home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
High Voltage Shareware
/
high1.zip
/
high1
/
DIR4
/
VIZ431.ZIP
/
VIZ.DAT
< prev
next >
Wrap
Text File
|
1993-10-30
|
12KB
|
226 lines
╒════════════════════════════════════════════╕
│ VIz │
│ Revision 4.31.0 │
│ │
│ Copyright (C) 1989-1993 FM de Monasterio │
│ Licensed Material. All rights reserved │
╘════════════════════════════════════════════╛
╦═╗
║ ║╔═╗╠═ ╔═╗
║ ║╔═╣║ ╔═╣
╩═╝╚═╝╚═╝╚═╝
BENCHMARKS
Benchmark timings were obtained with the utility VIZMARK.EXE, version 2.30,
which measures the performance of several video input and output functions,
and which is BBS-distributed as the archive VIZMK230.ZIP.
The first part of VIzMARK writes text to the video via several DOS and BIOS
services. Each category writes 100 lines of 80-characters starting from the
line at the top of the screen. (In a 25-line display screen, the categories
produce timings with a scrolling bias of 3:1). Window writing and scrolling
and other BIOS video functions are also timed. The results are expressed in
milliseconds per screen line.
The second part of VIzMARK sends a number of standard ANSI escape sequences
to the screen via interrupt 29h (Fast Console), which is used by MS-DOS for
character input/output. Such results are expressed as milliseconds per ANSI
screen line.
****
These data represent the mean of 5 sequential timing runs, after an initial
run that was discarded, expressed as an acceleration factor relative to the
timings obtained in the absence of a video accelerator. In other words, the
higher this factor the faster the actual time for a given video test. ~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The following machines were used for the testing:
1. 33-Mhz, 486 computer with a 32-bit bus, a COMPAQ BIOS, and a 16-bit VGA
planar adapter. Operating system: MS-DOS 5.00.
2. 25-Mhz, 486 computer with a 32-bit bus, an Award BIOS version 3.12.00d-
21, and a 16-bit VGA nonplanar adapter. Operating system: MS-DOS 4.01.
3. 16-Mhz, 386 computer with a 16-bit bus, a 1984-IBM AT BIOS and a 16-bit
VGA nonplanar adapter. Operating system: MS-DOS 5.00.
4. 10-Mhz, 286 computer with a 16-bit bus, a PREMIUM/286 BIOS version 1.10
and an EGA adapter. Operating system: MS-DOS 3.30.
In machines with an adjustable clock rate, the fastest rate was always used.
A 25 row, 80 column screen was used in all machines. All VIz versions being
tested used word OUT addressing.
The 386 and 486 computers had 386-type memory managers installed; to avoid a
bias in the comparisons across computers, the video BIOS was not remapped in
the 386 and 486 machines. Although these memory managers can decrease video
performance in a nontrivial manner, the performance penalty is not likely to
affect comparative testing results.
****
1. DOS AND BIOS TEXT VIDEO ACCELERATION FACTOR
(relative to benchmarks without accelerator)
CPU: 286-10Mhz 386-16Mhz 486-25Mhz 486-33Mhz
┌─────────┐ ┌─────────┐ ┌─────────┐ ┌─────────┐
│ VIz 286 │ │ VIz 386 │ │ VIz 486 │ │ VIz 486 │
│ 4.31 │ │ 4.31 │ │ 4.31 │ │ 4.31 │
╒════════════════╕ ╞═════════╡ ╞═════════╡ ╞═════════╡ ╞═════════╡
│ VIzMARK ------ │ │ │ │ │ │ │ │ │
│ DOS via StdOut │ │ 8.09 │ │ 6.67 │ │ 2.30 │ │ 2.20 │
│ CON fast INT29 │ │ 5.44 │ │ 3.06 │ │ 1.94 │ │ 1.65 │
│ ANSI sequences │ │ None │ │ None │ │ None │ │ None │
