• Please review our updated Terms and Rules here

Looking for volunteers to help test a new benchmark

Here's the result one of my other 486DX2/66s-An Acer Across

Code: 
MemoryTest=80
OpcodeTest=41
VidramTest=138
MemEATest=41
3DGameTest=30
Score=189
BIOSinfo=unknown
MachineModel=01FC
BIOSdate=19910701
BIOSCRC16=EB42
VideoSystem=VGA
Machine=AT clone
 
Trixter said:
Hey, this answers another long-time question of mine: A 386DX-40 really is faster than a 486sx-25. Thanks!
Click to expand...
hello.
My clone amd386 40 Mhz (with IIT 487DLC-40) , only terminates topbstub with runtime error : 150 at 02f4:0082.
- tried with other vga card (tseng4000/ati-wonder24xl) , tried with other HD/floppy card (DTC2280eb/ tecram DC600CF /winboard) ? :confused:
/cimonvg
 
cimonvg did you try the -s switch? I have to use it on all my 486s to get a result. Trixter what does this switch skip in the test? What effect does this have on overall test results?
 
Last edited:
Caluser2000 said:
cimonvg did you try the -s switch? I have to use it on all my 486s to get a result. Trixter what does this switch skip in the test? What effect does this have on overall test results?
Click to expand...

hello , yes.. no..
yes, did try the -s switch (anyhow in the first setup.. not with all the different cards I did try)
no, did not make any difference :|
could it bee the co-processor IIT 487 ??
/cimonvg
 
Cimonvg said:
hello , yes.. no..
yes, did try the -s switch (anyhow in the first setup.. not with all the different cards I did try)
no, did not make any difference :|
could it bee the co-processor IIT 487 ??
/cimonvg
Click to expand...
The test is not ment to test co-pros but as an experiment it would be interesting to see it it does make a difference by removing it.
 
Caluser2000 said:
A bit more detail on both systems would be usefull. A true comparison would have both systems having the same video, cache, etc as there is such a variety out there. I'll see if I can get hold of a similar system in the next week or so and see how that compares.

Here's Chucks results again:

MemoryTest=183
OpcodeTest=86
VidramTest=317
MemEATest=95
3DGameTest=77
Score=77
CPU=AMD Am386DX
CPUspeed=40 MHz
BIOSinfo=R(C)1985-1991,American Megatrends Inc.,All Rights Reserved.,1346 Oakbrook Dr.,#120,GA-30093,USA.(404)-263-8181. (12/12/91, rev. 0)
MachineModel=01FC
BIOSdate=19911212
BIOSCRC16=50F6
VideoSystem=VGA
VideoAdapter=VGA, Trident TR9000i, VESA, 768kb Video Memory, 512kb Video Memo
Machine=AT clone

Not much in it at all. I see another 386DX40 got a score of 96. It would interesting to see what the differences are between that, hardware wise, compared to Chucks. Results of that machine below:


[UID1BE22760F]
MemoryTest=186
OpcodeTest=122
VidramTest=194
MemEATest=139
3DGameTest=113
Score=96
Machine=AT clone
CPU=AMD Am386DX
CPUspeed=40 MHz
BIOSinfo=R(C)1985-1991,American Megatrends Inc.,All Rights Reserved.,1346 Oakbrook Dr.,#120,GA-30093,USA.(404)-263-8181. (12/12/91, rev. 0)
BIOSdate=19911212
BIOSCRC16=1BE2
VideoSystem=VGA
VideoAdapter=VGA, Tseng ET 4000 with HiColor RAMDAC, DRAM, 1024kb Video Memory
Machine=AT clone
Click to expand...

As Trixter said, the graphics card counts, and Tsengs were highly regarded for their speed while Tridents, not so much.

IBMMuseum said:
The Model 35SX (8535) and Model 40SX (8540) have a Texas Instruments SVGA chip with limited support. It is the same as on the Model 56SX/56SLC (8556), Model 57SX/57SLC (8557), and the microchannel IBM SVGA/A adapter. The 95xx-level of the Models 35, 40, 56, and 57 have XGA-2 on the system planar.
Click to expand...

