• Please review our updated Terms and Rules here

Looking for volunteers to help test a new benchmark

50Z again, with an SX25LCMD (AMD 386SX-25 with Austek L2 cache module) Kingston "SX Now!" CPU upgrade:

(cache not enabled)

Code: 
[UID495E41DD]
MemoryTest=709
OpcodeTest=369
VidramTest=1297
MemEATest=510
3DGameTest=343
Score=16
CPU=Intel 386SX
CPUspeed=25 MHz
BIOSinfo=        COPR. IBM 1981, 1987 (04/18/88, rev. 3)
MachineModel=04FC
BIOSdate=19880418
BIOSCRC16=495E
VideoSystem=VGA
VideoAdapter=VGA, unknown Chipset, 256kb Video Memory (BIOS)
Machine=PS/2 Model 50Z 286-10


(cache enabled)

Code: 
[UID495E75C4]
MemoryTest=437
OpcodeTest=300
VidramTest=1268
MemEATest=416
3DGameTest=272
Score=27
CPU=Intel 386SX
CPUspeed=25 MHz
BIOSinfo=        COPR. IBM 1981, 1987 (04/18/88, rev. 3)
MachineModel=04FC
BIOSdate=19880418
BIOSCRC16=495E
VideoSystem=VGA
VideoAdapter=VGA, unknown Chipset, 256kb Video Memory (BIOS)
Machine=PS/2 Model 50Z 286-10
 
50Z again, with an SX33LCMD (AMD 386SX-33 with Austek L2 cache module) Kingston "SX Now!" CPU upgrade:

(cache not enabled)

Code: 
[UID495E4A4D]
MemoryTest=627
OpcodeTest=354
VidramTest=1289
MemEATest=486
3DGameTest=335
Score=16
CPU=Intel 386SX
CPUspeed=30 MHz [not a typo, a mismeasurement]
BIOSinfo=        COPR. IBM 1981, 1987 (04/18/88, rev. 3)
MachineModel=04FC
BIOSdate=19880418
BIOSCRC16=495E
VideoSystem=VGA
VideoAdapter=VGA, unknown Chipset, 256kb Video Memory (BIOS)
Machine=PS/2 Model 50Z 286-10


(cache enabled)

Code: 
[UID495E687D]
MemoryTest=362
OpcodeTest=284
VidramTest=1276
MemEATest=399
3DGameTest=257
Score=30
CPU=Intel 386SX
CPUspeed=33 MHz
BIOSinfo=        COPR. IBM 1981, 1987 (04/18/88, rev. 3)
MachineModel=04FC
BIOSdate=19880418
BIOSCRC16=495E
VideoSystem=VGA
VideoAdapter=VGA, unknown Chipset, 256kb Video Memory (BIOS)
Machine=PS/2 Model 50Z 286-10
 
IBMMuseum said:
A third 50Z (and good floppy drive), same BIOS revision, with IBM/Hyperace 486SLC2-50 CPU upgrade (switches are probably set to enable L1 and L2 cache), which crashes:

Code: 
[UID495E168A1A]
MemoryTest=212
OpcodeTest=190
VidramTest=1270
MemEATest=229
3DGameTe
Click to expand...

Retested with the '-s' switch:

Code: 
[UID495E981A]
MemoryTest=204
OpcodeTest=188
VidramTest=1278
MemEATest=227
3DGameTest=154
Score=37
...

But I have no idea at the moment if the L1 and/or L2 cache is enabled by the current switch positions. One other odd thing when I shifted between CPU family levels (in other words, from a 286-10, to any of the Kingston SX Now! upgrades, to the 486SLC2-50 CPU daughtercard, but not between the Kingston 386SX upgrades), it wants me to run the Reference Diskette and Auto Configure. Meaning, there may be some basic CPUID functionality checks (only what is in the 286 CPU socket is being changed) on what is a 286-based system!

The Model 70 and 80 (386DX-based), released at the same time as the Model 50 and 60 (286-based), did, and had two bytes in an Extended CMOS location that stored the CPU "Reset Signature". What might be there on the Model 50/50Z/60? I'll try to look when I get an opportunity.
 
Does anyone have any results from the IBM PC AT machines? I am curious to see how the 6MHz AT compares to the XT/286 (which we have results for). I am also interested in results from the IBM AT @ 8MHz, and the Tandy TX, TL/2/3 systems with the stock 286 @ 8MHz.
 
