You probably need microcode support in the BIOS than not only does coppermines, but supports the specific stepping of coppermine. (I've come across "coppermine compatible" boards that work fine with a C0 stepping, but put a D0 in and they act dead unless updated)

BitWrangler wrote on 2022-06-24, 02:55:

You probably need microcode support in the BIOS than not only does coppermines, but supports the specific stepping of coppermine. (I've come across "coppermine compatible" boards that work fine with a C0 stepping, but put a D0 in and they act dead unless updated)

Hmm, well I'm updated to the latest BIOS (PDJT31A, July 19, 2000), so I really wonder. Does anyone have any further information about this board?

Looks like this gentleman was able to use a Tualatin with a Power Leap adapter and the latest BIOS update. This was a 6862, but it is essentially the same machine as the 6275 with an AGP port. This makes me believe that it will run this sort of chip, and something else is going on.

Further development. I tossed the same Slotket/CPU combo in my Dell XPS T550, and it was dead as a doornail as well. I know that these are both OEM machines, but I am beginning to think that either the Slotket is bad or the CPU is. Unfortunately, I don't have another Socket 370 CPU to test out. I guess I have to actually order one just to try it?

I've put an unsupported CPU in a board and had it not POST, but the fans came on. I hope it's a bad CPU rather than the adapter since 370 CPUs are much easier to find.

Verify that the CPU VRM on those mobos actually can output the requested 1.75V. Datasheets are usually available online. The VRM simply won't output anything at all if it gets a VID signal it doesn't recognize, the CPU never gets power, and the computer will appear dead as a post.

If you can't find any datasheets, you could try setting the voltage to 1.8V. It's often the lower limit on older pre-Coppermine VRM:s. The very slight overvoltage shouldn't hurt the CPU.

The auto setting just passes VID from the CPU on, the signal will be the same as if you jumpered it to 1.75V.

a bios mod can be done to add microcode of updated coppermines by request on bios forums

zapbuzz wrote on 2022-06-24, 09:25:

a bios mod can be done to add microcode of updated coppermines by request on bios forums

My experience with OEM Slot1/S370 boards might be very limited, but generally, the computer will still boot without the correct microcode for the specific CPU stepping. It will report the wrong CPU during POST, but that's usually purely cosmetical 😀

The only boards I've encountered that outright refused to boot without the correct microcode has been Intel ones.

must be a cpu fault?

This is a case when Coppermine Celerons come in handy. They cost like $2 each. Whenever I need to test a board, I use those.

My first suspicion would be the motherboard doesn't support Coppermine PIII. Try setting jumpers for 1.8V, that is the lowest Katmai 440BX boards can go.

AlexZ wrote on 2022-06-24, 10:11:

This is a case when Coppermine Celerons come in handy. They cost like $2 each. Whenever I need to test a board, I use those.

Totally agree with this. It's always good to have some (cheap/expendable) spare parts laying around to test unknown hardware with. As a bonus the Coppermine Celerons rule out any potential FSB issues.

Did you have the bios jumper in setup mode?

I always forget that on Intel OEM boards. I just did it last night testing 3 processors I got labeled scrap. I put the P3 700 and it boots fine, I put in the P3 450 and it doesn't boot... same with the P2 400. I thought it was weird as I have yet to find a bad slot 1 cpu, then I remembered the bios jumper. I put it in setup mode and all processors boot just fine.

I don't know why I always forget about it.

rootinit wrote on 2022-06-24, 07:45:

Further development. I tossed the same Slotket/CPU combo in my Dell XPS T550, and it was dead as a doornail as well. I know that these are both OEM machines, but I am beginning to think that either the Slotket is bad or the CPU is. Unfortunately, I don't have another Socket 370 CPU to test out. I guess I have to actually order one just to try it?

Whilst faulty is possible (cpu or slotket), don't rule out compatibility as well.

Although I don't have the IBM board for comparison, I do have a Dell XPS Txxx board with the latest A11 BIOS and, like you, a combination of the same Abit Slotket !!! / SL5QV cpu gives me nothing beyond the orange LED on the board, whereas the same cpu but on an MSI MS-6905 Master sloket starts first time.

