*Official Retail G0 Q6600 Overclocking Thread*

[NaeKuh]

Batch: L737B062

Date: 10/31/07

BTW turns out the xeon is a G0 not a B3 and performs similar to the q6600.

I got a buddy who has a bunch of these xeons for sale but not 4 long.

Your chips are nice, however, YOU DONT HAVE ANY VALID LONG TERM STABILITY TEST!

So therefore YOU CANT say your stable. Not 1 screenie has you posting a prime longer then 1 hour. All your primes show less then 1 hour stress testing.

I have a 37B and you need AT LEAST 1.375Vcore on it to hit PRIME95 longer then 2 hours. After 2 hours the chip will fail at that voltage.

PLEASE post a longer prime, at least 3 hours if you cant do the 7hour standard.


Otherwise i might as well bust out the 4.1ghz which last only 10 min...

My chip and cooling has the headroom:



Those are load temps BTW... NOT IDLE... And its WCG BIONIC.
And XS is always recruiting great quadcores to crunch for our team.

[rbsmontemor]

Surely you don't need that much Vcore to get you to 3.8GHz???? You have a very low VID Q6600 so I would say you could do it at 1.30V or less.

Yes I need. And I think that's a very good chip...and run's @ 3800Mhz with less vcore than my friend's QX6850. 3600mhz @ 1.28v too

[]'s

[WizardofOz]

Would you guys mind posting the complete Batch# as in L737B123 when you submit some results. I am compiling a list of Batch# with their respective results (screen shots) and the VID.

I think the VID is telling a story and that is, the lower the VID the less VCore is required to hit a standard speed in this case 3.6GHz and thus lower temps.
Not all L737B or any other Batch# has the same performance even with the same Batch# they are all different but the one variable that can explain performance is VID. I have not taken into account MB, RAM, person OC experience and these factors could explain differences.

[SNiiPE_DoGG]

Would you guys mind posting the complete Batch# as in L737B123 when you submit some results. I am compiling a list of Batch# with their respective results (screen shots) and the VID.

I think the VID is telling a story and that is, the lower the VID the less VCore is required to hit a standard speed in this case 3.6GHz and thus lower temps.
Not all L737B or any other Batch# has the same performance even with the same Batch# they are all different but the one variable that can explain performance is VID. I have not taken into account MB, RAM, person OC experience and these factors could explain differences.

Don't forget cooling as well. A lot of people don't post their systems in enough detail

[GFORCE100]

But you said



Ive not seen any 3.8Ghz under 1.3v without chiller/subzero.

If you have, i would like to see some links.

Yes my mistake, by "do it". Somehow 3.6GHz locked into my head last night even though he's at 3.8GHz hence 1.392V seems much more reasonable, or very good actually.

It seems VID is a very good indicator of yield quality, but not the only one that determines the end result (overclockability etc.). I guess some silicon behaves different than others above a certain MHz speed hence why VID only helps up to level X (by using a low Vcore), but then the silicion chemical properties come into play.

I would think it helps having a low VID version chip. Couple this with a good yield and 4GHz at around 1.45V effective should be attainable, no? This could be done on air I reckon.

[jmmtn4aj]

Ive not seen any 3.8Ghz under 1.3v without chiller/subzero.

If you have, i would like to see some links.

Wait, so you've seen under 1.3 with chiller/subzero? Shouldn't under 1.3 NOT require that level of cooling seeing as lower volts = lower temps?

[Sepander]

Just got this L739B250 Date: 11/22/07

First tests:
4206MHz (9x467) SP1M
FSB 500MHz

Need more time to find best GTL's for high FSB.

[mad_skills]

Wait, so you've seen under 1.3 with chiller/subzero? Shouldn't under 1.3 NOT require that level of cooling seeing as lower volts = lower temps?

Yes. I think below 1.3V you can even on box cooler.

[Devious]

Wait, so you've seen under 1.3 with chiller/subzero? Shouldn't under 1.3 NOT require that level of cooling seeing as lower volts = lower temps?

I haven't no, I was just trying to make the point that I think extreme cooling would be required to hit 3.8Ghz with under 1.3v. With extreme cooling you tend to need less vcore to get the same clocks as water/air.

[WizardofOz]

I am currently testing my Q6600 (L737B026), here is the results at 3.63GHz (403 FSB)

This is at Load Prime95 about 13h



This is at Idle after the run

[sonofander]

nice job Wiz!

[Demo]

Woah nice clocks!

[GFORCE100]

Just got this L739B250 Date: 11/22/07

First tests:
4206MHz (9x467) SP1M
FSB 500MHz

Need more time to find best GTL's for high FSB.

Please also post your VID...

[~aoe~]

Please also post your VID...

It's on the 1st screenshot (CoreTemp). 1.3000v

[GFORCE100]

It's on the 1st screenshot (CoreTemp). 1.3000v

Ahh sorry, silly me. 1.744V seems about right then at the best of the crop could do it at around 1.6V given they do 3.6GHz at 1.27V-1.288V.

[Sepander]

4233MHz (9x470MHz) SP1M
Yeah, VID is too high (1.3000V).

[LookAS]

4233MHz (9x470MHz) SP1M
Yeah, VID is too high (1.3000V).

what is the power consumption ?

[road-runner]

L738B233 VID 1.325v, not that good of a chip compared to my L734B164 with a VID of 1.2375.

L738B233


L734B164

[lazy]

hi guys, just got my q6600 and was wonderin if anyone has seen this batch #
L746A515?
Btw its an oem so i do not have the pack date

[Devious]

Thats a new one, let us know how it goes.

[lazy]

Thats a new one, let us know how it goes.

Hope I got a good one then

[Sadhana]

This is mine, batch L723B284, vid 1.2875v, can't load coretemp under Vista x64 but I do check the vid in win xp since I have dual booted os.

[sesdave]

This is mine, vid 1.2875v, can't load coretemp under Vista x64 but I do check the vid in win xp since I have dual booted os.

I dont have any problems with core temp since i followed this http://www.xtremesystems.org/forums/...8&postcount=43 . Before I was getting a load loop due to driver signing. on Vista x64 as latest KBs removed ability to disable driver signing permantly.
That said I dont use coretemp much anymore as I have switched to using PC wizard 2008 http://www.cpuid.com/pcwizard.php - when you minimise it you get a small display of core temps,CPU usage a,GPU and CPU temps in top right of your screen.Try it - its neat and its accurate plus it has a load of other features like everest - doesnt show VID though!.

[lazy]

well seems like the new batch # i have sucks.. vid is 1.3
having trouble trying to prime even at 1.45v

[peterpiper]

Anyone heard of this batch L731A553? think it might be oem chip.. Pack date 30/09/2007