Looking for volunteers to help test a new benchmark

evildragon · Nov 30, 2011

Here's my output.

Code:

;Data collected by: TOPBENCH | Benchmark and detection stub | Version 0.96
;This file contains fingerprinting information about your computer.  Please
;email this file to trixter@oldskool.org with a subject line of "Benchmark" to
;help test these routines and seed the TOPBENCH database.

[UIDB69D2DD]
MemoryTest=1214
OpcodeTest=602
VidramTest=949
MemEATest=778
3DGameTest=571
Score=12
CPU=NEC V30
CPUspeed=8 MHz
BIOSinfo=00F2092 COPR. IBM 1981, 1987    PARITY CHECK 1 (06/26/87, rev. 0)
MachineModel=01FA
BIOSdate=19870626
BIOSCRC16=B69D
VideoSystem=MCGA
VideoAdapter=MCGA
Machine=PS/2 Model 25/25L

Trixter · Nov 30, 2011

Chuck(G) said:
It would make testing a bit easier if you'd open OUTPUT.INI for "append" rather than "create". If I could add a one-line comment to each run, that would let me test more systems quickly.

I know; I was putting it off until someone complained :-D Okay, new version uploaded to http://dosbenchmark.wordpress.com/downloads/ that does just that. You can specify the filename, or leave it blank; it will append to existing files, or create a new one if it doesn't exist. You can add a description on the command-line. You can also skip dodgy detection routines that may hang clones. Use "-?" to get command-line usage.

deathshadow · Nov 30, 2011

wouldn't it be easier to say... use that "unique" UID to build the filename? :D

luckybob · Nov 30, 2011

Chuck(G) said:
It would make testing a bit easier if you'd open OUTPUT.INI for "append" rather than "create". If I could add a one-line comment to each run, that would let me test more systems quickly.

better yet, use a time stamp as a file name. Or something based off the bios info. I've been known to fiddle with settings in the bios to get different results for benchmarks.

hexsane · Dec 4, 2011

8Mhz V20 PCjr:

;Data collected by: TOPBENCH | Benchmark and detection stub | Version 0.95
;This file contains fingerprinting information about your computer. Please
;email this file to trixter@oldskool.org with a subject line of "Benchmark" to
;help test these routines and seed the TOPBENCH database.

[UID44F2A2B]
MemoryTest=1440
OpcodeTest=873
VidramTest=1273
MemEATest=1060
3DGameTest=805
Score=9
CPU=NEC V20
CPUspeed=8 MHz
BIOSinfo=1504037 COPR. IBM 1981,1983PS (06/01/83, rev. 0)
MachineModel=0000
BIOSdate=19830601
BIOSCRC16=44F2
VideoSystem=CGA
VideoAdapter=IBM PCjr
Machine=PCjr

Trixter · Dec 4, 2011

hexsane said:
8Mhz V20 PCjr

Indeed, the timings check out! How was this made possible? Did you replace the crystal, or was there some old retail product that did it?

hexsane · Dec 4, 2011

Its a dual 8284 turbo switch hack to use two crystals one at 14 Mhz and one at 24 Mhz. I could probably get it to go a little faster but a 29Mhz crystal was to much for the floppy controller and I don't have anything between 24 and 29.

I saw a request for 8088 PCjr. I have an 8Mhz 8088 so I'll do those when I get a chance.

-Matt

Cimonvg · Dec 5, 2011

hello
some results with the -s switch on version 0.97.
xt motherboard + inboard + rev to 486 (x1 clock), ATI-wonder:
;Data collected by: TOPBENCH | Benchmark and detection stub | Version 0.97
;This file contains fingerprinting information about your computer. Please
;email this file to trixter@oldskool.org with a subject line of "Benchmark" to
;help test these routines and seed the TOPBENCH database.

[UID1F689ED]
MemoryTest=430
OpcodeTest=196
VidramTest=950
MemEATest=227
3DGameTest=190
Score=24
BIOSinfo=Copyright (c) 1984,1985,1986 Phoenix Technologies Ltd (09/24/87, rev. 0)
MachineModel=0000
BIOSdate=19870924
BIOSCRC16=1F68
VideoSystem=VGA
Machine=PC/XT

the same PC , but with TIGA vga:
;Data collected by: TOPBENCH | Benchmark and detection stub | Version 0.97
;This file contains fingerprinting information about your computer. Please
;email this file to trixter@oldskool.org with a subject line of "Benchmark" to
;help test these routines and seed the TOPBENCH database.

