slivercr wrote on 2017-12-16, 18:10:

ACPI Windows XP: the end?

Yep. I ran into some trouble—finally—when attempting to replace the noisy Radeon 8500 with a GeForce4: after installing the nvidia drivers Windows would refuse to boot up just in the same way it refuses to do it when you force ACPI during install (POSTs, starts the boot up and shows you the WinXP loading screen, screen goes blank and resets). I removed the GF4, put the Radeon back in, uninstalled the nVidia drivers and installed Catalysts: the machine works again. If I try any other nvidia card, Windows won't boot after installing the drivers.

I guess I have to backtrack and recommend against doing this. Windows XP is still usable if you do a normal install from the CD: it will use the MPS Multiprocessor HAL and everything will be peachy.

So the ACPI makes the problem?

So the ACPI makes the problem?

I tried now as MPS Multiprocessor, and it boots.
The only problem, that the windows is not able to switch off the computer, I have to do it myself.

I made some benchmark on my OR840 with WinXP SP2.
I have a IDE->Sata converter from Delock, to attach the 600MByte/s Kingston SSD as a Windows System partition.
I made a HDD benchmark with crystaldisk, I was curious what IDE 66 is able to. The result was about 10% of the SSD speed, see picture below.

I did 3DMARK01 tests with 1 processor and then with 2 processors. VGA was Abit Siluro Geforce 4200 TI (driver detonator 30.87).
The result was in both case about the same. I did a test with the latest driver, but the result was less, only 5779 Pts with 1 CPU and 5836 Pts with 2 CPUs.
I have to mention again, that the RAMs are 4 Slot 512MB PC800 RDRAM in Dual-Channel with PC800.

I did also a Sysoft Sandra 2002 Test on the CPU, unfortunately it says "Current CPU" only, but it was a dual Pentium 3 933MHz.

I was not able to make a memory benchmark test in the Sysoft Sandra, I don't know why, but the result was zero, it just skipped the test.

Use older AIDA64 or AIDA32 to do the memory bandwidth and storage transfer performance and other host of things as well.

Cheers,

Hello!

I am starting this jurney with a AP550 motherboard.
I got the Compaq PSU for cheap but it is the 460w version.
It won’t perfectly fit in my case, but it will stay somehow…
It is not beautiful to see but all connections are plugged.
I was actually able to find the front panel pins, but I cannot find a scheme for it.

Any help?

Plus, I do not have experience with rimms, should I populate all slot?

Thanks

Pchoarderpatrol helped me with pinout.
Thanks.
Unfortunately, the machine do not work.
When I power it up I hear a little “click” on the psu, but nothing else. Nothing spins or bleeps.

I used the 460w version of the Compaq PSU, much cheaper but with different specs.
Could be this the reason?
The motherboard seems totally ok….

Mamba wrote on 2023-02-03, 08:23:

Pchoarderpatrol helped me with pinout. Thanks. Unfortunately, the machine do not work. When I power it up I hear a little “click […]
Show full quote

Pchoarderpatrol helped me with pinout.
Thanks.
Unfortunately, the machine do not work.
When I power it up I hear a little “click” on the psu, but nothing else. Nothing spins or bleeps.

I used the 460w version of the Compaq PSU, much cheaper but with different specs.
Could be this the reason?
The motherboard seems totally ok….

Have you tried bench testing the power supply to make sure it works?

How?

Jumper the atx connector to trigger the PSU to turn on and then use a volt meter to test all the voltage rails.

Isn’t it proprietary? Honestly it is beyond my expertise

Mamba wrote on 2023-02-03, 14:51:

Isn’t it proprietary? Honestly it is beyond my expertise

That I'm not sure do you have the pinout of the motherboard connector?

It is proprietary Compaq wtx pinout. Do not know how to jumpstart it.
Guess I will give up with this project.