│ │ │ │ │ │ │ │ │ │
│ BIOS teletype │ │ 6.15 │ │ 3.95 │ │ 2.23 │ │ 2.10 │
│ BIOS windows │ │ 4.20 │ │ 2.89 │ │ 1.95 │ │ 1.48 │
│ BIOS miscellan │ │ 17.20 │ │ 4.62 │ │ 2.15 │ │ 2.05 │
└────────────────┘ └─────────┘ └─────────┘ └─────────┘ └─────────┘
2. COMPARISONS WITH OTHER ACCELERATORS (1)
CPU ┌──────────────────────────────────────────────────────┐
386-16 Mhz │ VIz386 QUIKCRT PCKSCRN HYPERSCR VSCREEN THEFAST WARPF│
MS-DOS 5.0 │ 4.31 (2) (3) (4) (5) (6) (7) │
╒══════════════════╕╞══════════════════════════════════════════════════════╡
│ Resident (bytes) ││ 960 1456 4592 2560 1040 912 1296 │
│ Command switches ││ none /F /A+/Q+ F:+ none none /S │
├──────────────────┤├──────────────────────────────────────────────────────┤
│ VIzMARK ------- ││ │
│ DOS via StdOut ││ 6.67 2.49 2.13 1.65 1.36 0.95 0.98 │
│ CON fast INT29 ││ 3.06 2.88 2.38 1.75 1.25 0.93 0.98 │
│ ANSI sequences ││ None None None None None None None │
│ ││ │
│ BIOS teletype ││ 3.95 3.63 2.99 1.89 1.50 0.94 1.00 │
│ BIOS windows ││ 2.89 1.86 1.70 1.65 1.80 0.96 1.00 │
│ BIOS miscellan ││ 4.62 4.20 3.26 1.21 1.71 3.37 0.96 │
└──────────────────┘└──────────────────────────────────────────────────────┘
╒══════════════════╕╒══════════════════════════════════════════════════════╕
│ Output factor (8)││ 4.25 3.01 2.49 1.63 1.52 1.45 0.98 │
│ Factor/kbytes (9)││ 4.53 2.12 0.56 0.65 1.50 1.63 0.77 │
╘══════════════════╛╘══════════════════════════════════════════════════════╛
NOTES:
(1) ZENO.COM (version 2.20) could not included in the above comparisons
because its 'teletype' service garbled the DOS prompt/command line.
(2) Copyright 1986 T. Overfield (version 1.00).
(3) Copyright 1987-89 MultiSoft Corporation (version 1.05).
(4) Copyright 1989-92 R. Cross (version 2.30); provides screen blanking.
(5) Copyright 1986-87 TheSoft Programming Services (version 1.04).
(6) Copyright 1985-87 Golden Bow Systems (version 1.06).
(7) Written by E. Tauck (version 1.00); provides partial ANSI support.
(8) Sum of the acceleration factors divided by the 5 tests.
(9) Mean video acceleration factor divided by the resident size, rounded
to the next larger 16-byte boundary by DOS, in KB (1 KB=1024 bytes).
This value provides an approximate index of the effectiveness of the
code in implementing video acceleration.
3. STANDARD ANSI VIDEO ACCELERATION FACTOR
(relative to benchmarks under ANSI.SYS)
CPU: 286-10 Mhz 386-16 Mhz 486-25 Mhz 486-33 Mhz
┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐
│ VIz 286 │ │ VIz 386 │ │ VIz 486 │ │ VIz 486 │
│ 4.31 (1) │ │ 4.31 (2) │ │ 4.31 (3) │ │ 4.31 (2) │
╒══════════════════╕ ╞══════════╡ ╞══════════╡ ╞══════════╡ ╞══════════╡
│ Command switches │ │ /A+/Z- │ │ /A+/Z- │ │ /A+/Z- │ │ /A+/Z- │
├──────────────────┤ ├──────────┤ ├──────────┤ ├──────────┤ ├──────────┤
│ VIzMARK ------- │ │ │ │ │ │ │ │ │
│ DOS via StdOut │ │ 4.15 │ │ 4.95 │ │ 2.70 │ │ 2.02 │
│ CON fast INT29 │ │ 6.85 │ │ 6.10 │ │ 2.34 │ │ 1.97 │
│ ANSI sequences │ │ 10.08 │ │ 5.20 │ │ 4.52 │ │ 2.57 │
│ │ │ │ │ │ │ │ │ │
│ BIOS teletype │ │ 6.14 │ │ 3.95 │ │ 2.23 │ │ 2.10 │
│ BIOS windows │ │ 4.20 │ │ 2.89 │ │ 1.95 │ │ 1.48 │
│ BIOS miscellan │ │ 17.20 │ │ 4.62 │ │ 2.15 │ │ 2.05 │
└──────────────────┘ └──────────┘ └──────────┘ └──────────┘ └──────────┘
NOTES:
(1) Relative to ANSI.SYS (MS-DOS version 3.30).
