk for versions 4.50 or
newer.
If Stacker is found, will display drive letters used, and what
drive letters have been swapped, if any.
Page 19: Added display of what interrupts an AMIS TSR has hooked.
Changes by Andrew Rossmann v1.55 (4/21/92)
All : Added an AP command switch. When specified, all pages will be
sent to a file or printer. You will be asked the standard
screenprint questions before the first page.
Page 1: Added detection of many newer IBM PS/2 and PS/1 computers.
Added detection of HP Vectra series.
For AT's and up, will display if a special keyboard function
is supported.
For AT and up Phoenix BIOS's, display the version number.
Page 3: Reduced chances of non-AT's showing extended memory.
Page 4: Under MSDOS 5, DOS blocks show if code or data.
Page 7: Mouse version number should display correctly for Microsoft
version 8 and newer.
Mouse Systems and Logitech drivers are detected. Some show
additional information.
No longer reset the mouse hardware. Should retrieve mouse
information much faster.
Page 8: Removed detection of Tandy digital sound, which wasn't reliable.
NOTE: Under Windows 3.1, if you have the SoundBlaster drivers
enabled, you'll get a warning box from Windows, and Infoplus
will be unable to detect it. This is normal.
Page 9: The country represented by the country code is displayed.
Page 18:SMARTDRV detection should now work, including the new 4.0.
VDS information now recognizes more memory manager products.
Added detection of 386^Max version 6.01+.
Re-arranged screens by program types.
Added detection of EMM386 v4.20+.
Added detection of MICEMM (Micronics EMM) NOTE: I have version
4F, and it isn't detected on my system.
Page 19:NEW! This page displays programs using the Alternate Multiplex
Interrupt. This is a new standard, and not too many programs
use it yet.
Page 20:NEW! This page displays extra information about Memory Managers.
Currently, only QEMM 4.23 or newer is detected. You get information
about the various memory sizes used, and a memory map.
Page 21:Former page 19.
Changes by Andrew Rossmann v1.50 (1/1/92)
All: Added command line switches B, D, M, C, NV, H and '?'.
Added on-line help by pressing F1.
No longer reset the screen on exit unless you start in 40 column
mode, or a non-standard mode.
Page 1: Added detection of Tandy 3000.
Added detection of Tandy 1000 series.
Page 3: Having more than 64K of available UMB space will display
the correct byte size (the K size was correct.)
Page 6: Added additional information for ATI cards, and support for
newer versions.
Detect Tseng ET4000/HiColor DAC combo.
Only display Tseng memory size if it's an ET4000.
Hopefully corrected Trident problems. Also, tell true video
memory for 8900 series.
Hopefully corrected Cirrus problems.
Added detection of Oak chipset.
VGA chipset detections can be disabled by using the NV startup
switch.
Page 8: Modified sound card detections to reduce possible problems with
other cards using the same IO ports.
Added detection of Tandy 1000/2500 digital sound.
Page 9: A second 'page' that lists all currently open files, and lots
of information about each of them. Most useful under Windows
or Desqview.
Page 18:Detect and give information about SMARTDRV, and possibly other
caches.
Display version of QEMM (if newer than 5.10).
Detect JP Software's KSTACK.COM.
Detect Norton NCHACHE, Diskreet, DiskMon, FileSave/EraseProtect.
Added detection of Stacker.
INFOPLUS.ASM: Corrected bug in LONGCALL. It was restoring the wrong old DS
and SI!! (Thanks to Michael Holin for noticing that!)
Changes by Andrew Rossmann v 1.47 (8/2/91)
All: Several pages should work better under OS/2.
Corrected bug that caused lockups under Desqview. Could also
fix other unusual bugs. It appears to have fixed the occasional
'slow screen' bug, too.
Page 2: If an i386 is found, check for the POPAD bug. (This is a minor
bug that is nothing to worry about.)
If a 387 is 'found' with a 286 CPU, then it's identified as a
287XL.
Page 3: Tell if HMA is free or in use.
