The Extreme Pin Modding Guide - Page 4

Hmm... just to clarify, you got a 1.8 and pin-modded it to 2.4, yes? And when you boot up, is it stable or not in 3D applications?

Ahhh, ok, so then what's your stock voltage? Boot up with just the pin mod and open up a voltage modding program (or CPU-Z) to find your stock voltage. What pins you need to mod for what voltage depends on your stock voltage.

Guys, consolidate your posts , thank you kindly .

mr9865, I want you to run prime95 to test the CPU ALONE. Download the program here: http://files.extremeoverclocking.com/file.php?f=103. Install it and select "just stress testing" when you first open it. Go up to options>torture test and select large FFT and click "OK" to run the test. Let me know when it fails the test (could be minutes, could be an hour or more). Post when it fails and I'll be able to tell you what to ground (no worries, I'll explain it too ).

AD1984- I don't recommend running an instable 1.8@2.4 at 2.26GHz because the software control is shaky at best. There are several microseconds where the hardware settings "take over" while the software "catches up" (these are loose terms) in applying the moddified voltage. This can lead to instability where the CPU changes states. It's best just to add some extra voltage to the CPU and make it stable . And good thinking researching the seller you are buying from .

ikilledkenny- I, too, recommend checking out Tweak Guide's Tweaking Companion. It is a phenomenal achievement of Koroush and is just a great resource for tweaking your system and getting to know XP. The Intel Core Duo platform won't offer you much more performance in your current application because the source engine isn't multithreaded (as far as I know). While the source engine is very CPU-dependent, 2.0GHz for a P-M isn't too shabby. What kind of mods do you do exactly? 2GB of RAM might help you out .

Ah, ok. Yah, grounding vid 0 and 1 would only give you 1.388v, that won't keep you stable enough. I would try grounding VID0 (connect E2 and E3 with a wire) and VID 2 (connect F3 and F4 with a wire) which would give you 1.420v if your CPU calls 1.340v stock (because grounding changes the 1 to a 0). Try testing that and see how many hours prime95 goes for (if it goes for more than 15 hours, stop the test, you're stable as can be ).

If it's fine up until an hour then it might be a heat issue. How old is your laptop and have you cleaned the dust out of it recently?

Large FFT tests the CPU and the entire subsystem stability. It puts alot of stress on the CPU to transfer data. Both work equally well though.

And no, 2.0-2.66GHz doesn't seem feasible as the architecture of the Pentium M just doesn't scale that high. The clockspeed and voltage are just too high for "everyday" settings and cooling methods.