Yonah has been pinmodded! - Page 3

It just seems like you are on the last leg of the race. Keep pushing it will work. BTW glad you did not clip the pin off, that would have been sad & funny at the same time.

ok i get it but what i am telling you is that when BSEL0 is in high you run 13x, when it is on low you run 6x, that is my theory, since that is done from the OS, its done at a virtual level, have you tried it yet? then super pi again to see if there is any diff. What about comparing your result to a stock 2600?
here is a guide to change the voltage on desktop conroes. At end it says that he is going to post a fsb mod.
http://www.vr-zone.com/?i=3904&s=2

maybe you can force the voltage and vid with these.
On page 28 from the core duo data sheet there is more information on how the highest clock is managed.

*runs into room* someone doing hardware modifications?

Wow, using ATA100 cabling, great idea! Usually people use glue or clip the pin, both of which are a more permanent solution.

Background for everyone involved: Pin-modding is the sad remenant of overclocking that is left to laptop users due to OEM's locking their machines up tight. In the Dothan days, 533FSB boards were backwards compatible with 400FSB CPUs by means of the BSEL switch. To pin-mod, all one had to do was be throw the switch low via a wiremod and the chipset recognized and ran the CPU at 533FSB.
Fast forward to now, it appears Intel has done the same thing (what else do you do with crappy chips? ), only, it appears out of spite, the backwards compatibility goes with throwing the needed switch high, which is a little more difficult, as discussed. Even though there are two BSEL switches, only BSEL 1 should have to be moddified. Also note that by pin-modding, you are modifying the speed at which the FSB runs at, that is it. Simply equate it to overclocking the FSB as you would on a desktop. Multipliers, cache, and all other inherent properties of the CPU are not modified.

This said, I dont think your operating system should make a difference, however, drivers might. The first thing I'd guess is/was Intel Speed Step turned off (either manually or by some weird automatic functioning)? Also, try running a pre-OS program, such as memtest 86+, and see what it reads your CPU speed as ( www.memtest.org , grab the bootable iso, burn a CD, and boot to the CD; Your CPU frequency is in the upper lefthand corner). So far, your BIOS seems to recognize your CPU's frequency max as 2.16GHz, and it runs at full speed in the BIOS (as speed step CPUs do). If your CPU runs at 2.16GHz in memtest and speed step is enabled, then it would seem there is a driver, etc. blocking the CPU from functioning properly in windows. Therefore, the last thing I would suggest is try re-installing your intel chipset driver and seeing if that helps (I'm going out on a stretch, but hell, why not try it?).

Good luck with it and let me know how you make out. I'm glad to see someone had the cahones/time to try this awesome theory out .

i think that the high clock settings are different in the 2600

maybe use custom driver or maybe its a BIOS-Windows interaction?
I still think that reported speeds are not even close to be real time speeds.

lol, anyhow, i want to help you, i will try it myself with my 2250 next week, but if it doesnt work in windows xp its useless to most people, have you tried crystalcpuid real time clock reading? or maybe speedstep control (i know you have used the program, but what about the features in the menu?) i would like some pics on that too :P

I doubt intel would go to such lengths to squell the very small niche that is enthusiasts, it just wouldn't be worth it. That said, try uninstalling all your processor control related programs. Get rid of NHC, RMClock (remove the reg entries), Dell quickset, anything like that to get them out of the equation. Use CPU-Z and windows system properties to measure your CPU speed (its a good crossfire to double check accuracy). Use prime 95 or SuperPi to put the CPU underload and see if it cranks up the speed. Something in your windows environment is locking the CPU at 1000MHz, you just need to find what it is and disable it. The driver would be a good last ditch effort, but get rid of other programs first and see if they make a difference.

Something is wrong if it did 1M in 57 seconds. That's pretty slow. My 1.73 GHz Dothan did 1M in 42 seconds.

I got a question... what BIOS revision you using again? I'm betting your on A02, which is a "Support for more CPU's" update. BIOS may report 2.16ghz, but then knock it down once POSTing is complete after it verifies the CPUid.

Try running A01 or even A00 revisions to rule out the BIOS being the problem. I have an old A01 copy still, but never took a BIOS dump of A00 - sorry.

Yeap its in the hardware, there are diffrent configurations for the multiplier management in the datasheet. Try isolationg bsel O please. It could be a Bios issue too. Try flashing the A01 and the oldest one too, maybe then we will see different results. I am guessing too, that isolating the pin completetly is the problem, if where able to increase resistance without going to infinite, the multiplier management could work. AND PLEASE TRY CRYSTALCPUID its better for real time clock. post some pics too.