Detect if UMB (Upper Memory Blocks) are available, and report
size of the largest available.
Tell if DOS 5 is using the HMA, and how much is being used.
Page 9: Minor corrections to work properly with DOS 5 and OS/2.
Page 10:Added detection of DOS 5's DOSKEY.
If DOS 5's SETVER is loaded, the version list will be displayed.
Added a generic DOS extender check (new 'standard')
Page 13:Minor corrections to work properly with DOS 5 and OS/2.
Page 18:Added detection of Norton's NDOS (4DOS in disguise)
Added detection of Phar Lap DOS Extenders.
Added detection of the DOS 5 Task Swapper (which also identifies
itself as Windows in real/standard mode!!)
Added detection of HyperDisk 4.20 or newer.
INFOPLUS.ASM: New subroutine added for POPAD bug test.
Modified ALTINTR and LONGCALL to only pass the Overflow,
Sign, Zero, Aux Carry, Parity and Carry flags TO the called
interrupt/subroutine from the Flags variable in the
Registers record. The standard TP Intr doesn't pass ANY
flags! I need the ability for several detection routines.
The full Flags register will be passed back when the
interrupt/subroutine returns. This is what fixed the
Desqview problem. (Thanks to Heribert Eisele -
heribert@caen.engin.umich.edu for pointing this out!)
Changes by Andrew Rossmann v 1.46 (5/29/91)
Page 2: If an i486 is found without a coprocessor, then an i486SX will
be assumed. An i486SX w/ 487 will show as an i486.
Page 6: Most of the VGA chipsets will return more information.
Page 18:NEW PAGE (sort of).
Moved the TSR and drivers section of page 10 here.
Corrected VDS product version number display. Also tell if
VDS's BIOS data bit is set ($40:$7B, bit 5)
Added detection of TBScanX (a virus scanner)
Added detection of WildUnix (a Unix-like wildcard expander)
Added detection of DOS/16M (a protected mode interface)
Added detection of Inset (a text/graphics program)
Added detection of Microsoft CD-ROM extensions (MSCDEX)
Added detection of Fossil drivers
Added detection of Video Fossil drivers
WARNING! Page 18 could lock up under Desqview. I'm still investigating.
Page 19:Former page 18 (Thank you page.)
Changes by Andrew Rossmann v 1.45 (2/11/91)
All: If Desqview is detected, all screen output will be done using
the BIOS, allowing INFOPLUS to work in a window even on
non-386's. This may affect screen printing.
Page 1: Detailed info will only be shown on an AT or better (as told by
the machine byte at $FFFF:$E.)
Added detection of several computers, including Toshiba
T5200/100 and T1200/XE, IBM model 90's, and the PS/1.
Page 6: Added detection of XGA (MicroChannel systems only for now.)
Page 8: 16450 UART now detected (was shown as 8250 before.)
Added detection of AdLib, SoundBlaster, and Roland MPU-401.
If AdLib detected, it checks if the SOUND.COM driver is loaded
and tells you the version and interface address if found.
If COM port is at address $0000 (Desqview giving no access)
it shows N/A for address, and gives no other info.
Ports at address above $FFF should display correctly.
Page 9: DR-DOS should be detected.
Under DOS 5, will tell if UMBs are part of DOS memory chain.
Under DOS 4+, tell if BUFFERS are in EMS, and their handle.
Page 10:If Windows 3 in enhanced mode is detected, information about
the display is given.
If Desqview is detected, extra video and memory info given.
Added detection of TAME.
Added detection of QEMM/QRAM, MANIFEST (resident), and VIDRAM.
Added detection of pcAnywhere.
Added detection of Disk Spool II.
Added detection of DOS 4 GRAPHICS.COM.
Added detection of Win 3's EGA.SYS.
Added detection of DRIVER.SYS (note: DOS 4 appears to support
some of it's functions, so it may show yes.)
Added detection of the F-PROT package.
Added detection of ANARKEY.
Added detection of Virtual DMA Specification (VDS) [NOTE: Is
supported by QEMM!!]
