*Official Retail G0 Q6600 Overclocking Thread*

[Volkswagen]

how about this puppy? Wont be able to test it until the weekend...Got this and 2 x 2 GB DDR2 800mhz G.Skill sitting on my desk....

L737A822 10/17/07

[Darkstormz]

I´ve found a pretty good one. I got 5 CPU:s for test. And 4 of them had 1.30+ vid.

This is my baby
I primed last night 3GHz @ 1.11v - 8 hour stable.

Now I prime 3.6GHz @ 1.32v
Will update my screenshot in a few hours.

[lazy]

i got unlucky with my draw of the Q6600 getting one with 1.3vid

where do you guys find these low vid cpus?

[GFORCE100]

I´ve found a pretty good one. I got 5 CPU:s for test. And 4 of them had 1.30+ vid.

This is my baby
I primed last night 3GHz @ 1.11v - 8 hour stable.

Now I prime 3.6GHz @ 1.32v
Will update my screenshot in a few hours.

If it's a L737xxxx then it can do 3.6GHz around 1.27-1.288V, especially as yours has a low VID hence is one of the better yields. Give it a try, the chances are you can tweak it a little further. My L737B209 does 3.6GHz at around 1.27V and has a VID of 1.2250 so 120mv higher than yours.

I'm currently testing 3.825GHz at 1.37-1.38V with Core temps up to 70C.

[WizardofOz]

Great Going Darkstormz


Would you care to post the Full Batch so I can keep a log of the critical data from your tests.

I´ve found a pretty good one. I got 5 CPU:s for test. And 4 of them had 1.30+ vid.

This is my baby
I primed last night 3GHz @ 1.11v - 8 hour stable.

Now I prime 3.6GHz @ 1.32v
Will update my screenshot in a few hours.

[Darkstormz]

Great Going Darkstormz


Would you care to post the Full Batch so I can keep a log of the critical data from your tests.

It´s L739Bxxxx
I´ll get the box later tonight so i can check the full FPO/Batch for you.

I´ve updated my first post now
And also now prime @ 1.304v

[rabor]

Well,

first; not all L737's have low VIDs, so not all L737's will do 3.6GHz under 1.3V.

second; not all low VIDs mean high OCs..

I've had 8 L737's in the past couple of weeks, & only 4 of them did 3.6GHz @ (sub) 1.3V.. 2 of them were L737A & 2 L737B. What I found with all the low VID ones (exept 1) is that they run nice on or below 1.3V up untill about 3.65V, but than scale like cr#p needing 1.475-1.5V for 3.8-3.85GHz, & most of them hardly get to 4GHz with 1.6V.

However the one I have now (L737A679) has a VID of 1.3125 & needs 1.375V for 3.6GHz, but does 4GHZ easy just under 1.5V, so you just never know untill you test the critters.

Conclusion ; a SLACR with low VID is nice if you want to run at 3.6GHz, anything above 3.8GHz is a lottery.

[WizardofOz]

Rabor

Interesting anecdotal observation that you have from your 8 samples. I know from my E6600 experience that what you say is correct, that not all low VID will mean High Clocks and vice versa, but I think from what I have seen there is a high correlation between Low VID and High Clocks, it may not be 100% as evidenced by your sample, but from what I have seen so far already, a low VID is desirable if only to run it at 3.6GHz 24/7 which is what I am running mine at.

Mine is an L737B026 and I have an 8h Prime95 on 1.408v and will try for a 4.0+ SuperPi run, I won't be running Prime95 at 4GHz+ as the temps start to get high and I am on Air.
Your comments about not scaling linearly does hold true especially when you get above a certain threshold.

Would you like to document your 8 samples with their corresponding VID and Batch# and along with whatever data you have collected.

[lazy]

anyone here have a Q6600 G0 with a 1.3vid? how much vcore did you guys need to stable 3.6GHz?

[GFORCE100]

anyone here have a Q6600 G0 with a 1.3vid? how much vcore did you guys need to stable 3.6GHz?

Having a Q6600 with a 1.30V VID doesn't guarantee you it will run at 3.6GHz. I had two such chips with this VID and neither ran at 3.6GHz realiably enough to work in Prime95 hence be classified as stable.

It's down to luck really. Usually a lower VID means more chance to reach 3.6GHz or beyond at a lower voltage.

The 1.2875V VID Q6600 I had needed 1.43V to reach 3.6GHz but my 1.2250V VID Q6600 only needs 1.27V (one of the very good L737xxxx chips though).

For 3.6GHz it's really anything between 1.27V to 1.48V.

[sofarfrome]

FWIW...I have seen low vids that clock poorly and I have seen high vids that clock very well. You are putting too much emphasis on the vid. It will either clock or not, plain and simple. I am not so sure anymore that a particular batch will consistently clock well like the old L629Bxxx E6600 or Xeon3060 chips will do. It seems like any Q6600 G0 stepping will do 3.6GHz with reasonable vcore. The key is getting to 4.0GHz with less than 1.5vcore. If you can do that then you have a very good chip. Especially considering it is a quad.