[UID1F68544]
MemoryTest=124
OpcodeTest=115
VidramTest=1211
MemEATest=158
3DGameTest=97
Score=38
BIOSinfo=Copyright (c) 1984,1985,1986 Phoenix Technologies Ltd (09/24/87, rev. 0)
MachineModel=0000
BIOSdate=19870924
BIOSCRC16=1F68
VideoSystem=VGA
Machine=PC/XT

..then ATIwonder again with cpu (x3 clock, driver enablet):

;Data collected by: TOPBENCH | Benchmark and detection stub | Version 0.97
;This file contains fingerprinting information about your computer. Please
;email this file to trixter@oldskool.org with a subject line of "Benchmark" to
;help test these routines and seed the TOPBENCH database.

[UID1F6837C]
MemoryTest=126
OpcodeTest=115
VidramTest=1036
MemEATest=157
3DGameTest=96
Score=42
BIOSinfo=Copyright (c) 1984,1985,1986 Phoenix Technologies Ltd (09/24/87, rev. 0)
MachineModel=0000
BIOSdate=19870924
BIOSCRC16=1F68
VideoSystem=VGA
Machine=PC/XT

...score 42 on a xt motherboard.. hmmm
/cimonvg

Trixter · Dec 6, 2011

Something's not right with those results -- if you only changed to a TIGA in the second result, then why so much faster? Was the first result at the normal clock, and the second and third with the x3 clock?

EDIT: Yes, that must be it. It looks like you tested ATI x1, then TIGA x3, then ATI x3. I've recorded the results as such, thanks. And I agree, that is probably the fastest result from a 5150/5160 board anyone will ever see

Cimonvg · Dec 6, 2011

Trixter said:
Something's not right with those results -- if you only changed to a TIGA in the second result, then why so much faster? Was the first result at the normal clock, and the second and third with the x3 clock?

EDIT: Yes, that must be it. It looks like you tested ATI x1, then TIGA x3, then ATI x3. I've recorded the results as such, thanks. And I agree, that is probably the fastest result from a 5150/5160 board anyone will ever see

hello. yes, did notise it myself.
Test results with TIGA : boot on floppy (clean DOS) cpu ~16Mhz : score 23.
Boot on harddisk with driver for the CPU (which enables the onboard cache) cpu ~16 Mhz: score 31
Boot on harddisk with driver for CPU and x3 clock (cpu ~50Mhz) score 38.
Have read that a "rev to 486" x4 exists. But for the time beeing x3 is just about wery fine with me!

/cimonvg

evildragon · Dec 7, 2011

I have a new fixed disk now on my model 25. should i re-take this test, or will it not matter?

I went from a 128MB CF card, to a 2GB disk on module.

Trixter · Dec 7, 2011

No, it's a CPU/memory/video memory speed test only, but thanks for asking.

aitotat · Dec 10, 2011

I tested my 286 and 486 gaming systems and here are the results. I have tested with version 0.97.

The 286 system has the following specs:

Mobo: Biostar MB-1216H
CPU: Siemens 80286 16 MHz + Intel 80287 10 MHz
RAM: 4 MiB (zero wait state)
Graphics: Ati VGA Stereo F/X with 1 MiB RAM

First the results with Turbo enabled:

Code:

[UIDBA8810A1D4]
MemoryTest=393
OpcodeTest=152
VidramTest=290
MemEATest=183
3DGameTest=162
Score=42
CPU=Intel 80286
CPUspeed=15 MHz
BIOSinfo=R(C)1985-1990, American Megatrends Inc.,All Rights Reserved.1346 Oakbrook Drive, Suite-120, Norcross, GA-30093. Phone-(404)-263-8181. (06/13/90, rev. 0)
MachineModel=01FC
BIOSdate=19900613
BIOSCRC16=BA88
VideoSystem=VGA
VideoAdapter=VGA, ATI 28800-5VGAWonderBios : 0.16, 256kb Video Memory (BIOS)
Machine=AT clone

And with Turbo disabled:

Code:

[UIDBA8810AB9B]
MemoryTest=773
OpcodeTest=304
VidramTest=319
MemEATest=370
3DGameTest=322
Score=24
CPU=Intel 80286
CPUspeed=8 MHz

My 486 has the following specs:

Mobo: Asus VL/I-486SV2GX4
CPU: Intel 486DX4 100 MHz (WB cache)
RAM: 32MiB 60 ns
Graphics: #9 GXE64 (S3 Vision864, 2 MiB, VLB)

