VOGONS


First post, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie

I have this Cirrus Logic GD5428, VLB, that worked perfectly fine.
Yesterday it gave this error on POST, system goes on working, no beeps no other errors (I can see the HD blink).
At each supposed black screen it goes away to come back immediately after (when signal should be present).

CS 5428 - Cirrus Logic // PM-V128 REV:B / FCC ID:KC8PM-V116 / P/N :55V128000B
https://fccid.io/KC8PM-V116
Philips signetics 27C256-20 FA 32KB (bios chip)

Bios is working, I swapped it with another ISA card (CL-GD5420-750C-C) and it gives the same error, but on the ISA it displays everything fine. On the VLB still "out of range".

I cleaned the card thoroughly with contact cleaner, tried my best with VGA connector but nothing changed.

The monitor is a DELL 16:9 that worked well with it. I'll try an old CRT to see if it is an issue with low resolution for some reason (dead component?).

Old stuff brings up a lot of difficulties 😀

Any input is appreciated.

--- BIOS image in attachement ---
May help someone else.

Attachments

Last edited by Nexxen on 2022-11-23, 03:09. Edited 2 times in total.

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 1 of 18, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie

With a cold boot a central stripe (1/6 of the screen), blue background with letters/numbers in a pattern.
Could be a ram issue?
I'll post a pic when it'll happen again, as on reboot it doesn't show.

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 2 of 18, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie

This is what it displays with a cold boot.

Attachments

  • strip.jpg
    Filename
    strip.jpg
    File size
    488.72 KiB
    Views
    631 views
    File license
    Fair use/fair dealing exception

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 3 of 18, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie

https://pdf1.alldatasheet.com/datasheet-pdf/v … I/TMS45160.html

This is the ram. 80ns.
Could it be a ram issue? Can i use another card's ram to try a swap?

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 4 of 18, by mkarcher

User metadata
Rank l33t
Rank
l33t
Nexxen wrote on 2022-11-21, 15:13:

This is the ram. 80ns.
Could it be a ram issue?

I would be a fool to say "surely no" without careful testing, but getting the correct frequency from a Cirrus Logic chip doesn't require working RAM at all. Thus I conclude there has to be something different wrong with your card if your monitor refuses to sync to it.

Nexxen wrote on 2022-11-21, 15:13:

Can i use another card's ram to try a swap?

You can mix and match those 256K x 16 FPM RAM chips nearly as you like. There are two pinout variants: "Dual CAS" and "Dual WE", using a chip using the other pinout doesn't cause permanent damage. The chip you linked is "Dual CAS", recognizable by the two "CAS" pins UCAS and LCAS. Around the time of the Trio64/V+, EDO RAM started to appear on video cards, which might also be incompatible with cards designed for FPM.

Reply 5 of 18, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie
mkarcher wrote on 2022-11-21, 15:36:
I would be a fool to say "surely no" without careful testing, but getting the correct frequency from a Cirrus Logic chip doesn't […]
Show full quote
Nexxen wrote on 2022-11-21, 15:13:

This is the ram. 80ns.
Could it be a ram issue?

I would be a fool to say "surely no" without careful testing, but getting the correct frequency from a Cirrus Logic chip doesn't require working RAM at all. Thus I conclude there has to be something different wrong with your card if your monitor refuses to sync to it.

Nexxen wrote on 2022-11-21, 15:13:

Can i use another card's ram to try a swap?

You can mix and match those 256K x 16 FPM RAM chips nearly as you like. There are two pinout variants: "Dual CAS" and "Dual WE", using a chip using the other pinout doesn't cause permanent damage. The chip you linked is "Dual CAS", recognizable by the two "CAS" pins UCAS and LCAS. Around the time of the Trio64/V+, EDO RAM started to appear on video cards, which might also be incompatible with cards designed for FPM.

Thanks for the info.
On the card, what actually generated the signal? The GPU or another chip?

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 6 of 18, by mkarcher

User metadata
Rank l33t
Rank
l33t
Nexxen wrote on 2022-11-21, 17:12:

On the card, what actually generated the signal? The GPU or another chip?

The GPU generates the pixel clock, the horizontal sync signal and the vertical sync signal. Every timing is derived from the 14.318 MHz crystal on the graphics board.

Reply 7 of 18, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie
mkarcher wrote on 2022-11-21, 21:05:
Nexxen wrote on 2022-11-21, 17:12:

On the card, what actually generated the signal? The GPU or another chip?