Great Hierophant said:
Get better soon, and if you could always install remote desktop software so that others could continue the good work!
Click to expand...

It's a 500MB VirtualBox .ova file, actually. If it's ever determined I will never be working on it again, I will be happy to share it (has full networking with packet driver and mTCP running).

I have every intention of producing the full tool, because I can see it in my head and really want to get it into bits. I'm just sidelined for a few weeks.
 
Great Hierophant said:
Does anyone have any results from the IBM PC AT machines? I am curious to see how the 6MHz AT compares to the XT/286 (which we have results for).
Click to expand...

This. This is one of the five questions I was hoping to answer when I designed the benchmark. A few others have already been answered (true real-world speed differences between 808x and NEC Vx0, 386DX40 vs. 486sx25) but this one is still unanswered.
 
Hi! Here are the specifications for my computer.

It’s a Dell 325P. It was originally a 386 DX-25; however, I replaced the 386 DX-25 chip and installed a Cyrix 486 DLC-40. It also has a 387 DX-33 math co-processor, but the BUS speed of the motherboard is still only 25mhz.

It’s got a Headland “Video 7” 1mb graphics card, a Sound Blaster 16 ASP, and a 525mb DOM (Disk on Module, IDE).


Thanks,

Todd


[UIDC4E211C146]
MemoryTest=264
OpcodeTest=117
VidramTest=326
MemEATest=101
3DGameTest=76
Score=57
CPU=Cyrix/TI 486-class processor
CPUspeed=33 MHz
BIOSinfo=Copyright (C) 1985-1988 Phoenix Technologies Ltd. (05/02/91, rev. 0)
MachineModel=01FC
BIOSdate=19910502
BIOSCRC16=C4E2
VideoSystem=VGA
VideoAdapter=VGA, Video 7, VGA Version 5, HT208 rev CDVRAM, 1024kb Video Memo
Machine=Dell unknown; ID is15 /AT clone
 
Oops! These are the results from my Tandy 1000SL, with an 8086 in "fast" (8.00 MHz) and "slow" (4.00 MHz) modes, respectively. I guess I'll need to use the "debug" version next...

Code: 
[UID9CC31D3C]
MemoryTest=-1
OpcodeTest=806
VidramTest=-1
MemEATest=-1
3DGameTest=-1
Score=9
CPU=Intel 8086
CPUspeed=8 MHz
BIOSinfo=unknown
MachineModel=00FF
BIOSdate=19880824
BIOSCRC16=9CC3
VideoSystem=CGA
VideoAdapter=Tandy 1000 SL/TL
Machine=Tandy 1000

Code: 
[UID9CC314B7]
MemoryTest=-1
OpcodeTest=1624
VidramTest=-1
MemEATest=-1
3DGameTest=-1
Score=5
CPU=Intel 8086
CPUspeed=4 MHz
BIOSinfo=unknown
MachineModel=00FF
BIOSdate=19880824
BIOSCRC16=9CC3
VideoSystem=CGA
VideoAdapter=Tandy 1000 SL/TL
Machine=Tandy 1000
 
Cimonvg said:
hello
looked though the PRELIM.ZIP , but did not see 8088 10 Mhz. So one day soon ,I will test clone XT :)
/cimonvg
Click to expand...

Hello again
V20 xt motherboard - the JUKO type with 1Mb RAM.:
;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.
[UIDC2376967]
MemoryTest=2705
OpcodeTest=1607
VidramTest=1795
MemEATest=1968
3DGameTest=1490
Score=5
CPU=NEC V20
CPUspeed=4.77 MHz
BIOSinfo=Copyright 1988 Juko Electronics Industrial Co.,Ltd. (05/17/88, rev. 0)
MachineModel=0000
BIOSdate=19880517
BIOSCRC16=C237
VideoSystem=VGA
VideoAdapter=VGA, ATI 28800-5VGAWonderBios : 0.1, 256kb Video Memory (BIOS)
Machine=PC/XT
-----------------and in turbo mode-------------
[UIDC2379450]
MemoryTest=1050
OpcodeTest=626
VidramTest=899
MemEATest=766
3DGameTest=577
Score=13
CPU=NEC V20
CPUspeed=11 MHz
BIOSinfo=Copyright 1988 Juko Electronics Industrial Co.,Ltd. (05/17/88, rev. 0)
MachineModel=0000
BIOSdate=19880517
BIOSCRC16=C237
VideoSystem=VGA
VideoAdapter=VGA, ATI 28800-5VGAWonderBios : 0.1, 256kb Video Memory (BIOS)
Machine=PC/XT