Added detection of Common Access Method SCSI drivers.
Added detection of PC/TCP Packet drivers.
Page 16:You can use the up and down arrow keys to look at the BPB
information for all drives.
Page 17:CMOS info will only show for an AT or better (as told by the
machine byte at $FFFF:$E.)
NOTE: It seems as if 386 based PS/2's have non-standard CMOS
tables. Ignore warnings about bad CMOS checksum.
Tech: Program re-organized into Units.
Source code distributed separately.
Added information pages that you can print out.
NOTE!!->I'm having problems running under Desqview. I'm still trying to
figure out what's happening. Be aware that the program could
lock up at any time.
Changes by Andrew Rossmann v 1.41 (12/8/90)
All: Re-compiled under Turbo Pascal 6. No special features used.
Page 6: Check for Tseng chips should no longer cause problems if you
really have a Tseng chipset.
Page 8: Corrected bug in UART detection.
Page 16:When reading Boot info, no longer tries looking at the drive
that is one below the current default. (ie: B: if C: was the
default.) This primarily affects DOS 3.30 and 3.31 users only.
Technical: .INC files renamed to .PAS for easier use under the new IDE.
Changes by Andrew Rossmann v 1.40 (11/12/90)
All: New Command! Alt-P will pop up a window in which you can send a
copy of the data part of the screen to a file or the printer.
You can have any non-standard characters converted into Normal
ASCII. You can also have an extra line added to the printout.
You can use it to identify the computer, or whatever.
Page 6: Display 'Checking for xxxx' before trying to detect video
types and chipsets.
Paradise information checking now reads the chips directly,
instead of relying on the BIOS, which seemed to be causing
problems with some boards. It seems as if some may not be
detected.
Corrected ZyMOS detection.
CTI chips will only be detected if a CTI BIOS is found. Direct
chip detection caused problems with some other boards.
Page 8: Serial port info now displays the type of UART found. The
16450 (found on many AT's) will be listed as an 8250. The
16550AFN and 16550AN will display as 16550AN.
Page 9: No longer gets stuck in an infinite loop when trying to
determine STACKS on some computers.
Page 11:The environments' location, size, amount used, and free
space are now shown.
Page 17:Day of Week no longer shown since most BIOS's do not set it.
Changes by Andrew Rossmann v 1.35 (10/7/90)
All: If using <ENTER> to jump to a page, and you enter something illegal,
the page name will be restored on the bottom line.
Page 3: If no Free XMS is available, will now say 0K, instead of ERROR.
Page 6: More VGA chipsets detected, including Ahead, Cirrus, Chips&Technologies
Genoa, Trident, Tseng and ZyMOS.
VESA mode detection will only check values that the board or driver
says is available, instead of trying everything.
Page 8: Under Desqview, if not given priveleges for COM ports, will report
NA for baud, instead of bombing with 'Runtime Error 200'.
Page 17:Thank You info moved to page 18.
NEW PAGE! Lists information contained in your CMOS memory, if a
PC/AT or compatible is detected.
Page 18:Thank you information. Formerly page 17.
INFOPLUS.ASM: Several new routines added for detecting VGA chipsets.
DISKREAD modified to work with Compaq 3.31 and other
non-DOS 4's that support >32M partitions using the extended
read routine format.
Changes by Andrew Rossmann v 1.31 (9/4/90)
Page 2: Corrected bug that could cause lock-ups on PC's without coprocessor.
INFOPLUS.ASM: Corrected ndp routine to not generate WAITs!!
Changes by Andrew Rossmann v 1.30 (9/3/90)
All: If you are in MODE BW80 or MODE MONO, you will get black and white.
Page 1: More machine types identified.
Page 2: Some of the Machine Status Word bits are identified.
CPU and NDP detection routines further modified. NDP detection lockups
appear to be fixed (tested by George Toye.)
On 386/486 CPUs, check BIOS to see if Weitek coprocessor present.
Page 3: Conventional memory sizes now also given in K.
DPMI detection now works!!
