*Official Retail G0 Q6600 Overclocking Thread*

[Jor3lBR]

Ok, so this is my Everset information:
CPU Properties
CPU Type QuadCore Intel Core 2 Quad Q6600
CPU Alias Kentsfield
Engineering Sample No
CPUID CPU Name Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz
CPUID Revision 000006FBh
Core Voltage 1.275 V

CPU Speed
CPU Clock 4009.2 MHz (original: 2.4 MHz, overclock: 67%)
CPU Multiplier 9.0x
CPU FSB 445.5 MHz (original: 266 MHz, overclock: 67%)
Memory Bus 445.5 MHz
DRAM:FSB Ratio 1:1

I've run prime and everst, both for hours, but I'm having problems with some games and apps being stable (think it is a driver issue). My phase will hold -35c under load all night under prime or everst, so I think that it is stable.

Yes thats your VID

Lol create a signature we didn't know you had phase and yes with phase or a low vid cpu on water you can get Prime95 stable in that speed.

[wyemarn]

My L804 with 1.325V VID is doing 8x400=3200Mhz at 1.30V stable. I can push higher but the chip requires more voltage. At 3.2 my load temps are around 56-59C on a silent system. What do you guys think of my chip? Btw, should I enable C1E?

[jcniest5]

Just realized that now my local MicroCenter is completely out of the earlier G0 Q6600. Now, all their Q6600s are the light blue box.

[Shalamay]

I know why I can't get it stable, my ram took a dump. Memtest at 800mhz had 80+ errors in 30 seconds or so. Glad Patriot has a nice RMA service.

[Hans Gruber]

I've lowered my multiplier to 8x and increased the FSB to 450. Which means my system is still running at 3600Mhz. My ram is running at a 1:1 ratio at 900Mhz with 4-4-4-12 timings. My system is very stable at 1.42v

However, my Super Pi 32M calculations are taking 1 minute longer than running 9x400. 14:20 before and with a FSB of 450 my 32M time takes 15:22 minutes. My 1M calculations are slower as well.

My 3DMark06 scores are the same as before changing the FSB.

Shouldn't I gain a little performance considering my FSB is 450 instead of 400?

[Mtrx]

I've seen this so many times that hurts!
The old mentality that OCers had from AMD era does not apply for Intel. If you have a mid to good quality Mobo with a fair vdroop (nothing so big of course) that's within Intel specs of a healthy.

So VDROOP is GOOD for Intel cpu and stability (within a certain margin).

http://www.thetechrepository.com/showthread.php?t=126

Thanks for the link, it was interesting and as overclocking does the vdroop mod will affect the healthy of the quad although less vcore is needed.

Here is a pick from x-bit:

“I have to say that P5E is not free from a classical drawback of all Asus mainboards: significant processor Vcore drop under Vdroop load. We performed some tests with a quad-core CPU running at 3.6GHz and revealed that the processor voltage drops by 0.048V. Luckily, there is useful advice online on how to modify the mainboard in a quick way to eliminate this unpleasant effect that can hinder successful overclocking. Asus, unfortunately, didn’t offer any solution of their own to this problem. The BIOS of P5E doesn’t have the CPU Voltage Damper introduced in the P5K series, and the available Loadline Calibration option didn’t prove efficient enough. By the way, we noticed the processor voltage got simply reduced. For example, our mainboard sample sent 0.058V less to the processor than was set in the BIOS.”

And here is another from the last review at anandtech:

“Loadline Calibration - [Enabled, Disabled] Setting Enabled will effectively remove any negative feedback in the CPU supply voltage regulation circuit. This means that there will be little to no difference between the CPU supply voltage at idle and full-load. This does not mean however that there will be no offset seen between the idle voltage and the CPU Voltage VID as selected in BIOS. Disabled follows the Intel specification and allows the CPU supply voltage to droop under load as properly designed. We recommend you leave this Disabled and increase CPU Voltage as needed to satisfy the minimum operating voltage needed under load.”

[conzymaher]

Do the new Q6600s still clock as well as the good G0 chips?