and the xt-motherboard that gave me trouble proier to version 0.97b:
[UID84F5F0]
MemoryTest=3801
OpcodeTest=1764
VidramTest=2339
MemEATest=1949
3DGameTest=1873
Score=4
CPU=Intel 8088
CPUspeed=4.77 MHz
BIOSinfo=(C)Award 1986 IBM COMPATIBLE.* (01/30/87, rev. 0)
MachineModel=0000
BIOSdate=19870130
BIOSCRC16=084F
VideoSystem=VGA
VideoAdapter=VGA, ATI 28800-5VGAWonderBios : 0.1, 256kb Video Memory (BIOS)
Machine=PC
 
Here are the results of my IBM 5140 PC Convertible:

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.

[UID309D500A]
MemoryTest=3572
OpcodeTest=1646
VidramTest=2009
MemEATest=1804
3DGameTest=1764
Score=5
CPU=Intel 8088
CPUspeed=4.77 MHz
BIOSinfo=7396918 COPR. IBM 1985 (09/13/85, rev. 0)
MachineModel=00F9
BIOSdate=19850913
BIOSCRC16=309D
VideoSystem=CGA
VideoAdapter=CGA
Machine=PC Convertible

Comparing to previous 5140 results, mine seems to be faster:

IBMMuseum said:
Code: 
[UID309D530]
MemoryTest=3573
OpcodeTest=1795
VidramTest=2010
MemEATest=1804
3DGameTest=1763
Score=4
CPU=Intel 8088 [no indication with current code that it is an 80C88]
CPUspeed=4.77 MHz
BIOSinfo=7396918 COPR. IBM 1985, 1987 (09/13/85, rev. 0)
MachineModel=00F9
BIOSdate=19850913
BIOSCRC16=3090
VideoSystem=CGA
VideoAdapter=CGA
Machine=PC Convertible
Click to expand...

I replaced the original i80C88 (it was dead) with an Intersil CS80C88. How can this processor be faster?

cs80c88.jpg
 
nestor said:
Here are the results of my IBM 5140 PC Convertible:..

....Comparing to previous 5140 results, mine seems to be faster:

I replaced the original i80C88 (it was dead) with an Intersil CS80C88. How can this processor be faster?...
Click to expand...

Be aware that the PC Convertible is also designed around a default of CGA, and may have anti-snow timings...
 
IBMMuseum said:
Be aware that the PC Convertible is also designed around a default of CGA, and may have anti-snow timings...
Click to expand...

The benchmark shows that the original CGA's RAM speed was dog-slow. Tandy 1000 machines have noticeably faster RAM access at the same speed. I suppose this is why that the palette change in Jungle Hunt is a few scanlines "off" on a Tandy machine compared to an IBM machine.
 
This is my 5150 w/ Nec V20

;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.

[UID3B85D3F]
MemoryTest=2481
OpcodeTest=1510
VidramTest=1537
MemEATest=1858
3DGameTest=1396
Score=6
CPU=NEC V20
CPUspeed=4.77 MHz
BIOSinfo=1501476 COPR. IBM 1982 (10/27/82, rev. 0)
MachineModel=0000
BIOSdate=19821027
BIOSCRC16=3B85
VideoSystem=VGA
VideoAdapter=VGA, Paradise PVGA1A, 8-Bit-Video, 8-Bit-ROMfixed-sync, 256kb Vi
Machine=PC

This is my 5150 with the Orchid Tiny Turbo 286

;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.

[UID3B851352]
MemoryTest=1318
OpcodeTest=1258
VidramTest=1325
MemEATest=1661
3DGameTest=1132
Score=11
CPU=Intel 80286
CPUspeed=7.16 MHz
BIOSinfo=1501476 COPR. IBM 1982 (10/27/82, rev. 0)
MachineModel=0000
BIOSdate=19821027
BIOSCRC16=3B85
VideoSystem=VGA
VideoAdapter=VGA, Paradise PVGA1A, 8-Bit-Video, 8-Bit-ROMfixed-sync, 256kb Vi
Machine=PC
 
Last edited:
JDT said:
This is my 5150 with the Orchid Tiny Turbo 286

;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.

