Can't set the SL2Z4's multiplier to 4x on a GA-5AX

[CHips]

Oh, while we are at it... Is there any way to get 1x multiplier? It would be cool to run a Fdiv CPU at 1 x 100 MHz :p ...

1.5 x 75 MHz = 112.5 MHz is not stable with my Sx960 (and 1.5 x 83 MHz is not even Posting)!

[frag_]

Afaik for classic pentium minimum multiplier is x1.5, for pentium mmx it is x2.0.

And bios has nothing to do with pentium multiplier, because there is no way to set it from software (unlike e.g. K6-2+).
All that matters is logical levels on FB2..FB0 during reset.

[cuttingedgecs]

While I agree that ultimately the multiplier is controlled by the CPU based on the inputs it receives on BF0-2, and that on many older boards, those pins were connected directly to jumpers, and that in no way involved the BIOS, I don't think that's what this board is doing. While the physical layout and pin assignments on socket 7 CPUs was the same, there were minor electrical differences due to process choices in things such as the values of required pull up/down resistors etc. This meant that in order to support the most brands of chips boards would read the jumpers and control BF0-2 with gate logic. The fact that a 4x multiplier setting is yielding a 2x result is evidence that this is what's happening, as 2x is what a P55C defaults to when BF0-2 are neither tied high or low. If the jumpers on the board directly tied to BF0-2 4x would work, and if they directly tied to BF0-1 with a fault on the CPUs BF2 line, that should result in a multiplier for 2.5 not 2. I suspect the some chip on the board is detecting a Pentium, and detecting the 4x setting, and not knowing what to make of that, is not signalling BF0-2 at all. I doubt that would be the BIOS though, as the BIOS runs using the CPU, and BF0-2 need to be set the moment the CPU starts up.

[H3nrik V!]