No VCPI check if DPMI detected (DPMI takes precedence over VCPI).
Page 6: If a VESA driver or board is detected, you will get a listing of all
VESA modes supported and info about them.
Page 10:Added detection of Desqview and Windows 3.
Page 11:Show current program name and it's path.
Page 15:OS/2 High Performance File System (HPFS) added.
INFOPLUS.ASM: All routines are now using the FAR model.
Slightly modified 486 detection.
Removed FSAVE from NDP detection, appeared to be the culprit.
Modified LONGCALL to have a similar format to Intr and MsDos.
New routines AltIntr and AltMsDos. These use self-modifying
code to generate true interrupts. The standard TP routines
do not, and will not trigger protected mode routines.
Changes by Andrew Rossmann v 1.25 (8/10/90)
Page 2: The CPU and NDP detection has been overhauled. First, the
tests are now run seperately, making it easier to tell which
part is causing problems. The CPU detection code has beeen
modified AGAIN, this time from code sent to me by John Levine,
who said he got it from the Intel '486 manual. Also, the NDP
detection has been modified to reduce the chance of getting an
'unknown coprocessor' message. Test8087 is no longer used.
INFOPLUS.ASM: New coding in the CPU and NDP detections, increasing
reliability, and allowing seperate testing. This new
code should also detect the 486!!
Put some comments in the ATIINFO routine.
Changes by Andrew Rossmann v 1.21 (8/5/90)
Page 2: Slightly modified the CPU detection to disable interrupts
during the 286/386 tests. Hopefully, this will fix the
ongoing problems.
Page 6: The information returned when an ATI video board is detected
is now correct. NOTE: The monitor INFOPLUS says may not be
exactly what you have the board set to. There are only 16
possible selections, and many of the 'models' actually use
the same settings. I can only use the information that ATI
gives with the board. They probably have more information
(such as 8/16 bit mode) buried somewhere that they won't
release.
Changes by Andrew Rossmann v 1.20 (7/29/90)
Page 1: Thanks to information posted to USENET by James Howard of
Dell Computer Corporation, most later model Dell's will be
directly identified.
The adapter ROM scan now searches at 2K boundaries. This will
hopefully detect what QEMM calls 'split roms.'
Page 2: Due to continued problems with the CPU detection locking up
computers, I am now using the original detection code. This
uses differences in the Flags registers to distinguish a 386
from a 286. The 486 will not be directly identified. A side
benefit is that you can now run safely under Windows 3
enhanced mode.
The coprocessor detection code sometimes thinks you have one
when you don't. This gives an 'unknown coprocessor - FF'
message. Now, if that happens, I check to see if Turbo Pascal
found one (the Test8087 variable.) If it didn't find one, I
now say 'none.'
Page 9: The file handle count should now be more accurate. I search
where DOS keeps it's FILES lists, and get my counts that way.
Page 16:Tells whether an extended boot signature is found (DOS 4.)
If extended boot signature, tell the physical drive number
(hard disks start at 128, floppies at 0.)
INFOPLUS.ASM: Commented out 'invalid opcode' CPU detection method, and
re-enabled the original detection code. This results in
the inablity to detect the '486.
Changes by Andrew Rossmann v 1.10 (7/20/90)
Page 1: ROM search now displays size of ROMs found.
Page 3: If an XMS driver is found, various information about it
is now displayed.
All EMS memory is now given in K instead of pages.
Page 6: Added detection of Paradise, Video 7, and ATI VGA boards. Some
additional information and the correct memory size is given.
Page 7: Tell whether the Mouse driver is shadowing the EGA registers.
Page 10:A hopefully generic Network check was added.
The CDROM detection might work better. Also displays what
drives are actually CDROM.
Page 15:Now only search for drives $80 to $99. Going beyond that
caused exception errors under EMM386.SYS (the Win 3 version.)
Page 17:Corrected a bug that caused part of the screen to scroll
when running under Desqview (I hope.)
INFOPLUS.ASM: Added new LONGCALL routine. This is needed for the XMS
checking. LONGCALL calls a routine using a FAR CALL instead of
the more normal INT functions.
