Disruptor wrote on 2024-02-03, 20:23:

CoffeeOne wrote on 2024-02-03, 18:19:

Most likely you have got 20ns chips. 15ns chips usually can do 2-1-1-1 @40MHz

... when you drive 2 banks.

Yes, I agree. Double banked cache relaxes the timing.
I was able to get 512kB single bank cache stable at 40MHz on an Asus PVI-486SP3 with 15ns chips, but it was a really super hard to find 5 SRAM chips, that did it.
As a tag a Winbond chip with 10ns was used, but I am not sure if it was real 😁. But it worked in the end.

feipoa wrote on 2024-02-01, 12:54:

I may remember incorrectly, but I sorta recall motherboard versions which were sold with a UV EPROM chip contained a BIOS revision which did not try to update ESCD information. Did you see Updating ESCD... with your BIOS before you flashed it? Or did this only happen after you flashed it?

I'm pretty sure you need an EEPROM to update ESCD information. I don't think the UV EPROMs normally contain a EEPROM section to maintain ESCD.

For what it's worth, the original version of this board I used in the 90s had working Updating ESCD...Success only when a hardware change was made. As that was sold within an OEM system with a "Designed for Windows 95" sticker, I suspect the Windows Logo requirements might have included a working ESCD while those sold by the piece or to system builders may have used the cheaper part. The Updating ESCD also happened when encountering the "disappearing PS/2 mouse" problem mentioned in your manual, so the message was always bad news.

A few years back, I got one of these boards from eBay that came with a UV EPROM. It shipped with a BIOS named "UUD960326S." (notice the dot versus the normal version without). It printed "Updating ESCD..." at every boot but did not hang.
When I tried flashing a "modern" Flash chip (39SF010) it switched to hanging at Updating ESCD. I believe it has something to do with the original EEPROM being byte-programmable while the Flash chip is page oriented.
When I switched to a "period correct" EEPROM (29PHEE010) it went to "Updating ESCD...Success" only on hardware changes.

Later versions of Award BIOS, but not this old 4.50PG one, seem to have a list of allowed EEPROM/Flash IDs embedded in them, possibly to avoid the hang when it tries to program an incompatible chip.

2-1-1-1 @ 40 MHz single banked in an ASUS PVI-486SP3

CoffeeOne wrote on 2024-02-03, 22:55:

Yes, I agree. Double banked cache relaxes the timing.
I was able to get 512kB single bank cache stable at 40MHz on an Asus PVI-486SP3 with 15ns chips, but it was a really super hard to find 5 SRAM chips, that did it.
As a tag a Winbond chip with 10ns was used, but I am not sure if it was real 😁. But it worked in the end.

Congratulations!
Do you still have that board?
And may you post a photograph of that SRAMs? 😀
Do you use an Am5x86 @ 160 or Cx5x86 @ 120?

Disruptor wrote on 2024-02-07, 13:24:

Congratulations! Do you still have that board? And may you post a photograph of that SRAMs? :) Do you use an Am5x86 @ 160 or Cx5 […]
Show full quote

2-1-1-1 @ 40 MHz single banked in an ASUS PVI-486SP3

CoffeeOne wrote on 2024-02-03, 22:55:

Yes, I agree. Double banked cache relaxes the timing.
I was able to get 512kB single bank cache stable at 40MHz on an Asus PVI-486SP3 with 15ns chips, but it was a really super hard to find 5 SRAM chips, that did it.
As a tag a Winbond chip with 10ns was used, but I am not sure if it was real 😁. But it worked in the end.

Congratulations!
Do you still have that board?
And may you post a photograph of that SRAMs? 😀
Do you use an Am5x86 @ 160 or Cx5x86 @ 120?

Re: 3 (+3 more) retro battle stations
No, I don't have it anymore.
I used it with a Am5x86 @ 160.

CoffeeOne wrote on 2024-02-03, 22:55:

Yes, I agree. Double banked cache relaxes the timing. I was able to get 512kB single bank cache stable at 40MHz on an Asus PVI-4 […]
Show full quote

Disruptor wrote on 2024-02-03, 20:23:

CoffeeOne wrote on 2024-02-03, 18:19:

Most likely you have got 20ns chips. 15ns chips usually can do 2-1-1-1 @40MHz

... when you drive 2 banks.

Yes, I agree. Double banked cache relaxes the timing.
I was able to get 512kB single bank cache stable at 40MHz on an Asus PVI-486SP3 with 15ns chips, but it was a really super hard to find 5 SRAM chips, that did it.
As a tag a Winbond chip with 10ns was used, but I am not sure if it was real 😁. But it worked in the end.

Put back the original 15ns cache chips and it it's stable now at 40Mhz. Single banked cache btw.

Also played around with a Promise FastTrak TX2000 Ultra ATA/133 RAID card. I don't think it was mentioned here or in the unofficial guide yet but it works great. Only problem is that during a soft or hard reset it doesn't recognize my CF card. A cold boot works around that problem. Seems like a similar problem some people already mentioned with other Promise cards.