Guess its time to throw diagnostic programs at it (maybe start with memtest)

Already ran memtest and it passed.

What other diagnostic programs can I try?

PC Player Benchmark

Speedsys

Thanks a lot 😀

Running tests and will report back here...

Are the results from Speed Test:

CPU is in V86 mode: No
Processor : AMD Am486DX4WB(tm) 100 MHz
CPUID (TFMS) : 0494 Codename: P24C (0.50um)
Feature : 00000001h MMX(tm):No, IA SSE:No, IA SSE2:No
CPU speed index : 37.48

Total memory size : 64 MB
Memory Bandwidth : 76.44 MB/s

PCI Video : S3 325 ViRGE GUI Accelerator

VESA OEM String : S3 Incorporated. 86C325
VESA video memory : 4096 KB (18534 KB/s)

Hard drive 0 : 1023C 255H 63S 7.84 GB
ÀÄÄÄÄÄ Model (PM): ST380215A
Tested in FAST mode
Average/Max seek time : 5.35 ms / 6.66 ms
Random seek time : 5.13 ms
Track-to-track seek : 1.20 ms
Random access time : 9.50 ms
Linear verify speed : 73567 KB/s
Min/Max verify speed : 71848 KB/s / 73964 KB/s
Linear read speed : 11682 KB/s
Min/Max read speed : 11677 KB/s / 11685 KB/s
Buffered read speed : 12030 KB/s
Hard Drive speed index : 2684.80

Mainboard chipset : SiS 496/497
BIOS vendor : Award Modular BIOS v4.50PG (04/05/96)
Year2000 Bug : 31-12-1999 01-01-1900
OS version : MS-DOS 7.10
ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ

ATA/ATAPI Device Information

IDE0ÄÂÄMasterÄ<ATA-7>Ä ST380215A
³ Cylinders: 9729, Heads: 255, Sectors: 63 (74.53 GB)
³ Serial Number : 9QZDBRCY
³ Firmware Revision : 3.AAD
³ Maximum Transfer Mode : PIO 4, DMA 2, UDMA 5 (ATA-100)
³ Selected DMA Transfer Mode : DMA 2
³ Cache Buffer Size : 2048 KB
³
ÀÄSlaveÄÄ<ATAPI>Ä Pioneer DVD-ROM ATAPIModel DVD-119 0105
Firmware Revision : E1.05
Maximum Transfer Mode : PIO 4, DMA 2, UDMA 4 (UltraDMA/66)
Selected DMA Transfer Mode : DMA 2
Maximum Transfer Speed : 7060 KB/s (40X)
Selected Transfer Speed : 1633 KB/s (9X)
Cache Buffer Size : 256 KB

IDE1ÄÂÄMasterÄ<Empty>
³
ÀÄSlaveÄÄ<Empty>

IDE2ÄÂÄMasterÄ<Empty>
³
ÀÄSlaveÄÄ<Empty>

IDE3ÄÂÄMasterÄ<Empty>
³
ÀÄSlaveÄÄ<Empty>
ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ

Cache/Memory Benchmark
ÚÄÄÄÄÄÄÄÄÄÄÄÄÄÄÂÄÄÄÄÄÄÄÄÄÄÄÄÄÄÂÄÄÄÄÄÄÄÄÄÄÄÄÄÄÂÄÄÄÄÄÄÄÄÄÄÄÄÄÄ¿
³ Read ³ Write ³ Move ³ Average ³
ÚÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÅÄÄÄÄÄÄÄÄÄÄÄÄÄÄÅÄÄÄÄÄÄÄÄÄÄÄÄÄÄÅÄÄÄÄÄÄÄÄÄÄÄÄÄÄÅÄÄÄÄÄÄÄÄÄÄÄÄÄÄ´
³ Cache Level 1 ³ 95.10 MB/s³ 41.61 MB/s³ 124.53 MB/s³ 87.08 MB/s³
³ Cache Level 2 ³ 51.10 MB/s³ 41.50 MB/s³ 34.05 MB/s³ 42.22 MB/s³
³ Memory ³ 27.93 MB/s³ 42.45 MB/s³ 13.08 MB/s³ 27.82 MB/s³
ÀÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÁÄÄÄÄÄÄÄÄÄÄÄÄÄÄÁÄÄÄÄÄÄÄÄÄÄÄÄÄÄÁÄÄÄÄÄÄÄÄÄÄÄÄÄÄÁÄÄÄÄÄÄÄÄÄÄÄÄÄÄÙ
ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ

Memory Modules Information

SMBus/i2c Host Controller : Not Present
ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ

PCI/AGP Device Information [PCI 2.10]

[00-05-0]
ÀÄÄÄÄÄ VEN_1039, DEV_0496, REV_31, SUBSYS_00000000
SiS 85C496/7 Chipset, CPU to PCI to ISA Bridge
Class Code : 060000 - PCI to Host Bridge
[00-12-0]
ÀÄÄÄÄÄ VEN_5333, DEV_5631, REV_06, SUBSYS_00000000
S3 325 ViRGE GUI Accelerator
Class Code : 030000 - VGA Compatible Adapter
Memory Base Range 0 : F4000000 - F7FFFFFF (64MB)
ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ

End of report.

I ran PC Player Bench Mark, but as it was in a different language wasn't really sure what to make of it.

The Speedsys output looks OK to me.

I suggested the PC Player benchmark (only available in German, I'm afraid), because it is likely to crash when there is a problem with the cache or main memory.

What happens if you run Quake without Soundblaster digital sound? Does it still crash?

When I ran PC Player Benchmark it did not crash (actually ran it a few times). Quake does not have an option to chose sound cards etc, but I did REM out the lines from the autoexec.bat for the sound card and then ran Quake and it ran without crashing.

Just tried changing the option in CMOS that says "IRQ 5 Used by ISA" from "No/ICU" to "yes" (because I figured that is the IRQ the SB16 is supposed to use) and when the computer booted up and tried to load DOS Start, DOS Start crashes with the message "Device I/O error in line 0 of module DS18MAIN at address 0745:132F". I went back in to CMOS and changed the option back but DOS Start still says the same thing. Is DOS Start now corrupted somehow?

Hmmm. Everything points to a problem with the SB16 card.

I just looked at a few pics of the CT2230 on the net, it seems to be one of those "half-PnP" cards (which is a good thing, because "real" ISA/PnP is a pain on this mainboard).

This card has to be initialized before it works, either in your AUTOEXEC.BAT, or you can do it manually:

1. Set your BLASTER environment variable: SET BLASTER=A220 I5 D1 H5 P330 T6
2. Execute DIAGNOSE.EXE with the /S switch: DIAGNOSE.EXE /S

Then try Doom with SB sound again.

5u3 wrote:

Hmmm. Everything points to a problem with the SB16 card. […]
Show full quote

Hmmm. Everything points to a problem with the SB16 card.

I just looked at a few pics of the CT2230 on the net, it seems to be one of those "half-PnP" cards (which is a good thing, because "real" ISA/PnP is a pain on this mainboard).

This card has to be initialized before it works, either in your AUTOEXEC.BAT, or you can do it manually:

1. Set your BLASTER environment variable: SET BLASTER=A220 I5 D1 H5 P330 T6
2. Execute DIAGNOSE.EXE with the /S switch: DIAGNOSE.EXE /S

Then try Doom with SB sound again.

Already had that. These are my startup files:

Autoexec.bat:

@ECHO OFF
SET PATH=C:\DOS71;..;C:\PKZIP;C:\RAR;
SET SOUND=C:\SB16
SET BLASTER=A220 I5 D1 H5 P330 T6
SET MIDI=SYNTH:1 MAP:E
LH MSCDEX /D:IDE-CD
LH SMARTDRV
LH CTMOUSE
LH DOSKEY
C:\SB16\DIAGNOSE /S
C:\SB16\MIXERSET /P /Q
M.BAT

Config.sys:

DEVICE=C:\DOS71\HIMEM.SYS
DEVICE=C:\DOS71\EMM386.EXE RAM
DOS=HIGH,UMB
DEVICEHIGH=C:\DOS71\VIDE-CDD.SYS /D:IDE-CD

Also, because I did this:

"Just tried changing the option in CMOS that says "IRQ 5 Used by ISA" from "No/ICU" to "yes" (because I figured that is the IRQ the SB16 is supposed to use) and when the computer booted up and tried to load DOS Start, DOS Start crashes with the message "Device I/O error in line 0 of module DS18MAIN at address 0745:132F". I went back in to CMOS and changed the option back but DOS Start still says the same thing. Is DOS Start now corrupted somehow?"