Since TASM 2.0 supports the 486, the XADD macro is now
commented out.
Changes by Andrew Rossmann v 1.0 (6/25/90)
All: Due to confusions with Steve Grant continuing to release
new versions, I am renaming my version to INFOPLUS.
The documentation for SYSID 4.4 states that the source and
object code are public domain. SG's newer versions are
copywritten, and no longer contain source code. All
of my versions WILL ALWAYS continue to contain the source
code, and WILL ALWAYS be public domain!! Like FRACTINT, I
hope for this program to slowly grow as people add their own
ideas.
All: Changed colors to blue background, with text of LightGray
and LightCyan.
You now press the DOWN ARROW key to continue a long screen.
This lets you move to another page without looking at all
the information.
A title telling what is on the page appears at the bottom.
Page 0: New page, shows contents of each page. HOME always goes here.
Also, this eliminates the chance that scanning the ROM could
lock up the program, preventing its use. If that happens, you
can just go to another page directly.
Page 2: The V20/V30 should be correctly detected.
Page 3: Added detection of XMS, DPMI (untested) and VCPI.
DPMI detection is tentative, since it's still in the formative
stages. Windows 3.0 is the first program to support it.
Under EMS 4, handle names will be listed (not all programs
give their handles names.)
Page 9: Added count of FCB's (File Control Blocks.)
Added count of Dos STACKS (DOS 4 only.)
Page 10:Added detection of DOS (should always be YES!!)
Added detection of JP Software's 4DOS (and other info on it.)
Added detection of CD-ROM extensions.
Page 13:For default drive, tells what type it is, if it is removable,
has doorlock detection, is JOIN'd, or SUBST'd (and shows the
true path.)
Page 17:New page, just thanks you for using INFOPLUS, and gives
some places I can be reached.
The following are my versions of SYSID. It's very likely that some
of these version numbers may overlap Steve Grant's!!
Changes by Andrew Rossmann (5/28/90) v 5.1
Page 2: Corrected bug in 386 detection code actually in SYSID.ASM.
This bug could cause lock-ups or exception errors on
386, 386SX, or 486 systems.
Changes by Andrew Rossmann (1/19/90) v 5.0
Page 9: Added buffer count (and read-ahead) for DOS 4!
Page 15:Added several more partition types.
Changes by Andrew Rossmann (12/31/89) v 4.8
Page 2: Added recognition (I hope) of the '486. This involved
replacing the code that detects the 286/386 with newer
code by Robert Collins.
Page 5: Added the extended modes of the Paradise VGA boards.
(Considering that the CRT unit forces the screen into a
standard text mode at startup, most of the entries are
sort of useless right now.)
Page 15:Added Disk Manager, Speedstore, Split (?), and GB (?) to
the recognized partition types.
Corrected some erroneous numbers in the starting cylinder
and sector.
Changes by Andrew Rossmann (9/30/89) v 4.7
All: Re-compiled under Turbo Pascal 5.5.
Page 9: Reduced the number of files to 255 due to a potential bug
under DOS 3.30.
Page 10:Added a few extra programs to the multiplex listings. Some of
these will probably only be recognized under DOS 4 or newer.
Changes by Andrew Rossmann (7/2/89) v 4.6
Page 9: Under DOS 3.30 or newer, use a new DOS function call to set
up space for up to 256 files. The number of total files
available will now be closer to the FILES= setting, instead
of being stuck at the 20 that is the default size.
Changes by Andrew Rossmann (6/7/89)
Changed the version to 4.5.
ALL: Pressing <ENTER> will bring up a 'Page no.=>' prompt. You
can then type in the number of the page you want to jump to.
Page 9: Added OEM (Original Equipment Manufacturer) Serial number to
list. Not everyone uses this. Put file handle table sizes
on one line.
Changed current time to conform to the format of the current
country code, including 12-hour format.
