LocoMJ wrote on 2020-11-07, 11:03:Thanks vetz for all the advices :) […]
Show full quote
Thanks vetz for all the advices 😀
After thinking of all options, I think I'll try to install a P200 MMX and see how it works. I see that minimum vcore for all MMX is 2.7V, so 2.5V seems not enough. But, I think I can mod this motherboard.
This is a picture of the motherboard I found on the Internet.
You can see JP32 and JP33 near the processor. Those are the jumper you have to change to get 2.5v on the vcore. Also you can see near those jumpers a big voltage regulator. I'll have to open the PC to check it, but I think that's the voltage regulator that gives 2.5V.
Maybe I could change it for another regulator that gives 2.8V.
What do you think?
First things first, test what it's doing now. 2.5V was a very pre-pre-spec goal for the P55C, if your board actually has onboard VRM and not space for an add-in VRM module (like in the manual), it may well already output the required 2.8V.
Now, if it's 2.8V you can obviously use a P55C (why not 233MHz instead of 200MHz?). But the P55C is about the most over-engineered CPU on the planet. Even if it only outputs 2.5V, there's a good chance it will just work - those P55C things will work way out of clock and voltage spec, and undervolting is never dangerous for the hardware, at worst it could be unstable.
Now, if it is 2.5V, I'd second the K6-2 400 suggestion. There you're overvolting slightly, but given that the K6-3 ran at 2.5V anyway (and that was same core as K6-2) it's not a big issue. K6-3 is harder to find but would also add a few percent on top of K6-2 and actually run in-spec at 2.5V.