My original goal with this machine was to have a single PC that could competently play games from the early 1990s into the early 2000s. Not sure if anyone else is still following this thread, but at this point, I'm just trying different sorts of CPUs to see if anything reports an actual L2 cache. Otherwise, I've kind of given-up on this whole idea, and will end-up getting a Pentium 4 for the later games.

Anyway, I can also report that a 850MHz/100MHz/256KB/1.75V Coppermine (SL4CC) has the same behavior. No L2 cache is reported. At this point, I am also inclined to believe that this is a problem with the BIOS, but I'm not sure why because there are others who have reported the latest BIOS is working fine with these later CPUs.

I’m following, I wonder if it’s a wiring problem or a bios problem. Seems like maybe a pin isn’t being pulled or routed right..

Perhaps something similar to the tillamook issues.

Read the tillamook working l2 thread and the socket 370 tweaker thread, see if that gives you any ideas. If you find a pin mod for this not listed already in the tweaker thread, be sure to describe it there so it can be added

I don't think it will be a BIOS problem. More likely a damaged trace on board or a damaged socket.

AlexZ wrote on 2022-07-04, 22:49:

I don't think it will be a BIOS problem. More likely a damaged trace on board or a damaged socket.

Does a slot processor work?

Sphere478 wrote on 2022-07-05, 02:22:

AlexZ wrote on 2022-07-04, 22:49:

I don't think it will be a BIOS problem. More likely a damaged trace on board or a damaged socket.

Does a slot processor work?

The Katmai slot processors that I've used work just fine. It is the Coppermine ones (both Slot 1 and Socket 370 w/slotket) that fail to show an L2 cache.

rootinit wrote on 2022-07-05, 07:55:

Sphere478 wrote on 2022-07-05, 02:22:

AlexZ wrote on 2022-07-04, 22:49:

I don't think it will be a BIOS problem. More likely a damaged trace on board or a damaged socket.

Does a slot processor work?

The Katmai slot processors that I've used work just fine. It is the Coppermine ones (both Slot 1 and Socket 370 w/slotket) that fail to show an L2 cache.

So, it’s a bios or board problem then.

rootinit wrote on 2022-07-05, 07:55:

Sphere478 wrote on 2022-07-05, 02:22:

Does a slot processor work?

The Katmai slot processors that I've used work just fine. It is the Coppermine ones (both Slot 1 and Socket 370 w/slotket) that fail to show an L2 cache.

Have you tried to run one of the Coppermines at fsb 66 ?

PARKE wrote on 2022-07-05, 15:24:

rootinit wrote on 2022-07-05, 07:55:

Sphere478 wrote on 2022-07-05, 02:22:

Does a slot processor work?

The Katmai slot processors that I've used work just fine. It is the Coppermine ones (both Slot 1 and Socket 370 w/slotket) that fail to show an L2 cache.

Have you tried to run one of the Coppermines at fsb 66 ?

Not yet. I assume I just need to modify the jumpers on the Slotket to get it to do that. I'll report back soon.

Just realized that's not actually possible to do with this Slotket. Its jumper settings only allow for the main board to choose the FSB speed or for the CPU to do it. Either way, there are no Coppermines that are actually natively at 66 MHz FSB, so I cannot make it do that. 100 MHz FSB processors work fine here as Katmais though, so it shouldn't be a question of whether this board can do 100 MHz or not.

If anyone ever discovers what I'm doing wrong here, let me know, but as far as I can tell, there's nothing else I can really do short of finding yet another slotket.

Technically, green FCPGA Celerons should be 66FSB for anything inbetween 566MHz and 700MHz if I remember correctly. If you have any of those, it would be worth a test.

The 66Mhz ones run from 533A to 766. I think the 766 got popular for being both the highest 66mhz bus part, and the highest multi for overclockage, so they might be priced a bit higher. Can probably find middling ones for what loose change you can dig out of the couch still though.

Celeron 766 is not very useful for overclocking as the chance of it running at 1150Mhz is very slim. Celeron 600-633 is the best.

at least very late 733 and 766 did run at 100MHz

I have a bunch of slockets lying around if you want me to send you one to try out. They are all new so shouldn't have any issues, and will help in the process of elimination. Got a million P3's also if you want to try a different flavour CPU.