[UID3B851352]
MemoryTest=1318
OpcodeTest=1258
VidramTest=1325
MemEATest=1661
3DGameTest=1132
Score=11
CPU=Intel 80286
CPUspeed=7.16 MHz
BIOSinfo=1501476 COPR. IBM 1982 (10/27/82, rev. 0)
MachineModel=0000
BIOSdate=19821027
BIOSCRC16=3B85
VideoSystem=VGA
VideoAdapter=VGA, Paradise PVGA1A, 8-Bit-Video, 8-Bit-ROMfixed-sync, 256kb Vi
Machine=PC
Click to expand...

Almost identical with my Tandy 1000SX w/ 286 Express Card (80286 @ 7.16MHz) results.
 
These results are from running the stub on my Wang Microsystems PC 250/16

After doing a clean boot (bypassing autoexec.bat and config.sys by pressing F5):
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.

[UID482A10E648]
MemoryTest=389
OpcodeTest=180
VidramTest=336
MemEATest=206
3DGameTest=165
Score=38
CPU=Intel 80286
CPUspeed=16 MHz
BIOSinfo=Copyright (C) 1985-1989 Phoenix Technologies Ltd. (06/05/89, rev. 0)
MachineModel=01FC
BIOSdate=19890605
BIOSCRC16=482A
VideoSystem=VGA
VideoAdapter=VGA, Paradise PVGA1A, 16-Bit-Video, 8-Bit-ROMfixed-sync, VESA, 5
Machine=AT clone
Description=Wang Microsystems PC 250/16

After a normal boot:
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.

[UID482A11A890]
MemoryTest=386
OpcodeTest=161
VidramTest=331
MemEATest=207
3DGameTest=161
Score=39
CPU=Intel 80286
CPUspeed=16 MHz
BIOSinfo=Copyright (C) 1985-1989 Phoenix Technologies Ltd. (06/05/89, rev. 0)
MachineModel=01FC
BIOSdate=19890605
BIOSCRC16=482A
VideoSystem=VGA
VideoAdapter=VGA, Paradise PVGA1A, 16-Bit-Video, 8-Bit-ROMfixed-sync, VESA, 5
Machine=AT clone
Description=Wang Microsystems PC 250/16

The program was running off the harddrive both times. Note how the clean boot is actually slower?!? :confused:

Also, the CPU is made by AMD, not Intel.
 
Krille said:
...Also, the CPU is made by AMD, not Intel.
Click to expand...

In your example it can be determined that it is not an Intel CPU (being a 80286 at 16MHz), but the code cannot be expected to be that intuitive. It is however from the same licensed codemask, and will be indistinguishable in software otherwise. As I commented on my Intel 486 CPUID page, even if it was another manufacturer designing the codemask on their own, they are going to make the part that is designed to be compatible report the same identification (the same with FDC chips).
 
Samsung SPC3000V non turbo mode

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.

[UIDDCEA123]
MemoryTest=3652
OpcodeTest=1765
VidramTest=3471
MemEATest=1954
3DGameTest=1878
Score=4
CPU=Intel 8088
CPUspeed=4.77 MHz
BIOSinfo=Copyright (c) 1984,1985,1987 Phoenix Technologies Ltd (04/09/88, rev. 220)
MachineModel=12F0
BIOSdate=19880409
BIOSCRC16=0DCE
VideoSystem=CGA
VideoAdapter=CGA
Machine=PC/XT

turbo mode
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.

[UIDDCEA840]
MemoryTest=1850
OpcodeTest=874
VidramTest=2725
MemEATest=963
3DGameTest=933
Score=7
CPU=Intel 8088
CPUspeed=10 MHz
BIOSinfo=Copyright (c) 1984,1985,1987 Phoenix Technologies Ltd (04/09/88, rev. 220)
MachineModel=12F0
BIOSdate=19880409
BIOSCRC16=0DCE
VideoSystem=CGA
VideoAdapter=CGA
Machine=PC/XT

fast cpu in turbo mode for an xt, it's btw a fujitsu type 8088-1 cpu, shame it couldn't tell me the name 'n stuff of the graphics card.
 
Thanks for the result. For CGA cards, there is very little that can be probed to determine what make/model card it is. There are techniques you can use to probe the machine, such as trying to switch into an extended mode and seeing if it succeeds, but doing so corrupts the display and one of my requirements for the benchmark is that it not corrupt the display.
 
You must log in or register to reply here.
Back
Top