[WizardofOz]

Lazy

Xevipiu on this forum has a Q6600 with a 1.300v VID

http://www.xtremesystems.org/forums/...157690&page=12

You can see his results

[lazy]

i guess i just got a crappy Q6600 then.. one of the newer batches L747's
i have given it up to 1.48 trying to get stable at 3.6 but it wont even prime for 5mins..
i guess 3.5 is the limit for this.. which even at this clock it requires 1.46vcore

is there any particular store that has been selling good batches?

btw thnx for the link Wizard

[spinejam]

anyone here have a Q6600 G0 with a 1.3vid? how much vcore did you guys need to stable 3.6GHz?

3.6 Ghz reboots my 'puter when running prime! still working on it! (1.3VID)

[PCTools]

Well, I have been pretty happy with this crappy Q6600 (B3). You just have to shove it voltage to make her work!

VID = 1.25V

[road-runner]

I have two 1.325 VIDs, niether will do 3.6 stable. One does 3.3 the other 3.5 stable on air. My 1.2375 VID will do 3.9 stable, but is on water also...

[KidNovak]

What do I need to get to 3.6Ghz without water cooling?

I tried by upping the vCore to 1.36v, but got errors from Prime95 after 10 minutes. Also, CPU was getting Hot, about 68C.

Novak

[timber715]

What do I need to get to 3.6Ghz without water cooling?

I tried by upping the vCore to 1.36v, but got errors from Prime95 after 10 minutes. Also, CPU was getting Hot, about 68C.

hot for a quad is around 80C, 1.36V is low for most cases(except probably a few and the new batches), push the voltage a bit higher... have mine on 1.375V with 3420Mhz temps not exceeding 70C. but with 3.6Ghz, i hit 86C with vcore of 1.450V... LOL

BTW ambient here is 30C

[Dede99]

Hi Novak and happy first post
I have the same Batch as you and I need on my 680i board without Vdroop about 1.375v for 3.6 everything below want stay primestable.
Good Luck

What do I need to get to 3.6Ghz without water cooling?

I tried by upping the vCore to 1.36v, but got errors from Prime95 after 10 minutes. Also, CPU was getting Hot, about 68C.

Novak

[Isriam]

hopefully i'll be posting mine today. i just have some questions.

i'm using prime95 25.5 for q6600 G0 however when i load it up and start blend test, it basically locks everything up. i can move the mouse, and hear the cpu under load, but i can't alt-tab or ctrl-alt-del. even at stock fsb it seems to hog all resources and i can't even stop it. pc doesn't lock up, just seems to suck all resources. any ideas?

[sesdave]

What do I need to get to 3.6Ghz without water cooling?

I tried by upping the vCore to 1.36v, but got errors from Prime95 after 10 minutes. Also, CPU was getting Hot, about 68C.

You need whatever vcore and other tweaks it takes too get your CPU stable at 3.6(if it can do 3.6) Just cos other are achieving 3.6 it on lower than 1.4v doesnt mean your CPU will do it. I need 1.425v with PLL 1.6,FSB=1.4,NB-1.4 on a P5K premium. for prime stable and thats with good ballistix 8500 memory . I would start with upping the vcore and loosening the memory to 5,5,5,15 - if you get it stable then try tightening the memory. Better cooling than a Zalamsn 9500 would also be a good idea -e.g. 120 extreme or ultima-90(nice small profile) will give better performance.Try MX-2 paste - doesnt need to cure and good as AS5.

[sesdave]

hopefully i'll be posting mine today. i just have some questions.

i'm using prime95 25.5 for q6600 G0 however when i load it up and start blend test, it basically locks everything up. i can move the mouse, and hear the cpu under load, but i can't alt-tab or ctrl-alt-del. even at stock fsb it seems to hog all resources and i can't even stop it. pc doesn't lock up, just seems to suck all resources. any ideas?

whats your detailed PC setup and operating system.

[Isriam]

win xp
evga 680i (sucks, doing rma due to crickets)
q6600 G0 @ 2.66ghz
geil 800mhz 4-4-4-12-2T

all voltages are auto.

my problem is i can't test using orthos, prime95, anything because it seems to fill the processor up and i can't even close them.

i'm thinking its the board, my old board ran orthos, prime95 everything perfectly.

(i'll post my batch later, i hit 4ghz this morning on 1.550v and 1.3v fsb)

[Isriam]

fixed it! was 2 versions of prime95 conflicting.

L737B Q6600

1.336v on auto
3600mhz
35-55c temps

still testing, i did hit 4ghz this morning on 1.550v

[PCTools]

Move over G0, there is a new B3 in town!

Here is my overclocks on the B3.









And the Raid Array: