486 MB / DX2-66 issue <SOLVED> \ VOGONS

486 MB / DX2-66 issue <SOLVED>

Topic actions

Post a reply

First post, by flamingpentium

Posted on 2022-08-22, 03:10

flamingpentium Offline

Rank Newbie

Rank: Newbie
Posts: 6
Joined: 2022-08-22, 02:30

Hi all,

I'm in the middle of a 486 build and I've hit a problem I'm hoping someone might be able to help with. While the system seems to be working fine, all CPU speed tests seem to be well below expected - at least going off the scales provided:

Landmark: 52.28Mhz FPU: 209Mhz
Norton sysinfo: 16.3 which seems to be about half the speed of a 386DX-33
Topbench: 65
speedsys: 9.35 - which seems to be about 386DX-40 on the scale.

All above programs as well as CHKCPU and MSD all correctly identify the CPU as a DX2-66

Specs:

MB: 486 OPTI PAT48PG - 0.30 (manual can be found here https://manuals.plus/m/0b5246beaedf38f5916819 … a2e6bbd452c36d0
CPU is a 486DX2-66 intel overdrive version
Bus speed: 33Mhz checked on jumper and through software
Memory: 16MB (2x8MB SIMMS)
I/O: Winbond VLB
Vid: CL-GD5429 VLB
HDD: CF Card
PSU: AT (CWT-200) I assume its 200W
Sound: none installed yet.
OS: msdos 6.22 - only himem.sys and ansi.sys running so CPU is in real mode.
Internal CMOS battery is removed and disabled and using external battery header (3xAA)

My first thought was possibly a faulty cache issue. BIOS shows both internal and external cache as enabled and the MB shows 128K cache on boot but CACHECHK does throw up a "This machine does not seem to have any cache" message but then shows a "Cache ENABLED" message against the CPU - which it correctly detects. Turning off cache in bios does not seem to do anything to the benchmark numbers other than for BIOS to say there is no cache in the pre-OS boot message.

Not sure where to go from here. Any help greatly appreciated.

Last edited by flamingpentium on 2022-08-22, 08:11. Edited 1 time in total.

Reply 1 of 7, by Sphere478

Posted on 2022-08-22, 03:21

Sphere478 Offline

Rank l33t++

Rank: l33t++
Posts: 5643
Joined: 2021-01-13, 04:45

If the bios says it’s there then it probably is, unless it’s some sort of fake cache scam which was common back then

But then again, if disabling cache makes no difference then that is a red flag.

Try removing the cache physically

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 2 of 7, by TheMobRules

Posted on 2022-08-22, 05:11

TheMobRules Offline

Rank Oldbie

Rank: Oldbie
Posts: 1445
Joined: 2016-01-19, 23:29

Are you using a turbo switch? If not, try shorting pins 7 & 17 with a jumper as instructed in the manual to get the system to run at full speed. Disabling one or both caches is a very common way of handling the turbo switch slowdown on 486 motherboards.

Reply 3 of 7, by Sphere478

Posted on 2022-08-22, 05:51

Sphere478 Offline

Rank l33t++

Rank: l33t++
Posts: 5643
Joined: 2021-01-13, 04:45

TheMobRules wrote on 2022-08-22, 05:11:

Are you using a turbo switch? If not, try shorting pins 7 & 17 with a jumper as instructed in the manual to get the system to run at full speed. Disabling one or both caches is a very common way of handling the turbo switch slowdown on 486 motherboards.

Good idea.

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 4 of 7, by flamingpentium

Posted on 2022-08-22, 07:18

flamingpentium Offline

Rank Newbie

Rank: Newbie
Posts: 6
Joined: 2022-08-22, 02:30

@TheMobRules - I doff my hat to you and I am most chagrined for not picking this up myself. Thank you it worked. The turbo connector was loose and the front LED was saying turbo was on, so it never occurred to me to check.
@Sphere478 - thanks for the suggestion in any case it has prompted me to populate the empty cache sockets.

While I have either of you and at the risk of asking one more question about this MB. The manual makes mention of x36 72Pin SIMM (and x9 30 Pin) modules - I thought with these numbers it would be memory modules with parity. Would this be correct? I don't recall 486s being that particular about parity memory - let me know if I need to start another topic.

Reply 5 of 7, by TheMobRules

Posted on 2022-08-22, 07:31

TheMobRules Offline

Rank Oldbie

Rank: Oldbie
Posts: 1445
Joined: 2016-01-19, 23:29

Yes, those would be parity modules. But I doubt they're required, usually it means that they're supported and you can enable/disable parity checks by setting an option in the BIOS config.

In any case, I don't think it's really difficult to find parity FPM modules if necessary.

Reply 6 of 7, by flamingpentium

Posted on 2022-08-22, 08:11

flamingpentium Offline

Rank Newbie

Rank: Newbie
Posts: 6
Joined: 2022-08-22, 02:30

Makes sense. I've not noticed anything in BIOS like that but something to note if I go beyond 16MB. Many thanks for all your help I will mark this thread as solved.

Reply 7 of 7, by Sphere478

Posted on 2022-08-22, 18:01

Sphere478 Offline

Rank l33t++

Rank: l33t++
Posts: 5643
Joined: 2021-01-13, 04:45

Be careful when upgrading memory, sometimes you need to use double sided.

On my 486 gateway 2000 to get to 32 mb I had to use two double sided 16mb sticks. Which maxxed it out.

I think a single 32 may have worked?

But single sided 16s didn’t

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Go to top of page Go to top of page

Back to General Old Hardware

Main menu

Common searches