VOGONS


Reply 20 of 33, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

@feipoa
My message was cryptic i admit.
I never found the exact chipset, so sent you highres picture of what i have, if it is of any help to you.

@razs_pl
Thanks for the link. It norrows down the options to 2 or 3 possible candidates.

retro bits and bytes

Reply 22 of 33, by feipoa

User metadata
Rank l33t++
Rank
l33t++

pshipkov is saying that the BIOS for a board which contains an OPTI 495* chipset will work just fine on a board which contains an M1429 chipset. Is this true? Why would this work fine?

Plan your life wisely, you'll be dead before you know it.

Reply 23 of 33, by Deunan

User metadata
Rank Oldbie
Rank
Oldbie

I don't think so, most of these BIOSes were compiled with support for one particular chipset, and the menu options reflected that. It's not impossible to make a version that would detect the chipset it's running on, but why bother? It would only bloat the code and allow mobo manufacturers to just copy one BIOS over every board rather than pay separate license fee.

If this does work then perhaps the chipset had been re-labeled for some reason? To hide who made the mobo or with what parts to prevent copying?

Reply 24 of 33, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

I can confirm that this is the case for forex, umc, unichip and opti based motherboards. At least the ones i tried on. Extrapolating from this empirical data, one can conclude that there is a good chance it will work for other chipsets as well.
I guess the base of the 386 standard is consistent enough to allow for this to happen.

Now thinking about it, i am pretty sure i had success with swapping bioses of 286 systems as well, but cannot remember any details worth bringing here.

retro bits and bytes

Reply 25 of 33, by noshutdown

User metadata
Rank Oldbie
Rank
Oldbie

what is the topic now? i wanna know why this board performs so fast in cache/ram tests(cachechk, speedsys) yet so slow in game tests, is it just that slow or i got some setting wrong?

Reply 26 of 33, by feipoa

User metadata
Rank l33t++
Rank
l33t++
noshutdown wrote:

what is the topic now? i wanna know why this board performs so fast in cache/ram tests(cachechk, speedsys) yet so slow in game tests, is it just that slow or i got some setting wrong?

Seems pretty obvious (to use your words) that the comments are suggesting trying an alternate BIOS to see if therein lies your problem.

Plan your life wisely, you'll be dead before you know it.

Reply 27 of 33, by bakemono

User metadata
Rank Oldbie
Rank
Oldbie

i wanna know why this board performs so fast in cache/ram tests(cachechk, speedsys) yet so slow in game tests

Those tests cover only read performance? What about writes?

Maybe you could also try the old SPDUP.COM

mov al,0x74
out 0x43,al
mov ax,0x00A0
out 0x41,al
xchg al,ah
out 0x41,al

mov ax,0x4c00
int 0x21

Slowing down this timer can free up CPU time, sometimes even on systems that one wouldn't expect to use the timer for DRAM refresh.

Reply 28 of 33, by rasz_pl

User metadata
Rank l33t
Rank
l33t
noshutdown wrote:

what is the topic now? i wanna know why this board performs so fast in cache/ram tests(cachechk, speedsys) yet so slow in game tests, is it just that slow or i got some setting wrong?

ISA clock, tweakable boards allow 10, or even 12MHz

Open Source AT&T Globalyst/NCR/FIC 486-GAC-2 proprietary Cache Module reproduction

Reply 29 of 33, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

If you have perf_at.exe - wonder what it will report for the video write wait states.
These old benchmarks are usually all over the place with the results they report, but there is a pattern there - if it shows big numbers, this means that there is something clogging the pipe between chipset and vga

retro bits and bytes

Reply 30 of 33, by Am386DX-40

User metadata
Rank Member
Rank
Member
pshipkov wrote:

I usually use BIOSes from two motherboards:
PC Chips M321 rev 2.3 (later revisions are starting to suck)

Why do later revisions suck? Do you know what exactly is the difference among different revisions? I have a 2.6 and a 2.7, the only difference I see are the jumpers for the cache.

Reply 31 of 33, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

Rev 2.5 and later - the SRAM chips are soldered directly on the board.
Rev 2.6 and later - the crystal oscilator is replaced by clock generator.
Rev 3.0 - the CPU is soldered on the board.
And so on.
Basically, the later revisions of the motherboard don't bring anything better, but at the same time are less customizable.

retro bits and bytes

Reply 33 of 33, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

If the motherboard provides robust options to adjust the frequency - yes.
Otherwise no.

I cannot remember a 386 board that provides jumpers for this kind of stuff.
This mobo is no different. So, crystal oscillator is kind of better - it can be swapped easily, especially if socketed.

retro bits and bytes