I thought you had the pin out I guess I misunderstood your previous post. Sorry about that I wish I could help you more.

chrismeyer6 wrote on 2023-02-03, 18:41:

I thought you had the pin out I guess I misunderstood your previous post. Sorry about that I wish I could help you more.

One thing is the pinout of the motherboard, another thing is to jumpstart a psu alone that is also proprietary design.
I have no clue how to do that.

Mamba wrote on 2023-02-03, 18:56:

chrismeyer6 wrote on 2023-02-03, 18:41:

I thought you had the pin out I guess I misunderstood your previous post. Sorry about that I wish I could help you more.

One thing is the pinout of the motherboard, another thing is to jumpstart a psu alone that is also proprietary design.
I have no clue how to do that.

As discussed, it may still be an issue re using the 460W PSU but from your case pic above you don't seem to be showing the correct memory module setup, which will prevent the system from booting - all 4 slots need to be suitably filled with matching pairs...either by RIMMs or CRIMMs using the attached info.

Yes I know.
I used two identical pairs (2x512 + 2x64).
No boot.
That is why I am asking for help on the psu.

Mamba wrote on 2023-02-04, 06:57:

Yes I know. I used two identical pairs (2x512 + 2x64). No boot. That is why I am asking for help on the psu. […]
Show full quote

Yes I know.
I used two identical pairs (2x512 + 2x64).
No boot.
That is why I am asking for help on the psu.

If you haven't already done so, I might be inclined to test using only the two 64MB RIMMs (obviously this would also require a pair of CRIMMs) - these OR840 based board can be a little cranky with the 512s, through age (caps?), board version & BIOS support. Can't remember if I had any such issues with my AP550 when I got it a few years back, but I'm sure it was on the latest v3.05 BIOS anyway. Certainly none of matched pairs I use on any of my OR840 based boards cause booting issues now - what brand / model no / specs are yours?

As far as bench testing the PSU goes, this post from @nhattu1986 might suggest how to jump start it! Re: Weird Compaq Motherboard

An other OR840 is still alive. I got the Board in the year 2009, but was not such motivated, because the Board makes trobble. Now a friend of mine did a recap and now it is as new. It is an IBM Version.

The memory performance is not what i expected. I don't see any point in the dual-channel Rambus compared to other chipsets. Or what do you think?

And a benchmark with a Geforce 6800 Ultra.

In the End the OR840 is nice exotic Board, but a dual P-III Board with a VIA Apollo Pro133A (CUV4X-DLS) or VIA Apollo Pro266 (CUV266-DLS), even my i820 (P3C-D) is more performant. Or it feels like 😀

So you have 2 Tualatin on it?

Strange, not very good performances indeed.

MadYoshi wrote on 2023-04-20, 08:19:

An other OR840 is still alive. I got the Board in the year 2009, but was not such motivated, because the Board makes trobble. No […]
Show full quote

An other OR840 is still alive. I got the Board in the year 2009, but was not such motivated, because the Board makes trobble. Now a friend of mine did a recap and now it is as new. It is an IBM Version.

IMG_20230330_190233_1.jpg

The memory performance is not what i expected. I don't see any point in the dual-channel Rambus compared to other chipsets. Or what do you think?

cachemem.png

And a benchmark with a Geforce 6800 Ultra.

IMG_20230403_174648_2.jpg

3DMark01_10957.JPG

In the End the OR840 is nice exotic Board, but a dual P-III Board with a VIA Apollo Pro133A (CUV4X-DLS) or VIA Apollo Pro266 (CUV266-DLS), even my i820 (P3C-D) is more performant. Or it feels like 😀

I820 is slower and rambus on the 840 is bottle necked by the 133fsb, it wasn't untill P4 that rambus made since. Even still the 840 at stock clocks is faster than the apollo pro 133 at stock clocks. But unless youre using some benchmark that leverages dual CPU its pretty pointless especially for gaming benchmarks that are single threaded.

can you try this patch and compare it to single cpu
We found a dual Cpu patch (SMP) that works on old games.