A quick test on some other BX boards from QDI & Supermicro gave the same results - MSI works but Abit doesn't. I even changed the cpu to a 850/256/100 SL49G but still no go on any BX board with the Abit until I tried the P2B-DS (single cpu only) which worked without issue. Then moving up to a later chipset - the i810E on my Jetway 913AF slot 1 / skt 370 combo board - every cpu worked on the Abit without issue.

Most of the above were tested at both auto & hard FSB / voltage settings without effect, and can't be solely BIOS related either, given the Dell results. Strange indeed!

Dell XPS Txxx board is what I use a slotket on all the time (a cheap generic one) and I have never had a problem, i think i have that same cpu in it now.

Weird the Abit one is that picky but I don't have one of those.

kaputnik wrote on 2022-06-24, 08:45:

The auto setting just passes VID from the CPU on, the signal will be the same as if you jumpered it to 1.75V.

I trust auto about as much as I'd trust a wild polar bear to lick peanut butter off.... any part of my body. I've had boards tell me to just use auto settings whereupon they refuse to work until set manually.

Also a thing with slot 1 in general is it can tend to work only after you've pulled and reinstalled the CPU 3 times.

kaputnik wrote on 2022-06-24, 08:45:

Verify that the CPU VRM on those mobos actually can output the requested 1.75V. Datasheets are usually available online. The VRM simply won't output anything at all if it gets a VID signal it doesn't recognize, the CPU never gets power, and the computer will appear dead as a post.

If you can't find any datasheets, you could try setting the voltage to 1.8V. It's often the lower limit on older pre-Coppermine VRM:s. The very slight overvoltage shouldn't hurt the CPU.

The auto setting just passes VID from the CPU on, the signal will be the same as if you jumpered it to 1.75V.

I couldn't find that info from IBM. Unfortunately, setting to 1.8V and even 1.85V didn't resolve the issue. Just for kicks, I even put it at 2V, like a Katmai, and it still didn't work.

Grem Five wrote on 2022-06-24, 10:47:

Did you have the bios jumper in setup mode?

I don't believe this jumper exists on this board.

PC Hoarder Patrol wrote on 2022-06-24, 14:40:

Whilst faulty is possible (cpu or slotket), don't rule out compatibility as well. […]
Show full quote

rootinit wrote on 2022-06-24, 07:45:

Further development. I tossed the same Slotket/CPU combo in my Dell XPS T550, and it was dead as a doornail as well. I know that these are both OEM machines, but I am beginning to think that either the Slotket is bad or the CPU is. Unfortunately, I don't have another Socket 370 CPU to test out. I guess I have to actually order one just to try it?

Whilst faulty is possible (cpu or slotket), don't rule out compatibility as well.

Although I don't have the IBM board for comparison, I do have a Dell XPS Txxx board with the latest A11 BIOS and, like you, a combination of the same Abit Slotket !!! / SL5QV cpu gives me nothing beyond the orange LED on the board, whereas the same cpu but on an MSI MS-6905 Master sloket starts first time.

A quick test on some other BX boards from QDI & Supermicro gave the same results - MSI works but Abit doesn't. I even changed the cpu to a 850/256/100 SL49G but still no go on any BX board with the Abit until I tried the P2B-DS (single cpu only) which worked without issue. Then moving up to a later chipset - the i810E on my Jetway 913AF slot 1 / skt 370 combo board - every cpu worked on the Abit without issue.

Most of the above were tested at both auto & hard FSB / voltage settings without effect, and can't be solely BIOS related either, given the Dell results. Strange indeed!

That is great! You are doing testing where I am unable to. That is really helpful information. Thank you so much. So, perhaps the Abit is not so good with BX boards, and I should work on getting one of those MSIs. Sure the Dell and the IBM are not the same, but they are both 440BX boards, and that is a theme with the Abit.

rootinit wrote on 2022-06-24, 17:04:

Grem Five wrote on 2022-06-24, 10:47:

Did you have the bios jumper in setup mode?

I don't believe this jumper exists on this board.

Probably not on IBM but the Dell board has one.