Yonah has been pinmodded!

Ok, as I have said in my Merom pinmodding thread, I have successfully made my T2250 running at 1.73Ghz into a T2600 running at 2.16Ghz. Almost...

I have posted my CPU-Z page here, which shows the higher FSB speed...



...however, for some reason windows wont budge off of the 6X multiplier, so really I am stuck at 1Ghz right now. I need some assistance trying to figure out the culprit here. I have also reposted the picture of the processor internals and pinout to see if I missed something.

Basically I have isolated the BSel1 pin with a small piece of ATA100 wire shielding. I just slightly opened up the plastic part of the socket to accomodate the larger circumference. When I boot into the BIOS, it says:

Current: 2.16
Min: 1.00
Max: 2.16

As soon as I hit windows, I am forced into the 6x multiplier and cant force it higher by any means. So I need your help guys. Help me figure out this wierdness please!

Well, I suppose that is a possibility... However I have doubts that its in the driver, at least I would hope its not. I am so dang close, lol. I am really hoping that maybe there is just another pin that needs to be modified, if anything at all... Somebody out there might know the answer (or at least a good guess), so if ya have any thoughts, throw them out there!

Here is the link to the Datasheet for the Core Duo/Core Solo

ftp://download.intel.com/design/mobi...s/30922105.pdf

Just tried that, and got the same results... Seems like something is hardware locking it to that speed... I wonder if another pin needs to be insulated or shorted. The BSel pins are setup correctly now. There are two other pins labeled BClk0 and BClk1... wonder if they have anything to do with it. This is sure frustrating at this point... I have read that datasheet cover to cover 3 times now, and still can't make sense of why this method is not functioning 100%... Maybe it will come to me eventually, lol.

"sigh"...

I have attached a picture of my BIOS screen, just for reference. I really want to believe that there is just something simple I am missing, but I know better. I have yanked the cpu out and put it back in a few times, but that did not help. Guess I am back to the drawing board.

Just tried it also, with similar results. I just can't put 2 and 2 together here. The BIOS shows the higher clock speed... when I set I8KFanGUI to the "Ask BIOS about speed" setting, it shows the 2.16Ghz value. However Sandra CPU Benchmarks show otherwise... I am performing less than half speed of a T2600 right now. Fortunately its easy enough to revert back to the regular speed with my method, but I really want to make this work!

Now here is one thing I have been trying to decipher. Other forums have shown that by insulating a BSel pin, you effectively make it go HIGH, whereas shorting one to ground gives you LOW. Is it possible that maybe it really needs some kind of actual signal to operate correctly? On a 667Mhz FSB Yonah, the pin still exists on the cpu. Is it just disconnected inside the cpu on these models? Or is there some other necessary voltage that is required to be fed to this pin to make it work properly? I certainly dont want to go shorting pins out on my machine without a clue of what effect it may have... I really need somebody with hardware experience to ring in here and help me out. Anybody?

BTW, thanks to everybody so far for the suggestions. Everything is appreciated right now.

Thanks to everyone for the replies. Yes I have set windows to always on, and I use NHC (tried several, but this one usually works for me...) to force it to high perfomance mode anyway. Doesn't seem like any software programs can get me to the point of success.

As for using the Dothan technique, this is nothing like that. With Dothan you used a jumper wire to ground out a pin that was already HIGH, which in turn changes it to LOW. I had to go the other way around, from LOW to HIGH. If you look into Xeon pin modding a little bit you can see that people have done basically what I have done to get the FSB a serious boost. With a Xeon, if you insulate the pin as I have done, it effectively throws the socket hole to HIGH. Now that part of the mod has worked well, because you can see I have successfully hit the 166FSB/667MHz without issue.

The pinout is in the picture attached to my first post if you want to take a look at it. There is only one group of BSel pins... BSel0, BSel1, and BSel2. BSel1 is the only pin that needs to change status to make the overclock go through. However I am thinking that something else needs to change via the pinout to successfully open the multipliers back up. This is where I cannot figure it out. As I have said also, the BIOS shows the correct speed, so why doesnt it transfer through to windows...? Intel has made some changes, that is apparant. I believe this is still possible though. I am halfway there now, and I dont want to throw in the towel just yet!