The Celeron 766 is designed to run on 1.65V, 1.70V or 1.75V. I believe that it would tolerate 1.80V.

STX gave very detailed and good response, my hat is off to him for patience.

Alessandro, have you ever heard of overvolting CPU when overclocking? CuMine Celerons worked perfectly even at 2.2volts when pushed to higher frequencies (like 667 to 1 Ghz etc).
Of course it will tolerate 1.8 volts, that's why I said just to set the voltage on the adapter to 1.8 volts and forget about it.

I gave advice that was based on first hand experience, sometimes it is just wiser to follow the advice and not to overthink and -complicate things, if there is deeper interest then google it, it has been beaten to death...

I found out I had these..

https://drive.google.com/file/d/16hGWpBe_LDhx … ew?usp=drivesdk

from wich one i start?

Paadam wrote:

STX gave very detailed and good response, my hat is off to him for patience. […]
Show full quote

STX gave very detailed and good response, my hat is off to him for patience.

Alessandro, have you ever heard of overvolting CPU when overclocking? CuMine Celerons worked perfectly even at 2.2volts when pushed to higher frequencies (like 667 to 1 Ghz etc).
Of course it will tolerate 1.8 volts, that's why I said just to set the voltage on the adapter to 1.8 volts and forget about it.

I gave advice that was based on first hand experience, sometimes it is just wiser to follow the advice and not to overthink and -complicate things, if there is deeper interest then google it, it has been beaten to death...

Yes, i overclock and overvolt in the past... but i forget the value, and today this peace of hardware become hard to find, is better not to do dangerous experiment

ok, i have made some test with some cpu that i have.

P3 slot1 500/100/2v seen by the system as a 66x5=333
P2 slot1 400/100/2v seen by the system as a 66x4=266

if the sloket force the mainboard to go at 100mhz in the same way to that cpu force the mainboard to go at 100mhz... no way to force my system to go at 100mhz. My scenario is correct?

AlessandroB wrote:

ok, i have made some test with some cpu that i have. […]
Show full quote

ok, i have made some test with some cpu that i have.

P3 slot1 500/100/2v seen by the system as a 66x5=333
P2 slot1 400/100/2v seen by the system as a 66x4=266

if the sloket force the mainboard to go at 100mhz in the same way to that cpu force the mainboard to go at 100mhz... no way to force my system to go at 100mhz. My scenario is correct?

It certainly seems that way. If you use a 100FSB jumper on a slocket adapter it's just going to set the same FSB request values that your existing 100FSB CPUs are already setting. So if those don't work, I don't see how it's going to be any different with a jumper on an adapter.

I wonder if there's a jumper or DIP switch on the motherboard that can be changed to allow 100FSB CPUs. Normally I'd expect an IBM board to use autodetection though. If there is a jumper, it could even be soldered so not easy to change (or find).

I can't remember for sure, but I think I have an IBM PC300GL motherboard with a VIA chipset on it. If that's true then it makes me think there's lots of versions of these boards with different chipsets.
Are you sure yours is a 440BX (marked FW82443BX) and not a 440ZX-66 or some other chipset? Some Celeron boards of that period didn't have 100FSB capable chipsets on them. Perhaps IBM had different versions of the motherboard depending what CPU it was purchased with.

Paadam has experience with these systems so maybe he knows how to resolve it, but it sounds like your board won't do 100FSB even when the CPU signals are calling for it, so it sounds like you need to stick with 66FSB Celerons.