(2) Relative to ANSI.SYS (MS-DOS version 5.00).
(3) Relative to ANSI.SYS (MS-DOS version 4.01).
4. COMPARISONS WITH OTHER ANSI FILTERS (1)
CPU ┌─────────────────────────────────────────────────────┐
386-16 Mhz │ VIz386 NNANSI ZANSI ZAVT NANSI ANSI │
MS-DOS 5.0 │ (2) COM (3) SYS (4) SYS (5) SYS (6) COM (7) │
╒══════════════════╕╞═════════════════════════════════════════════════════╡
│ Resident (bytes) ││ 1760 3876 2720 3536 4240 2336 │
│ Command switches ││ /A+/Z- none none none none FAST/B0 │
├──────────────────┤├─────────────────────────────────────────────────────┤
│ VIzMARK ------- ││ │
│ DOS via StdOut ││ 4.95 4.95 4.36 3.20 2.82 2.23 │
│ CON fast INT29 ││ 6.10 5.87 5.11 3.68 3.17 2.36 │
│ ANSI sequences ││ 5.20 4.44 3.68 3.84 3.01 1.99 │
│ ││ │
│ BIOS teletype ││ 4.95 1.00 1.00 1.00 1.00 1.00 │
│ BIOS windows ││ 2.89 1.00 1.00 1.00 1.00 1.00 │
│ BIOS miscellan ││ 4.62 1.00 1.00 1.00 1.00 1.00 │
└──────────────────┘└─────────────────────────────────────────────────────┘
╒══════════════════╕╒═════════════════════════════════════════════════════╕
│ Output factor (8)││ 5.41 5.09 4.38 3.57 3.00 2.19 │
│ Factor/kbytes (9)││ 3.15 1.34 1.65 1.03 0.73 0.96 │
╘══════════════════╛╘═════════════════════════════════════════════════════╛
NOTES:
(1) Relative to the ANSI.SYS driver distributed with MS-DOS 5.00.
(2) Revision 4.31. The screen accelerator uses 960 bytes of memory, and
the ANSI plus Cursor display resident use 800 bytes together. Notice
that the DOS acceleration switch is off (/Z-) so that all of the DOS
output is passed to the console driver; although switch /Z+ provides
a much faster StdOut output (10.67 ms/line) it may not implement all
ANSI requests.
(3) Written by T. Almy, version 9/92 (based on NANSI.SYS 2.2 (c) 1986 D.
Kegel). Includes keyboard mapping service and graphics support. The
driver may be removed from memory in many cases. (Faster times could
be obtained with NNANSI fast mode and BIOS write_tty mode, but these
modes fail some VIzMARK tests, thus vitiating comparisons.)
(4) Copyright 1986, 1987, T. Hanlin III (version 1.2). Does not include
keyboard remapping services. Driver cannot be turned off or removed
from memory.
(5) Copyright 1989, 1990, Luns Tee (version 1.1). Also includes an Avatar
level 0 handler. Used ZAVTNK.SYS, which is smaller than ZAVT.SYS and
has the keyboard remapping service disabled. Driver cannot be turned
off or removed from memory.
(6) Copyright 1986, D. Kegel (version 3.1). Includes a keyboard remapping
service. Driver cannot be turned off or removed from memory.
(7) Copyright 1989 Ziff-Davis Corp., Author: M.J. Mefford (version 1.20).
Switch /B0 yields the smallest resident size; part of such a resident
handles keyboard remapping services (not included in the benchmarks).
Switch 'SLOW' generates an output slower than that of ANSI.SYS alone.
Resident can be turned off and removed from memory.
(8) Sum of the acceleration factors for DOS, CON, and ANSI divided by 3.
(9) Mean DOS/CON/ANSI output factor divided by the resident size, rounded
to the next larger 16-byte boundary by DOS, in KB (1 KB=1,024 bytes).
This value provides an approximate index of the effectiveness of the
code in implementing this video output subset.
----------------
┌─────────────────────────────────────────────────────────────┐
│ Trademarked names are the property of the respective owners │
└─────────────────────────────────────────────────────────────┘
[End]