It seems that my file allocation table is bad now. Am running CHKDSK on it right now.

Maybe you have one of the revisions with the buggy IDE controller. Try to limit the IDE transfer modes to PIO2 in the BIOS setup ("IDE 0 Master Mode", etc... in the Chipset Features section).

It's alright to set the IRQ and DMA of your SB16 to "yes" in the PCI/PnP setup.

5u3 wrote:

Maybe you have one of the revisions with the buggy IDE controller. Try to limit the IDE transfer modes to PIO2 in the BIOS setup ("IDE 0 Master Mode", etc... in the Chipset Features section).

It's alright to set the IRQ and DMA of your SB16 to "yes" in the PCI/PnP setup.

Well, it's weird because I did not have any HD errors until after I did that.

I just set IDE0 Master Mode to 2 (options were Auto, 1, 2, 3 or 4). Is that right?

Just set the "ISA Used By" settings for IRQ and DMA for the SB to Yes on IRQ 5 and DMA 1 and DMA 5. Sound right?

I'll see what this does...

You sir, are a genius!

Everything now working!

What can I use to defrag this drive? Defrag.exe doesn't work with MSDOS 7.10 as far as I can tell. And I didn't really want to install Windows, although I found some original Windows 3.11 disks the other day.

Glad to hear it works now. Early versions of this board have a nasty hardware bug in the integrated CMD640 IDE controller, which results in data corruption when using the faster IDE transfer modes (PIO3 and PIO4). Everything up to PIO2 should be safe.

About the defrag question: FreeDOS has an open-source defragmentation program that supports FAT32 filesystems.
Download it here.

5u3 wrote:

About the defrag question: FreeDOS has an open-source defragmentation program that supports FAT32 filesystems.
Download it here.

When I tried to run that defrag it said it "simply refuses to run in Windows". How do I get round that?

rick12373 wrote:

You sir, are a genius!

Everything now working!

Awesome advice, 5u3! 😁

rick12373 wrote:

5u3 wrote:

About the defrag question: FreeDOS has an open-source defragmentation program that supports FAT32 filesystems.
Download it here.

When I tried to run that defrag it said it "simply refuses to run in Windows". How do I get round that?

Does anybody know how to get the FreeDOS defrag to work under MSDOS 7.10?

Why are you so hell bent on defragging? 😁 It doesn't help much at all in practice, especially if you have a semi-modern HDD waiting on that 486. Any drive from like 1998 onward is faster than that CPU can feed.

If you want mad HDD speed and have PCI, put in a UDMA card. I run a Promise Ultra66 in my 486, with a nice'n'quiet 80GB Maxtor attached. The system can move about 30MB/s according to measurements. The DMA HDD access probably helps smooth out Windows a bit too. The onboard IDE on those mobos is almost certainly non-busmastering and uses PIO Mode 3 or perhaps 4.

swaaye wrote:

Why are you so hell bent on defragging? 😁 It doesn't help much at all in practice, especially if you have a semi-modern HDD waiting on that 486. Any drive from like 1998 onward is faster than that CPU can feed.

If you want mad HDD speed and have PCI, put in a UDMA card. I run a Promise Ultra66 in my 486, with a nice'n'quiet 80GB Maxtor attached. The system can move about 30MB/s according to measurements. The DMA HDD access probably helps smooth out Windows a bit too. The onboard IDE on those mobos is almost certainly non-busmastering and uses PIO Mode 3 or perhaps 4.

OK, well if it doesn't matter on an old system like this then I won't worry about it.

Just got a new serial mouse from eBay. I hope it works. I had a bunch of old ones at home and a couple of them didn't work (they weren't damaged physically). The old serial mouse I ended up using was a Microsoft serial mouse. Not sure what make this eBay one is, but it is definitely new.

I don't think SIS 496's IDE controller is CMD640-based.