DoZator wrote on 2025-06-20, 19:14:

...
All that's left is to figure out a way to reset the multiplier to "40" under Windows XP (installed on the same PC, on a separate partition). CrystalCPUID, in principle, allows you to change the multiplier on a given processor (using the /Fxx parameter), but it hasn't been updated for a while, and it doesn't seem to fully support the 4th generation. It has some internal limitations that prevent the multiplier from exceeding a certain value (based on my rough estimates based on benchmarks, it's around "31"), even if you use the "/F99" parameter (which should theoretically increase the multiplier to its maximum allowed value). Unfortunately, I haven't found any other XP-compatible tools (CPUSPD doesn't officially support XP).

Older versions of ThrottleStop work with Windows XP. I have just shared a proven to work version of it (ThrottleStop_600) in another topic here:
Re: x86 microarchitecture benchmark (MandelX)

Falcosoft wrote on 2025-06-20, 19:35:

DoZator wrote on 2025-06-20, 19:14:

...
All that's left is to figure out a way to reset the multiplier to "40" under Windows XP (installed on the same PC, on a separate partition). CrystalCPUID, in principle, allows you to change the multiplier on a given processor (using the /Fxx parameter), but it hasn't been updated for a while, and it doesn't seem to fully support the 4th generation. It has some internal limitations that prevent the multiplier from exceeding a certain value (based on my rough estimates based on benchmarks, it's around "31"), even if you use the "/F99" parameter (which should theoretically increase the multiplier to its maximum allowed value). Unfortunately, I haven't found any other XP-compatible tools (CPUSPD doesn't officially support XP).

Older versions of ThrottleStop work with Windows XP. I have just shared a proven to work version of it (ThrottleStop_600) in another topic here:
Re: x86 microarchitecture benchmark (MandelX)

All right, I figured it out. In fact, my mistake was that I changed not only the multiplier to "8" in the BIOS, but also the "Uncore Ratio" to the same "8" (assuming that these two settings should change synchronously). However, this is not the case. As a result, the performance did not return to its previous level. In fact, there is no need to adjust the "Uncore Ratio" at all. I've returned the "Uncore Ratio" to its default value. In general, both "ThrottleStop" and "CrystalCPUID" (with the /F99 parameter) correctly return the multiplier to the "40" value, and the performance is now fully restored. And those of my assumptions that you quoted above, regarding "CrystalCPUID" - NOT TRUE (CrystalCPUID, when launched with the parameter "/F99", quite correctly returns the multiplier back to the value "40", in combination with this processor, which is confirmed by the readings "ThrottleStop" and benchmark results). The problem is completely solved. Thank you for your help.

Hi All,

Long time no see, hope all is well.

Attached is a spreadsheet that might be of interest to you. I used it to automatically look up the equivalent cpu according to the DOOM realticks when developing CPUSPD. It also has the results for some of the CPUs I tested and recorded.

Just plug in the DOOM realticks from your test into column L, and column M ("Equivalent CPU", blue) will look it up from the "Reference" sheet and show you the results. The "Reference" sheet contains the results posted by one member on Vogons (sorry it was too long ago and I've forgotten who). Note that DOOM can't really run on 286 and below so the realticks for those cpus are interpolated.

Cheers!