Both of these badboys look like they have plenty of jumpers / switches for messing with the FSB / multipliers / voltage etc...

https://ibb.co/VJkbBt0

https://ibb.co/HT0DkRG

I have lots of others. Just packed away "neatly", as in - underneath a million other things which are also packed away neatly.

Maybe try to use another CPU or another slotket.. Also you can try another slot motherboard (preferably normal one, not OEM) just for the sake of testing the components...

The fact that this slotket is NOS, does not make it automatically an working one.. Maybe the caps did dry out and they dont do anything now. There is also the chance of some incompatibility between these exact parts that you are trying to make work together... For a start try that slotket + CPU on non OEM slot board such as ABIT, ASUS, GIGAbyte etc...

rootinit wrote on 2022-07-06, 17:24:

Just realized that's not actually possible to do with this Slotket. Its jumper settings only allow for the main board to choose the FSB speed or for the CPU to do it. Either way, there are no Coppermines that are actually natively at 66 MHz FSB, so I cannot make it do that. 100 MHz FSB processors work fine here as Katmais though, so it shouldn't be a question of whether this board can do 100 MHz or not.

If anyone ever discovers what I'm doing wrong here, let me know, but as far as I can tell, there's nothing else I can really do short of finding yet another slotket.

Still think this is most likely now a BIOS mis-reporting / limitation issue (with any 'workig' sloket) on this board. The earlier mention of similar success on the 300PL / 6862 version with AGP, is down to the fact that this board received explicit BIOS support for coppermines plus numerous coppermine microcode updates - something which was completely lacking on the 300GL / 6275.

PC Hoarder Patrol wrote on 2022-07-09, 11:38:

rootinit wrote on 2022-07-06, 17:24:

Just realized that's not actually possible to do with this Slotket. Its jumper settings only allow for the main board to choose the FSB speed or for the CPU to do it. Either way, there are no Coppermines that are actually natively at 66 MHz FSB, so I cannot make it do that. 100 MHz FSB processors work fine here as Katmais though, so it shouldn't be a question of whether this board can do 100 MHz or not.

If anyone ever discovers what I'm doing wrong here, let me know, but as far as I can tell, there's nothing else I can really do short of finding yet another slotket.

Still think this is most likely now a BIOS mis-reporting / limitation issue (with any 'workig' sloket) on this board. The earlier mention of similar success on the 300PL / 6862 version with AGP, is down to the fact that this board received explicit BIOS support for coppermines plus numerous coppermine microcode updates - something which was completely lacking on the 300GL / 6275.

I'm inclined to agree. While I sincerely appreciate the offer from @Iris030380 to try different slotkets/CPUs, my understanding is (esp. for anyone reading who hasn't followed the whole thread), we could rule out the slotket incompatibility because it has the same bad behavior with a native Slot 1 coppermine CPU. I have no clue about this, but I wonder if there's a way to hack in the coppermine microcode from the 6285 BIOS if that truly exists. I would've thought that they would have nearly the same BIOS already though, because the only difference between these two models was the fact that the PL had an AGP lot, and the 6275 didn't.

rootinit wrote on 2022-07-10, 20:04:

I'm inclined to agree. While I sincerely appreciate the offer from @Iris030380 to try different slotkets/CPUs, my understanding is (esp. for anyone reading who hasn't followed the whole thread), we could rule out the slotket incompatibility because it has the same bad behavior with a native Slot 1 coppermine CPU. I have no clue about this, but I wonder if there's a way to hack in the coppermine microcode from the 6285 BIOS if that truly exists. I would've thought that they would have nearly the same BIOS already though, because the only difference between these two models was the fact that the PL had an AGP lot, and the 6275 didn't.

Hi, finally you tried another slotlek or best CPU?

I have same machine with a lastest bios and a 550 katmai. Thinks on install a slotket or a slot1 800 / 100mhz (i suppose that not support 133).

Can you tell if your project is active or finished?

Regards