With the Q9450 costing 100 euro more than the Q6600 and only having a 8X multi the Q6600 seems like a more attractive chip if it will still hit 3.6-3.8ghz

So, Q6600 or Q9450? :p

[okorn]

Q6600

#Batch: 740A740
Vid: 1.2000V

[Leeghoofd]

Is that one of the newer ligher blue boxes Okorn ? nice clocker you got there

[Emerica]

Q6600

#Batch: 740A740
Vid: 1.2000V

Nice chip!

[zlojack]

I know this is the Q6600 thread, but since there's no thread for it:

I just picked up a Q6700 this afternoon. I was at the shop as they got new Q6600's in, but they had one Q6700 on the shelf and the price drop has come into effect. $300 for this chip! 10x multi and what should be a higher binned Q6600 so I figured I'd pay the extra $50 and grab it.

L727B332 - I'll post results later.

[ZenEffect]

Do the new Q6600s still clock as well as the good G0 chips?

With the Q9450 costing 100 euro more than the Q6600 and only having a 8X multi the Q6600 seems like a more attractive chip if it will still hit 3.6-3.8ghz

So, Q6600 or Q9450? :p

q6600 ftw!

im sure intel is going to do something to improve the batches of the q9450. they are getting a lot of p/o'd oc'ers that thought they were upgrading.

[drnick79]

I just received my q6600- I was afraid I was going to get one from the L8xx batch, and I did , but the price (online only-not honored in store) of $180 was too tempting.

The VID is 1.325 and the batch is L804. Granted I've only tried it on my 780i mobo (which I consider to be a flaky overclocker), but I'm not sure it will even hit 3.6 ghz. I've got it OCCT stable for a half hour so far at 3.15ghz (450x7), but that required 1.6125v vcore (bios)/1.56v (everest). I need to set the vcore to around 1.675v (bios)/1.62ish (everest) to reliably get into windows at 3.6ghz (either 400x9 or 450x8), but at this speed OCCT crashes within a minute.

I've tried the memory (2x2gb gskill 1066) both linked and unlinked (and usually underclocked), and used a range of GTL REF values, and it doesn't seem to make any difference. On this board, I don't think it is capable of 3.6ghz unless I'm willing to go over 1.7v vcore in the bios !

I'm considering returning it and seeing if I can find any locally from an L7xx batch.

[wyemarn]

I just received my q6600- I was afraid I was going to get one from the L8xx batch, and I did , but the price (online only-not honored in store) of $180 was too tempting.

The VID is 1.325 and the batch is L804. Granted I've only tried it on my 780i mobo (which I consider to be a flaky overclocker), but I'm not sure it will even hit 3.6 ghz. I've got it OCCT stable for a half hour so far at 3.15ghz (450x7), but that required 1.6125v vcore (bios)/1.56v (everest). I need to set the vcore to around 1.675v (bios)/1.62ish (everest) to reliably get into windows at 3.6ghz (either 400x9 or 450x8), but at this speed OCCT crashes within a minute.

I've tried the memory (2x2gb gskill 1066) both linked and unlinked (and usually underclocked), and used a range of GTL REF values, and it doesn't seem to make any difference. On this board, I don't think it is capable of 3.6ghz unless I'm willing to go over 1.7v vcore in the bios !

I'm considering returning it and seeing if I can find any locally from an L7xx batch.

Yup, the latest batch are bad clockers. Mine also L80x with VID 1.325V. I can do 3.2Ghz at 1.3V stable. Anything higher than that requires much more voltage. 3.4Ghz requires at least 1.375V. Tried 1.45V at 3.6Ghz and Windows will crash after priming a few minutes.

[Jor3lBR]

I just received my q6600- I was afraid I was going to get one from the L8xx batch, and I did , but the price (online only-not honored in store) of $180 was too tempting.

The VID is 1.325 and the batch is L804. Granted I've only tried it on my 780i mobo (which I consider to be a flaky overclocker), but I'm not sure it will even hit 3.6 ghz. I've got it OCCT stable for a half hour so far at 3.15ghz (450x7), but that required 1.6125v vcore (bios)/1.56v (everest). I need to set the vcore to around 1.675v (bios)/1.62ish (everest) to reliably get into windows at 3.6ghz (either 400x9 or 450x8), but at this speed OCCT crashes within a minute.