So if I wanted true IBM VGA, I would be stuck with the 25/286, 30/286, 50, 50z, 56, 57, 60, 70, & 80.
 
Great Hierophant said:
...So if I wanted true IBM VGA, I would be stuck with the 25/286, 30/286, 50, 50z, 56, 57, 60, 70, & 80.
Click to expand...

The 8556/8557 is SVGA, the 9556/9557 is XGA-2...

Don't forget the 55SX (the 65SX also has baseboard VGA, but was released with an additional XGA-2 adapter)...

The P70 also has standard VGA out in addition to the plasma screen (the P75 is XGA)...

There is also the 8-bit VGA adapter for the XT, AT, and 8086-based Model 30...
 
Great Hierophant said:
As Trixter said, the graphics card counts, and Tsengs were highly regarded for their speed while Tridents, not so much.
Click to expand...
I do realise that but it isn't varifying that a 386DX40 is that much better than a 486SX25. From what I can see performance is quite similar given the same hardware.
 
Last edited:
Cimonvg said:
hello.
My clone amd386 40 Mhz (with IIT 487DLC-40) , only terminates topbstub with runtime error : 150 at 02f4:0082.
- tried with other vga card (tseng4000/ati-wonder24xl) , tried with other HD/floppy card (DTC2280eb/ tecram DC600CF /winboard) ? :confused:
/cimonvg
Click to expand...

Ack, sorry it's causing trouble. Grab the latest stub from dosbenchmark.wordpress.com (I patched around the AMD 386sx false positive an hour ago, the filename is the same) and run "DSTUB", not topbstub. If you still get a runtime error with DSTUB I will be able to troubleshoot it. If it still crashes, use "-s" which bypasses the CPU model and speed detection routines. I obviously haven't tested them on every combo, and the info for those routines came from several different libraries and sources (even some old USENET archives) so it is always a possibility the CPU routines will crash.
 
Last edited:
Caluser2000 said:
Score=77

Score=96
Click to expand...

I agree the two systems should have been more similar; Chuck's CPU and memory tests are individually faster than the second machine, yet it has a lower score. The Score is the number of times the code of each test can run once in a 50ms period, so for some reason, on Chuck's machine, it ran 77 times whereas it ran 96 times on the second. Based on the individual test timings, the 77 score is the more accurate one; the 96 score seems too high. (A rough ballpark for what the Score should be on 486s and lower is to add up the five microsecond timing numbers and divide 50,000 by that sum.)

I intentionally do not disable interrupts when calculating the score, so maybe if it wasn't a clean boot and there were things like a packet driver loaded, it might have affected the Score calculation? Without having both systems in front of me there's really no way to tell. At least both systems are somewhat comparable.

(For the worried, I do disable interrupts when calculating the microsecond timings, so they are as accurate as can be within the constraints of a system performing regular DRAM refresh cycles.)
 
Trixter said:
Hey, this answers another long-time question of mine: A 386DX-40 really is faster than a 486sx-25. Thanks!
Click to expand...

I want to see where an IBM 486SLC3-75 compares to the Intel/AMD 486 class...

And the difference between a 386SLC-20 (with 8Kb of L1 cache) to a 386SX-20...

486DX2-66 "Enhanced" model with the 8Kb of L1 cache set to Write-Back compared to Write-Through...

486DX4-100 with 16Kb L1 compared to 486DX4-100 with 8Kb L1...

The list goes on, and I realize I am one that could do all of those tests...
 
hello - happy new year
with version 0.97b and the IBM 8535-043 , looks like this:
;Data collected by: TOPBENCH | Benchmark and detection stub | Version 0.97b
;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.