Full speed results:

Code:

[UID99F1128C8D]
MemoryTest=97
OpcodeTest=38
VidramTest=99
MemEATest=41
3DGameTest=24
Score=267
CPU=Intel i486DX4
CPUspeed=100 MHz
BIOSinfo=,d Software Inc. Awar,Copyright (C) 1984-94, Award Software, Inc. (01/11/95, rev. 0)
MachineModel=01FC
BIOSdate=19950111
BIOSCRC16=99F1
VideoSystem=VGA
VideoAdapter=VGA, S3 unknown, VESA, 256kb Video Memory (BIOS)
Machine=AT clone

Turbo disabled:

Code:

[UID99F1129028]
MemoryTest=238
OpcodeTest=83
VidramTest=274
MemEATest=119
3DGameTest=76
Score=111
CPU=Intel i486DX4
CPUspeed=100 MHz

Turbo enabled but internal cache disabled with icd.exe:

Code:

[UID99F11294FA]
MemoryTest=205
OpcodeTest=65
VidramTest=156
MemEATest=88
3DGameTest=63
Score=85
CPU=Intel i486DX4
CPUspeed=100 MHz

And last Turbo and cache disabled:

Code:

[UID99F1129882]
MemoryTest=667
OpcodeTest=219
VidramTest=476
MemEATest=290
3DGameTest=200
Score=27
CPU=Intel i486DX4
CPUspeed=40 MHz

Trixter · Dec 10, 2011

aitotat said:
And last Turbo and cache disabled:

Roughly the speed of a 286-12

Thanks.

hexsane · Dec 15, 2011

A slew of stuff:

PCjr 8088 @ 4.77Mhz

Code:

[UID44F2F4A]
MemoryTest=3615
OpcodeTest=1688
VidramTest=2586
MemEATest=1843
3DGameTest=1792
Score=4
CPU=Intel 8088
CPUspeed=4.77 MHz
BIOSinfo=1504037 COPR. IBM 1981,1983PS (06/01/83, rev. 0)
MachineModel=0000
BIOSdate=19830601
BIOSCRC16=44F2
VideoSystem=CGA
VideoAdapter=IBM PCjr
Machine=PCjr

PCjr 8088 @ 8Mhz

Code:

[UID44F2897]
MemoryTest=2157
OpcodeTest=1007
VidramTest=1601
MemEATest=1102
3DGameTest=1068
Score=7
CPU=Intel 8088
CPUspeed=8 MHz
BIOSinfo=1504037 COPR. IBM 1981,1983PS (06/01/83, rev. 0)
MachineModel=0000
BIOSdate=19830601
BIOSCRC16=44F2
VideoSystem=CGA
VideoAdapter=IBM PCjr
Machine=PCjr

Amiga 80286 Bridgeboard with MakeIt 486 @ 33Mhz No Cache:

Code:

[UIDC60716BEE1]
MemoryTest=786
OpcodeTest=381
VidramTest=695
MemEATest=495
3DGameTest=375
Score=18
CPU=Cyrix/TI 486-class processor
CPUspeed=33 MHz
BIOSinfo=A2286 BIOS, COPYRIGHT (C) 1988  Commodore Business Machines  Version 3.6 (09/xx/88, rev. 0)
MachineModel=01FC
BIOSdate=19880900
BIOSCRC16=C607
VideoSystem=CGA
VideoAdapter=CGA
Machine=AT clone

Amiga 80286 Bridgeboard with MakeIt 486 @ 33Mhz With Cache:

Code:

[UIDC60716D4DF]
MemoryTest=559
OpcodeTest=295
VidramTest=734
MemEATest=404
3DGameTest=275
Score=23
CPU=Cyrix/TI 486-class processor
CPUspeed=33 MHz
BIOSinfo=A2286 BIOS, COPYRIGHT (C) 1988  Commodore Business Machines  Version 3.6 (09/xx/88, rev. 0)
MachineModel=01FC
BIOSdate=19880900
BIOSCRC16=C607
VideoSystem=CGA
VideoAdapter=CGA
Machine=AT clone

hexsane · Dec 15, 2011

vwestlife said:

Results from my NEC PowerMate SX/16:

Code:

;Data collected by: TOPBENCH | Benchmark and detection stub | Version 0.95
;This file contains fingerprinting information about your computer.  Please
;email this file to trixter@oldskool.org with a subject line of "Benchmark" to
;help test these routines and seed the TOPBENCH database.