I've tried the memory (2x2gb gskill 1066) both linked and unlinked (and usually underclocked), and used a range of GTL REF values, and it doesn't seem to make any difference. On this board, I don't think it is capable of 3.6ghz unless I'm willing to go over 1.7v vcore in the bios !

I'm considering returning it and seeing if I can find any locally from an L7xx batch.

Wow sorry but thats a rubbish chip! Why don't search for a good batch or pay a little more for a good one?!

I have the 1.21vid L745A793 it is Prime95 blend 8hrs stable with 1.26v (bios) and 1.28v CPUZ.

Also it hits 4005Mhz with 1.39v (bios) and 1.42v cpuz (wPrime, SuperPi, 3dMark0103,05,06 and Crysis stable).

So just do some research before you buy, I saw somewhere in this thread for a eBay add with a 1.20vid cpu.

[Jor3lBR]

Heres 24/7 settings on my batch:

http://i28.tinypic.com/28lcjgk.png


* temp readings: closed case, dual rad w/ 7v fans, 24oC room.

[drnick79]

It's definitely not what I would consider a winner of a chip. I thought I would take a chance that all of the L8xx batch might not be such overclocking duds. I just didn't think I would get such a voltage hog. RMA time I suppose (luckily I have a 15 day return option at Fry's).

Is the VID printed anywhere on the retail box? I know how to find it once in windows with programs like Coretemp- just wondering if there is a way to find the diamond in the rough without having to install the cpu...

[Jor3lBR]

It's definitely not what I would consider a winner of a chip. I thought I would take a chance that all of the L8xx batch might not be such overclocking duds. I just didn't think I would get such a voltage hog. RMA time I suppose (luckily I have a 15 day return option at Fry's).

Is the VID printed anywhere on the retail box? I know how to find it once in windows with programs like Coretemp- just wondering if there is a way to find the diamond in the rough without having to install the cpu...

No it's not printed anywhere, the only way to get a good chip it's to google batches #.

Like I did, so you already have a winner batch on my last post, you can also see some VID's here:
http://www.forumdeluxx.de/forum/showthread.php?t=438595

And here: http://www.xtremesystems.org/forums/...d.php?t=178263

[drnick79]

Hey Jor3lBR,

Thanks for the info! I'll see if I can find any from the desirable batches at the local brick and mortar purveyors!

On a semi-related topic- isn't Intel cutting CPU prices soon? (tomorrow?)

[wyemarn]

Hey Jor3lBR,

Thanks for the info! I'll see if I can find any from the desirable batches at the local brick and mortar purveyors!

On a semi-related topic- isn't Intel cutting CPU prices soon? (tomorrow?)

Good luck on your search. Too bad I can't exchange mine. Over here in Japan, some retailers already cut their price. I got my q6600 two weeks ago and now its around $30 cheaper.

[C'DaleRider]

As for the price cut, Intel has one "scheduled" for today, the 20th. The Q6600 is supposed to drop by $43 in lot sizes....don't know what it'll do individually. Time will tell, I suppose.


I do own a Q6600 myself.....lot# L745....


I am somewhat of a very lazy OC'er......the cpu is runing on an Asus X38 board, a Maximus Formula.

Leaving all setting at auto except for forcing my PCI-e speed to run at 100MHz, cpu voltage set to 1.35V, NB voltage at 1.5V, and RAM voltage at 2.0V, I've been able to run my cpu at 3.6GHz. That's stable under hour after hour of OCCT.

I have had it boot at 3.9GHz, 1.51V, all other setting again left in auto, but fails OCCT after just a couple of minutes.

Not really sure which or what setting to try tweaking, but 3.6 @ 1.35V seems to me to be a decent OC. My temps, under watercooling, really never seem to hit more than 52C on the "hotter" cores of the two pairs of cores. (The other "core" never gets near 50C running OCCT at 3.6GHz.)

