First attemt at oc'ing my q9550

setter

setter

setter

Caporegime
Joined
14 Dec 2005
Posts
28,067
Location
armoy, n. ireland
After updating to a more recent bios i decided to have a crack at clocking my q9550, 1.2 vid EO stepping chip. Went straight in at 3.4ghz (8.5x400) on stock volts, ran a quick small fft test, temps at 50-47-43-43, no errors. Ill probably try for a longer run when i get the time, but i dont intend to keep it at 3.4ghz for long:D

q955034ghz.jpg
 
They will go to 4.5 no trouble see my spec mine runs faultless at 3.8 all the time have had it at 4.5 but with the Zalman cooling it warms up a bit after a few hours so backed of to 3.8 and no trouble also VID at 1.175
 
Yep im hoping to see 4ghz from it, although i may do some rearranging of fans, the 2 pwm sharkoons i have on my TRUE are only rated at 400-1200, not the 400-2000 that they were described as:o, may switch them round with the 2 full blown 2000 rpm models im currently running as front intakes via a fan controller, i think with those 2 fans i could drop the temps a fair bit.
 
Jay Bee said:
They will go to 4.5 no trouble see my spec mine runs faultless at 3.8 all the time have had it at 4.5 but with the Zalman cooling it warms up a bit after a few hours so backed of to 3.8 and no trouble also VID at 1.175
Click to expand...
Did you have to up the pll, nb and fsb term much when pushing past 4ghz jay bee? on my previous q6600 @3.8ghz i needed 1.6 pll, 1.3 nb and 1.34 fsb term.
 
Last edited:
Good luck Setter! :)

I think you will be getting to grips with the wonderful world of GTL tweaking at some point soon, if you pull out a big overclock with all three of them on [auto] I will be in awe! :D

I've just swapped an E8400 for an E8400 (big upgrade!) and my previous GTL settings are not quite right on the newer chip, this is leading me to believe that every chip is a lot more individual once you push past 400MHz-FSB than I previously thought . . .

I'll be interested how you get on! :cool:

[edit] I think PLL is another individual CPU setting but vNB and vFSB are generic I think i.e you shouldn't have to change either of them if all the other kit are the same and you have just swapped CPU?

[double edit]I also believe that vFSB shouldn't be set higher than vNB as it reduces the effectiveness of the GTL design, I'm at 1.18vFSB with a 500MHz-FSB dually so can't see you needing much more than that unless you aiming above 500MHZ-FSB you crazy man! :p
 
Last edited:
Cheers Big Wayne, had a glance at your findings on GTL tweaking in the main p5q thread, interesting stuff. I left all gtl settings on auto when i had the q6600 at 3.8ghz, but with that chip i was running 9x422 fsb (not that high on a p45) so i just left them on auto.

EDIT: Yep ive heard that PLL will depend on the particular cpu, and that as far as NB and FSB, 1.4v is the max to go with theese on a 45nm cpu. Currently running.

PLL, 1.52
FSB, 1.12
NB, 1.12
 
Last edited:
So you at 400MHz-FSB now then?

Have you tried manually dialling down vTT to 1.12 and vNB to 1.14 and see if its still stable?

It would be on a dually @ 400Mhz-FSB so interested to see how much extra strain a 45nm quad gives the system?
 
Those were set to +0.02v above auto, i cant see me going near 1.4 tbh. I upgraded the bios from ver 0803 to 1406 so im not sure if this p5q overvolting on auto or lowest manual issue is still the case, but i think its best to get away from auto settings anyway. I ran p95 with those settings for half an hour or so without error. Will give it a longer run at some point as itll be a good stable base to work from.:)
 
Nothing more so far, been a bit busy with work and such, im gonna have a good run at it over the weekend.
 
You must log in or register to reply here.
Back
Top Bottom