[UIDC63F10E08A]
MemoryTest=425
OpcodeTest=172
VidramTest=320
MemEATest=216
3DGameTest=175
Score=32
CPU=AMD Am386SX
CPUspeed=16 MHz
BIOSinfo=Copyright (C) 1985-1990 Phoenix Technologies Ltd. (01/15/88, rev. 0)
MachineModel=01FC
BIOSdate=19880115
BIOSCRC16=C63F
VideoSystem=VGA
VideoAdapter=VGA, Paradise PVGA1A, 16-Bit-Video, 8-Bit-ROMMulti-Sync, 256kb V
Machine=AT clone

The 8Mhz bus bandwidth on the Bridgeboard is the bottleneck. I should be getting performance close to the above with the Make It 486 @33 Mhz :/
If anyone has a 386 Bridgeboard I'd like to see those numbers.

-Matt

Vanilla BB:

Amiga 80286 Bridge Board:

[UIDC6074502E]
MemoryTest=924
OpcodeTest=398
VidramTest=684
MemEATest=517
3DGameTest=391
Score=17
CPU=Intel 80286
CPUspeed=8 MHz
BIOSinfo=A2286 BIOS, COPYRIGHT (C) 1988 Commodore Business Machines Version 3.6 (09/xx/88, rev. 0)
MachineModel=01FC
BIOSdate=19880900
BIOSCRC16=C607
VideoSystem=CGA
VideoAdapter=CGA
Machine=AT clone

Great Hierophant · Dec 21, 2011

Trixter posted these results from his XT with CGA and PCjr. in another post on this forum, but they should be here for easy reference :

4.77MHz 8088 IBM PC/XT w/CGA

Code:

[UID9485166A30]
memory_test=3774
opcode_test=1753
vidram_test=2652
mem_ea_test=1935
3dgame_test=1851
score=4
machine=PC/XT (enhanced)
cpu=Intel 8088
cpuspeed=4.77 MHz
biosinfo=62X0851 COPR. IBM 1986 (01/10/86, rev. 1)
biosdate=19860110
bioscrc16=9485
videosystem=CGA
videoadapter=CGA

4.77MHz 8088 128K IBM PCjr.

Code:

[UID7F5C71D]
memory_test=5926
opcode_test=3584
vidram_test=3373
mem_ea_test=4392
3dgame_test=3490
score=2
machine=PCjr
cpu=Intel 8088
cpuspeed=4.77 MHz
biosinfo=1504037 COPR. IBM 1981,1983PS (06/01/83, rev. 86)
biosdate=19830601
bioscrc16=7F5C
videosystem=CGA
videoadapter=IBM PCjr

If a stock PCjr. gets a score of 2, what kind of PC Compatible would get a 1? Also, other than in systems like the Jr., am I correct in that adding more memory would not increase the scores? I mean at least until you get to those systems that allow you to fiddle with memory timings.

Comparing both results to Hexane's PCjr., it is amazing what adding the extra memory can do. The Jr. is a real dog without extra memory. Once the memory is added, it is slightly faster than an XT according to the benchmarks.

Trixter · Dec 21, 2011

Great Hierophant said:
If a stock PCjr. gets a score of 2, what kind of PC Compatible would get a 1?

I don't think there is one

although I hope someday that someone submits timings for a Sanyo MBC-550 (http://www.old-computers.com/museum/computer.asp?st=1&c=473) which runs at 3.6MHz instead of 4.77MHz. I am guessing that will have a score of 3 instead of the common 4 for most PC/XT clones.

(Note that I did not calibrate the score to correlate with 8088 MHz speed in any way; it is just a coincidence.)

Also, other than in systems like the Jr., am I correct in that adding more memory would not increase the scores? I mean at least until you get to those systems that allow you to fiddle with memory timings.

Right, adding memory to regular systems shouldn't increase the scores. It works on PCjr because the first 128KB of PCjr memory is very slow due to added wait states (so the display circuitry can access it as well as the CPU). Running program code in the first 128KB is like trying to cross the street as a crossing guard is constantly interrupting you and telling you to wait a second. When you add more memory to a PCjr, the extra memory doesn't have extra wait states added, so as long as you can force programs to run in it (using a memory driver like JRMEM.SYS or equivalent), you should see speeds roughly equal to that of a regular PC.

Comparing both results to Hexane's PCjr., it is amazing what adding the extra memory can do. The Jr. is a real dog without extra memory. Once the memory is added, it is slightly faster than an XT according to the benchmarks.

Make sure you're not confusing with his second set of numbers, which are for an 8MHz crystal modification.

