More important than the BIOS is the VRM on that board. If it's able to give the 1,475/1,5 required by the Tualatins, you're good to go with an adaptor (Lin Lin on a Coppermine Slot 1-> Slot370 or Slot-T adaptor). If not, you'll need an active adapter like the Powerleap (if you're able to source one, let us know where).

Driving two of these babies on a dual Slot 1 is another issue altogether, but you'll be already happy as a lark with one functioning.

flupke11 wrote on 2022-04-28, 21:10:

More important than the BIOS is the VRM on that board. If it's able to give the 1,475/1,5 required by the Tualatins, you're good to go with an adaptor (Lin Lin on a Coppermine Slot 1-> Slot370 or Slot-T adaptor). If not, you'll need an active adapter like the Powerleap (if you're able to source one, let us know where).

Driving two of these babies on a dual Slot 1 is another issue altogether, but you'll be already happy as a lark with one functioning.

Voltage is not a problem at all.
As soon as the board support Coppermine and goes down to 1.65 (which the 3.0 does), there are zero problems.
I can tell because I had zero problems myself on P2B or other old BX board and 1.7V on P-IIIs

No, the issue lies somewhere else as this guy:
https://pcrebuilding.altervista.org/1/28/PC+R … e+Tualatin.html
was not able to run a tually on a P6DGU that is pretty the same, even with a proper adapter.

Any direct experience?

It would be lovely to have an Intel system with Tualatin and more than 1GB on it.

The only 440GX I have is a Slot 2 system, so I cannot test Tualatin cpu's on that one.
Most 440BX will accept the Tualatin Celeron, with the correct adapter. They might fuss at boot about an incorrectly identified cpu,but will boot either way.
Running the BX out of spec at 133 and stressing the VRMs to accommodate the 1,4/512 is however not always possible. I am lucky to have an Asus P3-BF which is one of the fastest boards for Tualatin if oc'd at 133.

If you absolutely need more than 1GB, these are options I was able to test myself: Compaq AP500 (Rdram!), Supermicro P3TDE6 (4GB ram, but AGP issues), Aopen Ax34 (fastest Via694X I have seen), Asus P3B-F.

flupke11 wrote on 2022-04-29, 06:42:

The only 440GX I have is a Slot 2 system, so I cannot test Tualatin cpu's on that one. Most 440BX will accept the Tualatin Celer […]
Show full quote

The only 440GX I have is a Slot 2 system, so I cannot test Tualatin cpu's on that one.
Most 440BX will accept the Tualatin Celeron, with the correct adapter. They might fuss at boot about an incorrectly identified cpu,but will boot either way.
Running the BX out of spec at 133 and stressing the VRMs to accommodate the 1,4/512 is however not always possible. I am lucky to have an Asus P3-BF which is one of the fastest boards for Tualatin if oc'd at 133.

If you absolutely need more than 1GB, these are options I was able to test myself: Compaq AP500 (Rdram!), Supermicro P3TDE6 (4GB ram, but AGP issues), Aopen Ax34 (fastest Via694X I have seen), Asus P3B-F.

Thanks,
I already know. It is not my request nor my will actually but thanks for your insight.

The board is in my hands now.
It has two HIP 6004BCB voltage regulators, they should be capable of going down to 1.4V

I really do not see reason why a Tually cannot boot, with adapter of course.

No way, the Tualatin is not working.
I tried at 1,45v and at 1,7V both, at 100Mhz FSB of course.

Black screen.

My SL5QW 1100 works but at 1Ghz only and the motherboard shows microcode error.

The bios is 3.1, so the latest one.
Maybe it should be updated with proper microcodes (Tualatin and later cumine)?

Any assistance?

Never found much info online re. Tualatins on the 440GX, and you already linked the best of them with the pcrebuilding one. There was a thread on here where a member (@Dant) apparently got some to work on a P6DGH, but never any of the top end ones - Pentium III-S (1400/512/133) @ 100mhz FSB?

What board version do you have - the earlier P6DGEs used a completely different BIOS set, both available here https://www.supermicro.com/support/bios/archive.cfm
What slokets are you using now (same as last time we PM'd?)
Although the POST screen shows 1GHz for the 1.1GHz chip, have you checked what it's actually running at with any utilities / benchmak apps?
Microcode errors are a pain but can generally be ignored
Do you have JP11 (FSB jumper) set to Pin 1-2 AUTO, and are you using a terminator board in the 2nd slot? (both recommended by SM for higher frequency processors)?

Hello,
The board is rev 3.0, the latest, with VRMs capable of going down to 1.4v.
The SL5QW is a 1.1Ghz 100Mhz FSB chip so it is strange to see 1.0ghz.
The adapter is a Asus s370-DL with a modified Tualatin that worked literally everywhere I tried (P2B included).
I think is a bios/microcode issue, but the procedure described on Pc rebuilding is beyond my expertise.