Page16: Corrected more DOS4 errors in the Dos Disk Parameter Block
listings. It seems the number of sectors per FAT has grown
from 1 byte to 1 word. This pushed the sector # of directory,
address of device header, media descriptor, access check,
and next block pointer up 1 byte.
Changes by Andrew Rossmann (6/3/89):
I upped the letter to 'V'.
Page 9: Added 'Boot disk was'. This will display which disk drive
was used to boot the computer under DOS 4. Under DOS 3,
this will always display (unknown).
I changed the file used to figure out the number of available
file handles from COMMAND.COM to the program itself. This is
because COMMAND.COM may not always be available (especially
if you are on floppies.) It's more likely the program itself
will be available. This uses a DOS 3 or 4 feature that lets
a program know it's full pathname. (ParamStr(0) in TP5).
Page10: Added extra information about the APPEND command under DOS
4. Will display which switches are active, and the path
currently in use.
Page15: Added several parition types, including 'Big DOS-16' (DOS 4
partitions >32M), XENIX, NOVELL, PCIX, CP/M, and BBT.
Page16: Modified the DISKREAD portion of SYSID.ASM for use under
DOS 4. Will now correctly read disks with partitions >32M.
Disks >32M under Compaq DOS 3.31 will probably not work.
Changes by Andrew Rossmann (5/29/89):
I named this version 4.44U for 'U'pdated. Don't know of any
newer versions.
Page 1: The search for the Copyright message in ROM has been
expanded to start at F000:C000 instead of F000:E000. This
test had returned nothing on my Tandy 1000SX. This may
cause longer waits on other machines. NOTE: the CPUID
routine does not detect my V20. It returns an unknown type.
I have no idea why, and I would like to know why. (I also
have an 8087 if that makes any difference.)
Page 7: If you had a mouse, some information about your keyboard
ended up scrolling off the screen. Some related info
has been put on 1 line to prevent scrolling.
Page 9: The method to find the number of available file handles
had a minor problem. It used the COMSPEC environment
variable to continuously open COMMAND.COM until an error
occurred, presumably a 'Too Many open files' error. BUT,
the files were opened as untyped files, which have a
default FileMode = 2 (read/write). If your copy of
COMMAND.COM was read-only (for virus protection), an
error 5 (Access denied) occured, and 0 available handles
was shown. This is fixed by setting FileMode:=0 (Read only)
before opening files.
The printer echo status is supported under DOS 4.
The number of buffers is still unknown as I am unable to
figure it out. Since DOS 4 can have a huge number of
buffers, and even put them in EMS memory, finding out their
count may be difficult.
Page13: Information about a floppy now supports the parameters
for 3.5" drives, and displays number of tracks.
Page16: Now supports DOS 4.0 and Compaq DOS 3.31 with partitions
>32M (maybe others if they follow the same format.)
If you have a disk formatted under DOS 4, the boot
record section will display the volume label, serial number,
and FAT type of the disk.
Known bugs: (Steve Grants ver 4.44)
1) Page 2: The CPU test for interrupts of multi-prefix string
instructions is reliable only on machines whose clock speeds are
less than about 15 MHz. The 8086 and 8088 are the only CPU's that
don't handle such interrupts correctly, however, and they aren't
(to my knowledge) ever run at anything like 15 MHz. (This 15 MHz
limit assumes that the timer tick interrupt occurs at the standard
rate of 18.2 Hz. SYSID could check that, too, I suppose.)
2) Page 5: The description of foreground color will not mention the
blinking attribute, even if it was enabled before you invoked
SYSID.
3) Page 10: SYSID used to report incorrectly the statuses of some of
the executable files which use the "multiplex interrupt" (INT
2FH). I have commented these status checks out of the source
code, determined to do battle with them another day. Can anyone
supply the correct INT 2FH functions for these files? Or are some
of them red herrings that simply check INT 2FH to see if *other*
files have been loaded (e.g. APPEND/ASSIGN)?
4) The error beep sometimes fails to sound when you press PgDn
while on the last page (or PgUp while on the first page). I have no
idea why. (!!corrected in version 4.5)