That PM54CPA-A1 sounds simple but is thoroughly non-googlable. The -A1 on the end sounds reminiscent of Tekram, but Tekram P1 boards start with P5, not PM...

Is it maybe an generic OEM motherboard perhaps? The strange thing is, it doesnt have an onboard IDE / floppy / LPT section of ports. Considering that later 486 boards had that intergrated.

Robin4 wrote on 2020-02-22, 23:34:

Is it maybe an generic OEM motherboard perhaps? The strange thing is, it doesnt have an onboard IDE / floppy / LPT section of ports. Considering that later 486 boards had that intergrated.

I think that A1 is only a revision number.

Searching for the bios string 2A5KBP51C (and including ALi 1449/61/51) gives 2A5KB as Ali 1449/61/51 Chipset, the P5 yields nothing. (P4 is Asus, P6 is Pro-Tech) and 2A5KBP51 lists ?? A diff search and parts there of yielded 2A5KBP51 and 2A5KBP52 as Proect as manufacture but nothing good comes up after that..

Robin4 wrote on 2020-02-22, 22:43:

It should have alladin chipset.. But i dont know the maker or i dont have any manual for the jumper setting.
Also dont know if there are chipset drivers ever released for this board.

I think it is safe to assume that there is no such thing as "chipset driver" for this board. There is no I/O 😀

Which jumper are you not sure about?

Got the same chipset on a Socket 4 board. It is the original Alladin.

I am surprised they hacked it to work with a 3V CPU. Look at the small army of 5V <-> 3.3V bus transceivers above the CPU. That's hilarious.

I not familiar with this board settings.. Its trail and error now.. But i can make a list in the message window.

JP1. (unknown
JP2 Hardwired
JP3 ? (maybe isnt there)
Jp5 Single Jumper North east of the cpu socket (doesnt know what it does) Jumper block 1 by 2
JP6
JP7
JP8

JP9 Two horizontal jumpers on 1-2 and 3-4 bus speed setting.

Pin 1-2 & Pin 3-4 open - closed 33 mhz FSB with jp23 open = 50 Mhz CPUclock closed = 66Mhz CPUclock
Pin 1-2 & Pin 3-4 open - open 50mhz with jp23 open = 75 Mhz CPUclock closed = 100 Mhz CPUclock
Pin 1-2 & Pin 3-4 closed - open 60 mhz FSB with jp23 open = 90 Mhz CPUclock closed = 120Mhz CPUclock
Pin 1-2 & Pin 3-4 closed - closed 66mhz FSB with Jp23 open = 100 Mhz CPUclock closed = 133 Mhz CPUclock

JP10
JP11 Front panel Reset
JP12 Front panel turbo switch
JP13 Front panel Turbo led
JP14
JP15
JP16
JP17 Has anything to do to set the cache size (dont know the config yet)
JP18 Has anything to do to set the cache size
JP19
JP20
JP21
Jp22
JP23 Does anything with the processor speed.. Its set the 133 Mhz to 100 Mhz when removed. (multiplier?) Removed set at 1.5x times as all earlier boards., closed set it to 2.0 x so processor runs at 133 Mhz with 66Mhz system bus.
JP24
JP25
JP26 (here is a jumper blok of 2 by 5 rows (thought it could be a com 1 connector, but isnt silkscreened as com1.) No jumper caps here.

Robin4 wrote on 2020-02-23, 01:28:

I not familiar with this board settings.. Its trail and error now.. But i can make a list in the message window. […]
Show full quote

I not familiar with this board settings.. Its trail and error now.. But i can make a list in the message window.

JP1. (unknown
JP2 Hardwired
JP3 ? (maybe isnt there)
Jp5 Single Jumper North east of the cpu socket (doesnt know what it does) Jumper block 1 by 2
JP6
JP7
JP8

JP9 A jumper block of 2 by 2. And both jumpers are installed (thought something for configuring for EPP or so) but could be something different.
JP10
JP11 Front panel Reset
JP12 Front panel turbo switch
JP13 Front panel Turbo led
JP14
JP15
JP16
JP17 Has anything to do to set the cache size (dont know the config yet)
JP18 Has anything to do to set the cache size
JP19
JP20
JP21
Jp22
JP23 Does anything with the processor speed.. Its set the 133 Mhz to 100 Mhz when removed.
JP24
JP25
JP26 (here is a jumper blok of 2 by 5 rows (thought it could be a com 1 connector, but isnt silkscreened as com1.) No jumper caps here.

First iam going to purchase new dallas DS12887 clock chips.. Hopefully it would fix the cmos error and low battery error.

The good news on this board is that the CMOS checksum error is gone becausei swapped out the RTC module, for one that has some life in it. So the bios seems to be fine.

The floppy drive error C0 is there because there is no controller on board of this motherboard. Iam guess it will be gone when i populate an ISA IDE / floppy controller on this board.

Also would really like of someone had some information about this motherboard.

My best guess is this is some industrial motherboard meant to be useable with a broad variety of controller cards and whatever resources they require so the design is as lightweight as possible.
Usually documentation is hard to come around since those boards weren't shipped to consumers individually.

Tried 1024KB cache on this board, doesnt seems to work. Also trying to set the nearst by jumpers on every setting possible. Ending in a board that locks on boot, or dont want to start (black screen)
Before using the 1024KB chips i verified them with my Mini pro and they seems to be good. Does that mean that not every board with 32 pin dip cache sockets will actually will support 128x8 sram chips? PCB say so. 128*8.
Then i going further with 256KB srams that i have in spare. Installed them.. Board is starting.. and detects 256Kb cache. Only bad thing its slow as hell on boot trying detecting ide controller..