I am really trying hard to find microcodes for Intel Pentium III late coppermine (like my SL5QW) and tualatin, with no luck.
Can someone help on this?

Mamba wrote on 2022-05-10, 19:31:

I am really trying hard to find microcodes for Intel Pentium III late coppermine (like my SL5QW) and tualatin, with no luck.

Microcodes are still available from Intel. SL5QW has a CPUID of 068A, so you'll need the file intel-ucode\06-08-0a. (The first 2048 bytes are enough; the rest of the file contains updates for other platforms.)
(I haven't checked how to get microcodes into an AMIBIOS, though.)

Thanks,
PC Hoarder Patrol helped me already.
I need to test the bios.
Combining the site and the discussion linked, I have the feeling that somehow the SM board prevents any chip with 133 FSB to work.
If it is the case… Well, maybe only a very AMI bios expert could actually look at it.

Just tested the modded bios.
Same behaviour, a SL5QW 1100 chip post at 1000Mhz.
I suspect there are two blocks:
- one prevents multiplier over 10X
- one prevents cpu with 133mhz to post

Do not know if this can be unlockable somehow.

Same behaviour as here:
Coppermine 1100 doesn't work via slotket

Mamba wrote on 2022-05-18, 05:31:

Just tested the modded bios. Same behaviour, a SL5QW 1100 chip post at 1000Mhz. I suspect there are two blocks: - one prevents […]
Show full quote

Just tested the modded bios.
Same behaviour, a SL5QW 1100 chip post at 1000Mhz.
I suspect there are two blocks:
- one prevents multiplier over 10X
- one prevents cpu with 133mhz to post

Do not know if this can be unlockable somehow.

Same behaviour as here:
Coppermine 1100 doesn't work via slotket

Not sure either 'block' is relevant -

- strictly speaking these LX / BX / GX boards are considered by SM to max out at a core / bus ratio of X 9.5 (which is what they provided jumper settings for), but this is irrelevant to locked processors at X 10 and beyond. It'll be a cosmetic problem on the initial BIOS POST screen (I have a BX board which claims all 1000MHz processors are actually 0MHz - can't deal with 4 digits). What do programs like speedsys or hwinfo say the processor identifies as / is running at - my guess would be the correct values rather than what the BIOS mistakenly claims.

- based on my P6DGU, 133MHz processors shouldn't be an issue in general. I have run this board with dual slot 1 1GHz / 256 / 133 processors (SL4BS) both at the default 100MHz FSB for 750MHz (7.5 X 100) and overclocked thru a BIOS setting at 133MHz for 1GHz (7.5 X 133) without issue.

Problem is, that a slot-1 800/256/133 chip does not post.
That is why I ended up with the two assumptions.
Seems like the board can go 133mhz (in bios)
and can go 1.4v as VRMs are meant to go that low.
But it simply does not at post.

Mamba wrote on 2022-05-18, 06:38:

Problem is, that a slot-1 800/256/133 chip does not post. That is why I ended up with the two assumptions. Seems like the board […]
Show full quote

Problem is, that a slot-1 800/256/133 chip does not post.
That is why I ended up with the two assumptions.
Seems like the board can go 133mhz (in bios)
and can go 1.4v as VRMs are meant to go that low.
But it simply does not at post.

Hmm, that seems a little hinky!!

So I rechecked the theory, this time on the sister board the P6DBU (same PCB / BIOS but BX instaed of GX chipset - couldn't try the P6DGU as its not booting atm...think it needs a full recap) with the same results being 2 x 1GHz / 256 /133 SL4BS happily boot at 750MHz.

More relevant though, is that the P6DGE is clearly just as capable of the same as this old post describes - https://www.motherboardpoint.com/threads/p6dg … -support.56173/ - which also links to these

- https://web.archive.org/web/20060426190853/ht … q/Processor.asp
- https://groups.google.com/g/alt.comp.periphs. … 0ba9ddacc8502e1

Supermicro support also kind of refer to the possibility back in 2004 (P6DBE / P6DGE??) - https://www.supermicro.com/support/faqs/faq.cfm?faq=1311

Do not know what to say…

Mamba wrote on 2022-05-18, 09:54:

Do not know what to say…

Back to basics I guess -

- whats the S-spec (SLxxx) # of the 800MHz slot 1
- what SDRAM memory
- what jumper settings for JP11 (Bus Speed) & JB1 thru JB4 (CPU/Bus Ratio)
- are you using a terminator board in the 2nd cpu slot as suggested
- any sign of caps failure (blown tops or popping / tilting from below)
- got a clear hi res pic of the whole board without cpu / memory
- got a pic of the post screen (something like this) - download/file.php?id=89024&mode=view