[UID4F258C]
MemoryTest=434
OpcodeTest=158
VidramTest=282
MemEATest=209
3DGameTest=156
Score=40
CPU=Intel 80386SX
CPUspeed=20 MHz
BIOSinfo= COPR. IBM 1981, 1991 (09/25/91, rev. 10)
MachineModel=19F8
BIOSdate=19910925
BIOSCRC16=04F2
VideoSystem=VGA
VideoAdapter=VGA, unknown Chipset, 256kb Video Memory (BIOS)
Machine=PS/2 Model 90/95?
/cimonvg
 
I've released all of the submissions I've collected at http://dosbenchmark.wordpress.com/downloads/ along with the source code blocks for the five metric sections. Development on the full tool is going slower than I thought it would, so I thought it would be reasonable to release what's been collected so far.

I really want the tool to have a nice professional interface, so I chose to finally learn Turbo Vision. TV is not just a CUI library, but an entire event-driven OOP framework, and learning it from scratch is taking a while. If it's any consolation, I've completed the tutorial, and the main system entry dialog is finished. Sorry it's taking so long.

dlgdsn_001.jpg
 
Managed to get another video card sorted, just pressed hard on the vram and it boots fine-

Code: 
MemoryTest=209
OpcodeTest=132
VidramTest=269
MemEATest=143
3DGameTest=94
Score=66
CPU=AMD Am386SX
CPUspeed=33 MHz
BIOSinfo=R(C)1985-1991,American Megatrends Inc.,All Rights Reserved.,1346 Oakbrook Dr.,#120,GA-30093,USA.(404)-263-8181. (12/12/91, rev. 0)
MachineModel=01FC
BIOSdate=19911212
BIOSCRC16=2115
VideoSystem=VGA
VideoAdapter=VGA, Trident TR9000, VESA, 768kb Video Memory, 512kb Video Memor
Machine=AT clone
Almost three times the original score.

One question. Why the two vram sizes Trixter? The video card posts as 512kb vram. I noticed the same on Chucks output too. Could this just be a Trident detection issue?

It's certainly interesting seeing the improvements.
 
Last edited:
Neoware CA19 thin client (run from a USB flash drive, MS-DOS 6.22):

Code: 
;Data collected by: TOPBENCH | Benchmark and detection stub | Version 0.97b
;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.

[UIDAA7921484]
MemoryTest=5
OpcodeTest=4
VidramTest=122
MemEATest=2
3DGameTest=3
Score=369
CPU=Unknown 686-class CPU (Make : CentaurHauls)
CPUspeed=400 MHz
BIOSinfo=es, Ltd. Phoenix T.Copyright (C) 2003, Phoenix Technologies, LTD (08/17/06, rev. 0)
MachineModel=01FC
BIOSdate=20060817
BIOSCRC16=AA79
VideoSystem=VGA
VideoAdapter=VGA, VESA, 256kb Video Memory (BIOS)
Machine=AT clone

What's interesting is that DSTUB says the CPU is 366MHz. It's a VIA C7 running at 400MHz.
 
Chuck(G) said:
What's interesting is that DSTUB says the CPU is 366MHz. It's a VIA C7 running at 400MHz.
Click to expand...

But the output you posted correctly shows 400Mhz...?

BTW, the CPU speed detection is based on known timings from known CPUs. If the code can't correctly detect the CPU, or has no timings for that CPU, the MHz speed is a best guess which will likely be wrong.
 
Last edited:
Caluser2000 said:
One question. Why the two vram sizes Trixter? The video card posts as 512kb vram. I noticed the same on Chucks output too. Could this just be a Trident detection issue?
Click to expand...

Easily. It is nearly impossible to non-destructively probe for all video cards and their features accurately. What the code tries to do is first identify the chipset, then go deeper. Sometimes you can only go so deep before you risk hanging the system, card, or both.

This is why all of the detection is optional in the full tool -- you can click the DETECT button for various sections and hopefully it will autofill with something helpful. If it hangs, just reboot and don't click DETECT :)

Unfortunately, it would be very difficult for me to fix the Trident detection code for your particular card without access to the card. I got the chipset right without crashing the system, and I'm happy to leave it at that :)
 
You must log in or register to reply here.
Back
Top