Everest gives my VID at 1.238V.

Guess I'll just let sleeping dogs lie and be very satisfied that it does what it does.

[ChaosAD]

I would like to ask you something. My previous Q6600 had a vid of 1.325. Used to run 8x425@1.41v while the absolut minimun Vcore to boot in Vistax64 was 1.34375 in Bios, 1.328 in Cpu-z. After some search to find a new chip i found a friend with absolut no knowledge of pc's and o/c ofc and i traded my old chip for this new one. When i asked him to check his VID before trading he read CoreTemp and reported 1.1625.

So long story short we made the trade, i installed my old cpu in his mobo and his cpu in mine.After i logged in Vista my core temp/everest report a vid of 1.3 (not 1.325 from old cpu, or 1.1625 from the new one), while his core temp report the old vid of 1.1625. Is this normal?Is this bug?If i reinstall Vista will this fix this error?

Now i test my new Q6600 8x425@1.304 in cpu and passed 30 min of prime

[bluescreen]

I just received my q6600- I was afraid I was going to get one from the L8xx batch, and I did , but the price (online only-not honored in store) of $180 was too tempting.

The VID is 1.325 and the batch is L804. Granted I've only tried it on my 780i mobo (which I consider to be a flaky overclocker), but I'm not sure it will even hit 3.6 ghz. I've got it OCCT stable for a half hour so far at 3.15ghz (450x7), but that required 1.6125v vcore (bios)/1.56v (everest). I need to set the vcore to around 1.675v (bios)/1.62ish (everest) to reliably get into windows at 3.6ghz (either 400x9 or 450x8), but at this speed OCCT crashes within a minute.

I've tried the memory (2x2gb gskill 1066) both linked and unlinked (and usually underclocked), and used a range of GTL REF values, and it doesn't seem to make any difference. On this board, I don't think it is capable of 3.6ghz unless I'm willing to go over 1.7v vcore in the bios !

I'm considering returning it and seeing if I can find any locally from an L7xx batch.

The L804 I got from frys a few days ago had a 1.262 VID reported by EVEREST. I have it up to 3465Mhz completly stable on my striker extreme going to try for 3.6+ when I get some time.

[The0men]

I just received my q6600- I was afraid I was going to get one from the L8xx batch, and I did , but the price (online only-not honored in store) of $180 was too tempting.

The VID is 1.325 and the batch is L804. Granted I've only tried it on my 780i mobo (which I consider to be a flaky overclocker), but I'm not sure it will even hit 3.6 ghz. I've got it OCCT stable for a half hour so far at 3.15ghz (450x7), but that required 1.6125v vcore (bios)/1.56v (everest). I need to set the vcore to around 1.675v (bios)/1.62ish (everest) to reliably get into windows at 3.6ghz (either 400x9 or 450x8), but at this speed OCCT crashes within a minute.

I've tried the memory (2x2gb gskill 1066) both linked and unlinked (and usually underclocked), and used a range of GTL REF values, and it doesn't seem to make any difference. On this board, I don't think it is capable of 3.6ghz unless I'm willing to go over 1.7v vcore in the bios !

I'm considering returning it and seeing if I can find any locally from an L7xx batch.

Don't give up on the chip just yet.... Mine was the same, in the same board, untill I give like 1.8+ volts to the CPU PLL, try that, then see how it does. also have 1.4v VTT and 1.5v on the 1.2v HT. Also I have 1.64v set in bios for NB... try those and see how it does. Only using +30mv on the GTL ref's

[Gen]

Don't give up on the chip just yet.... Mine was the same, in the same board, untill I give like 1.8+ volts to the CPU PLL, try that, then see how it does. also have 1.4v VTT and 1.5v on the 1.2v HT. Also I have 1.64v set in bios for NB... try those and see how it does. Only using +30mv on the GTL ref's

you suggest 1.8v pll? i understand that's your opinion but most chips doing 1.8vpll for 24/7 will not last long. try every GTL reference before you increase VTT and PLL for your overclock...