Great Hierophant · Dec 21, 2011

Let me add my humble contributions to the common knowledge

Tandy 1000SX w/ 286 Express Card (80286 @ 7.16MHz), Cache On (8KB)

Code:

[UID9890754]
MemoryTest=1321
OpcodeTest=1276
VidramTest=1232
MemEATest=1682
3DGameTest=1153
Score=11
CPU=Intel 80286
CPUspeed=7.16 MHz
BIOSinfo=unknown
MachineModel=0000
BIOSdate=19860714
BIOSCRC16=9890
VideoSystem=CGA
VideoAdapter=Tandy 1000
Machine=Tandy 1000

Tandy 1000SX w/ 286 Express Card (80286 @ 7.16MHz), Cache Off

Code:

MemoryTest=2877
OpcodeTest=1547
VidramTest=1471
MemEATest=2054
3DGameTest=1428
Score=5
CPU=Intel 80286
CPUspeed=6 MHz
BIOSinfo=unknown
MachineModel=0000
BIOSdate=19860714
BIOSCRC16=9890
VideoSystem=CGA
VideoAdapter=Tandy 1000
Machine=Tandy 1000

Tandy 1000SX 8088 @ 7.16MHz

Code:

[UID9890FEE]
MemoryTest=2732
OpcodeTest=1392
VidramTest=1731
MemEATest=1584
3DGameTest=1435
Score=6
CPU=Intel 8088
CPUspeed=7.16 MHz
BIOSinfo=unknown
MachineModel=0000
BIOSdate=19860714
BIOSCRC16=9890
VideoSystem=CGA
VideoAdapter=Tandy 1000
Machine=Tandy 1000

Tandy 1000SX 8088 @ 4.77MHz

Code:

[UID989015B1]
MemoryTest=3742
OpcodeTest=1770
VidramTest=2082
MemEATest=1962
3DGameTest=1883
Score=4
CPU=Intel 8088
CPUspeed=4.77 MHz
BIOSinfo=unknown
MachineModel=0000
BIOSdate=19860714
BIOSCRC16=9890
VideoSystem=CGA
VideoAdapter=Tandy 1000
Machine=Tandy 1000

IBM PC 5150 8088 @ 4.77MHz

Code:

[UID3B8536B]
MemoryTest=3774
OpcodeTest=1753
VidramTest=2648
MemEATest=1935
3DGameTest=1851
Score=4
CPU=Intel 8088
CPUspeed=4.77 MHz
BIOSinfo=1501476 COPR. IBM 1982 (10/27/82, rev. 0)
MachineModel=0000
BIOSdate=19821027
BIOSCRC16=3B85
VideoSystem=CGA
VideoAdapter=CGA
Machine=PC

Interesting comparisons between three 8088 4.77MHz machines :

My 5150 and Trixter's 5160 give exactly the same scores in 4 out of 5 tests and the 5th score the difference is negligible (I am assuming we were both using IBM CGA cards). The IBM vs. Tandy scores show variances of less than 2% at best, except for the vidram score, where the difference is 27%.

If the video card were changed to an MDA, Hercules, EGA or VGA card, wouldn't that score change? Also, if that score changed significantly, the overall score may change as well.

DOS lives on!! · Dec 21, 2011

IBM EduQuest 55.

Code:

[UID9B48F1B23]
MemoryTest=89
OpcodeTest=49
VidramTest=144
MemEATest=68
3DGameTest=34
Score=177
CPU=Intel i486DX2
CPUspeed=66 MHz
BIOSinfo=Copyright (C) 1989-1994 Phoenix Technologies Ltd. (05/25/94, rev. 0)
MachineModel=01FC
BIOSdate=19940525
BIOSCRC16=9B48
VideoSystem=VGA
VideoAdapter=VGA, unknown Cirrus CL-GD54, VESA, 256kb Video Memory (BIOS)
Machine=AT clone

Now to the interesting part. The instant TOPBENCH finished writing the output file to the floppy disk, all the text became garbled on the screen. I don't know what went wrong, but a restart fixed it.

Here's what it looked like.

Looking for volunteers to help test a new benchmark

Veteran Member

Veteran Member

Veteran Member

Veteran Member

Experienced Member

Veteran Member

Experienced Member

Experienced Member

Veteran Member

Experienced Member

Veteran Member

Veteran Member

Experienced Member

Veteran Member

Experienced Member

Experienced Member

Veteran Member

Veteran Member

Veteran Member

Veteran Member