2.6c / IS7-E Max Stable OC - Any suggestions?

[batmang]

I battled with my cpu all weekend over the holiday break, and the highest rock solid overclock i got it at was 3.033ghz. I am in no way unsatisfied, because before the highest rock solid overclock i could get was settling for a 2.8ghz 1:1 oc. Currently its at 233fsb, 5:4 ratio, 2-3-3-7 timings, 1.65vcore in bios, actual ranges from 1.584v - 1.6v, vagp 1.65v, vdimm 2.8v, bios cacheable, video cacheable both disabled. GAT in order, auto/normal/auto/disable/disable. Using bios version 17 with my IS7-E. I ran 2 instances of prime95 for about 2 hours with priority on 10, so thats enough for me to declare my system stable. The thing that is bothering me is that i could get all the way to 3.1ghz on stock voltage. And in CPU-Z it was showing 1.504v, and when underload it would range from 1.43v to 1.48v, but failed prime95. I had to set it to 1.65v in bios to find a rock solid stable oc, at 3.033ghz / 233fsb. Unless im doing something wrong, i think ive done everything possible to find the sweet spot, which i think i found with this oc. I had tried all sorts of different timings, fsb's, and this seems to be my chips max rock solid overclock. I'm happy with it, but i hate knowing my bios vcore is 1,65v, but in reality i know the is7-e undervolts, so i should be good. I've had the cpu/mobo since April 02, have been running it on various vcore since then, but all have been well over stock. So as for the cpu dying on me, i dont think its gonna die anything soon. All i need now is a better video card.... mx's really suck. Oh yeah, if any of you read my post about the radeon 9600np i had, well... it had something wrong with it. I think im gonna save up for an nvidia card, ive always liked them better. If any of you have any suggestions about how i can increase me oc, id really appreciate it, thanks in advance.

 

[MDE]

I'm inclined to say that your power supply can't keep up with the load (although it should definately be able to). Check the voltages with a multimeter and see what you find. I believe the P4 runs off the 12V rail, so that's the first suspect IMO.

 

[batmang]

i suspected the same thing, and i checked the voltages in the bios and they are:

psu voltages
-------------------------------------------
agp vddq - 1.64v
ddr vtt - 1.39v
atx +3.3v - 3.34v
atx +5v - 5.13v
atx+12v - 11.97v - 12.03v
battery (+3v) - 3.31v
standby (+5v) - 4.89v

I dont happen to have a multimeter, but ive checked the voltages in windows with the abit utility, and in the bios. Seems to have enough juice flowing.
I also ran memtest86 overnite with no error or failure, and i even tried 3:2 ratio to knockout all possible memory problems with the oc, still no errors or failures.

 

[myocardia]

It sounds as if your "glass ceiling" is just a bit lower than some other people's 2.6C's.

 

[batmang]

yeah, figured that, but i aint complaining, still a 400mhz oc.

 

[SpeedFreak03]

My glass ceiling on my 2.6C is 3.12GHz (240MHz) FSB, but on stock voltage. Thats a nice OC you have though! I had 3.25 but it was unstable in P95 after a while. Just dont go over 1.6 actual (CPU-Z reported voltage is actual), and you should be okay. Congrats on the stable OC!

-Josh

 

[Haden]

I've got my 2.6C today, goes 3.3 (255FSB) w/o any voltage changes using stock cooler (thought I'll replace it, way too noisy).
Seems I'm pretty lucky with it.

 

[batmang]

indeed you are, nice oc

 

[Shimmishim]

goes to show once again that overclocking is a lot of pure luck!

that's why i've given up my search for the magic 3.6 ghz 2.4C



time for 3.0's baby!