The GPU generates the pixel clock, the horizontal sync signal and the vertical sync signal. Every timing is derived from the 14.318 MHz crystal on the graphics board.

I'll replace the crystal to see if had anything to do.

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 8 of 18, by rasz_pl

User metadata
Rank l33t
Rank
l33t

Afaik the frequency is programmed by the bios, without video bios running no output will be generated. Garbage on screen and bad frequencies make me think something wrong with ISA slot, or power supply to the card?

>Bios is working, I swapped it with another ISA card (CL-GD5420-750C-C) and it gives the same error, but on the ISA it displays everything fine. On the VLB still "out of range".

so the ISA card plugged in to ISA slot with VLB glued to its ass is also doing weird things? sure sounds like that ISA slot went bad

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

Reply 9 of 18, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie
rasz_pl wrote on 2022-11-21, 22:50:

so the ISA card plugged in to ISA slot with VLB glued to its ass is also doing weird things? sure sounds like that ISA slot went bad

I don't understand this. No, I didn't "glue it to its ass". 😀

I took the bios chip from a GD5422, tried on VLB: same.
Tried VLB card Bios on ISA card: works as expected.
I tried the VLB card on all VLB slots, nothing changed.

I don't think it's the bios chip or data on it. Issue is somewhere else, but it's not easy like with all old things.

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 10 of 18, by rasz_pl

User metadata
Rank l33t
Rank
l33t

so again, does that:

>>Bios is working, I swapped it with another ISA card (CL-GD5420-750C-C) and it gives the same error, but on the ISA it displays everything fine. On the VLB still "out of range".

mean CL-GD5420-750C-C works fine in short only ISA slots, but fails when plugged into combined ISA/VLB slots?

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

Reply 11 of 18, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie
rasz_pl wrote on 2022-11-22, 00:06:

so again, does that:

>>Bios is working, I swapped it with another ISA card (CL-GD5420-750C-C) and it gives the same error, but on the ISA it displays everything fine. On the VLB still "out of range".

mean CL-GD5420-750C-C works fine in short only ISA slots, but fails when plugged into combined ISA/VLB slots?

I was looking for a BIOS image to test if it was a BIOS issue.
I came across a site with Bioses and it stated that one of the images was a universal 542x model. Turned out it didn't work.

I have this ISA card with a 5422 chip, tried its bios chip to see if it worked on the VLB card.
Same problem.

To test the opposite, putting the VLB card's chip in the ISA card (the "swap"), the ISA card worked perfectly. No issues at all. I just ran Doom benchmark and it as fast as it should be.

My conclusion is that the card has a problem but not the bios, as it works.
The card worked until it didn't with that bios, can't rule it out 100% though, but I'm calling it close enough.

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 13 of 18, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie

Crystal was replaced but no change, still out of range.

This is what happens:
1) display bios info // out of range
2) blaks out in transition to mobo post info // goes black
3) bios post // out of range

Basically when it should output a signal the signal is not a valid input for the screen.

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 15 of 18, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie

All ISA slots works with 16bit cards.
VLB slots work with VLB controllers 100%

To me it's a matter of signal not being processed correctly by the GD5428, 74F32PC (close to crystal) or MC74F245N (block of 4 close to sockets).
Can't guess anything else.

https://pdf1.alldatasheet.com/datasheet-pdf/v … LD/74F32PC.html

https://pdf1.alldatasheet.com/datasheet-pdf/v … /MC74F245N.html

Attachments

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 16 of 18, by Tiido

User metadata
Rank l33t
Rank
l33t

You might need to try a different video BIOS, chances are another version has better video timings (they're all under software control).

T-04YBSC, a new YMF71x based sound card & Official VOGONS thread about it
Newly made 4MB 60ns 30pin SIMMs ~
mida sa loed ? nagunii aru ei saa 😜

Reply 17 of 18, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie

Ok, it wad broken traces.
I feel like an idiot. 😀
But it was a straight cut, razor like, probably clashed against another sharp pin in the box it was stored. Took me 10 times and light at angle to spot it.

It was in fact connecting BUS to MC74F245N

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 18 of 18, by Nexxen

User metadata
Rank Oldbie
Rank
Oldbie
Tiido wrote on 2022-11-23, 03:03:

You might need to try a different video BIOS, chances are another version has better video timings (they're all under software control).

I'll need a new rom, compatible, to make the upgrade.
Will do in the future.

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K