Iam also thinking it could be an industrial motherboard. But it its not strange that pre-pentium class motherboard didnt had floppy or IDE on board. I think that this board was released between socket 4 and the first socket 5 boards.

Board works with a PTI-277w version 3 isa controller.. No errors anymore. Yeah..

Only have to find out the jumper settings of this motherboard. And that is pretty hard without any documentation or manual. any suggestions?
And iam waiting on some RTC replacements. I dont know if the max of this board is Pentium 133. But i find this already special for a pre-socket 5 board.

Iam guessing that JP23 Jumper is for the multiplier.. Because when removed it makes from the pentium 133 to 100 Mhz.. When close its 133mhz. (so i think its a 1.5 / 2.0 multiplier..) I dont think that the board have a higher multiplier then 2x for its time period.

Somebody knows if there are dos programs that detects the processor frontside bus as the multiplier?

EDIT:

And i found the FSB Jumpers @ JP9 (horizontal position)

Pin 1-2 & Pin 3-4 closed - closed 66mhz FSB with Jp23 open = 100 Mhz CPUclock closed = 133 Mhz CPUclock
Pin 1-2 & Pin 3-4 open - closed 33 mhz FSB with jp23 open = 50 Mhz CPUclock closed = 66Mhz CPUclock
Pin 1-2 & Pin 3-4 closed - open 60 mhz FSB with jp23 open = 90 Mhz CPUclock closed = 120Mhz CPUclock
Pin 1-2 & Pin 3-4 open - open 50mhz with jp23 open = 75 Mhz CPUclock closed = 100 Mhz CPUclock

Hi,

I bought the same board on ebay. Thank you very much for the clock speed settings.

I spented a lot of time to find this board in the stason database (with the features: 4x ISA, 4xPCI and RTC / Bios between it) but I didn't find anything (also not a similar board).

Two hints:
- be careful with the battery holder: in my opinion it has the wrong polarity: plus pin of the battery is connected to ground.
(if you will use an DS12887 again, you doesn't need an external battery, so it doesn't matter).
- I got a PCI-IDE controller (Tekram DC-290N) up and running on this board but only with using the ISA interrupts via a "paddle-board" (legacy PCI mode). Up to now I haven't had success with native PCI mode for IDE.

Hi Robin4,

today I did some reverse-engineering...

For the cache jumpers JP17, JP18 the following configurations work for me (TAG-Chip is always an UMC UM61256FK-15 with 32KByte) :
JP17 JP18
1-2 1-2 256kBytes (8x ISSI IS61C256)
1-2 2-3 512kBytes (8x ISSI IS61C512)
2-3 2-3 1MBytes (8x ISSI IS61C1024)

JP5 can connect socket 5 pin W33 with Pin 4 of the PCI-ISA bridge Ali M1449 but Intel and AMD say: W33 is not connected...?!
JP6 is on one side connected to pin S5 of the socket 5. According to the Intel datasheet this pin is used for debugging: "BreakPoint3". In case of an AMD K5 this pin is not connected.
JP10 can set the pin 204 of Ali M1449 to 5V (default is 10k to ground). I haven't found a datasheet for this chip, so I will try it out later - perhaps anything happens...
The function of JP29 is unknown for me (one row is ground) - I found in the manual of the Chicony CH1451A a similar construct: "Green function connector" but in this case: the pins are outputs and not for jumpers.

LeoAndFanny wrote on 2020-02-28, 13:39:

Hi, […]
Show full quote

Hi,

I bought the same board on ebay. Thank you very much for the clock speed settings.

I spented a lot of time to find this board in the stason database (with the features: 4x ISA, 4xPCI and RTC / Bios between it) but I didn't find anything (also not a similar board).

Two hints:
- be careful with the battery holder: in my opinion it has the wrong polarity: plus pin of the battery is connected to ground.
(if you will use an DS12887 again, you doesn't need an external battery, so it doesn't matter).
- I got a PCI-IDE controller (Tekram DC-290N) up and running on this board but only with using the ISA interrupts via a "paddle-board" (legacy PCI mode). Up to now I haven't had success with native PCI mode for IDE.

Yeah i noticed too that the battery coin holder wouldnt work.. Tried it out for some minutes, then removing the battery.. But ill keep it witih the battery in the Dallas RTC. Only dont know why there is a battery holder installed when its not in use.
Iam just using a regular combo IDE controller which resides al I/O on one board, so it wont use more slots then needed.

In the bios i set the setting for controlling the IDE controller to ISA instead of PCI..

Does anyone know where to get the ULI / ALI M1451 chipset drivers for windows 95 / windows 98 perhaps dos / windows 3.1 / 3.11

Robin4 wrote on 2020-03-08, 00:47:

Does anyone know where to get the ULI / ALI M1451 chipset drivers for windows 95 / windows 98 perhaps dos / windows 3.1 / 3.11

You don't need drivers, as there is no onboard i/o. Even DOS and the BIOS support the basic functions of such a board. You may need drivers for your i/o card, but I doubt it if you use anything standard.

Is it true that all SIS 85Cxxx chipsets also dont have chipset drivers?

Maybe thats way i dont find any for SIS 85C501/2/3

So. We have started the massive task of identifying and adding unicore-made bios upgrades that were released for the spectra333/spectra400 upgrade boards. We have a makers name for the P5 manufacturers code: Proect.

https://theretroweb.com/motherboards/s/unknown-pm54cpa