Temps for dif CPUs

[Dawgdoc]

I have 2 systems.

1 is a Q6600, P5K vanilla, Tuniq Tower.

The other is an X3220, P5K Premium, TRUE 120.

The temps with the TRUE 120/X3220 are far far far cooler than the TT/Q6600 when comparing identical BIOS settings with Vcore, NBV, Vdimm, etc.....

Is it a fair assumption that if I switch the Q6600 chip to the X3220 rig and keep everything else the same, that I would expect the same temps with the same settings as when I had the X3220 in there? Or.....could the temps be dif bc of different heat output from the 2 dif CPUs?

Also, my Q6600 takes PRESCIOUS little voltage to run stable. FAR less than my X3220. To me, that means that it likely will be a better overclocker. Is this a fair assumption?

Until I change the cooling on the hotter rig, I am going to switch the Q6600 to the better cooled rig and see how far I can push it. As of right now the limiting factor have been the temps. Anything I have thrown at it, it has been stable for the first few mins, but then the temps spike so much I stop the testing.

Think the voltages in the same range will work when the Q6600 is switched into a new rig, and do you think that the temps will be roughly the same as the same setting for when the X3220 was in there?

Thanks!

[bing]

Imo, there is no easy explanation on those difference cause there are so many factors could affect that temp, unless you try both CPUs at the same mobo & HSF.

Btw, how much difference ?

[Dawgdoc]

How much dif for temp or voltage?

At fixed 450 X 8, Vcore 1.5, Vdimm 2.2, NBV 1.55:

X3220 TRUE 120
Idle 43/42/44/42
Load 68/68/69/69

Q6600 Tuniq Tower
Idle 46/46/43/43
Load 77/77/74/74

Thats a big big dif in load temps IMHO. Load was applied via P95 small fft, and only kept for 3 mins or so. I didnt want to push the Q6600 more than that 77 for any comparisons beyond that.

As far as the voltage dif, the Q6600 will boot to windows as low as 1.3V
I have not done in depth stability testing bc the temps spike to 65+ on that machine no matter what the voltage. I think I got a bum TT. It cools ok, and I can CO the Q6600 to 3.5 Ghz stable (not P95 sm fft stable, but stable enough that its been folding 36 hours straight with no problems) and I think it can do significantly more with better cooling.

The whole "its roughly the same temp no matter what the Vcore" means to me my HS aint doing enough. I lapped it and the CPU, and have remounted several times with dif fan directions, etc......and this is as good as that HS is getting for me. Ill be putting another TRUE 120 in there within the next week to check and see, but in the meaning I just wanted to see how far I can get that Q6600 with good cooling.

Thanks for the info! Keep it coming :)

[bing]

Yeah, that is quite big difference.

Could you try a quick uvoc-ing on both your CPUs to see what happened to that difference on this low heat setting ?

[Dawgdoc]

Yeah, that is quite big difference.

Could you try a quick uvoc-ing on both your CPUs to see what happened to that difference on this low heat setting ?

I would be happy to, but these 2 beautiful rigs I put together (and they ARE sweet!!!) are my first 2 OC attempts (3.6Ghz and 3.5Ghz both stable FTW!!!! but i want more....) so....uvoc I assume is undervolt OCing, but I really dont know how to do it.

[Dawgdoc]

And bing...if you have a moment could you address some of the other questions above? Im so very glad to have the ear of someone with your experience.

In the orig post.

[bing]

Is it a fair assumption that if I switch the Q6600 chip to the X3220 rig and keep everything else the same, that I would expect the same temps with the same settings as when I had the X3220 in there? Or.....could the temps be dif bc of different heat output from the 2 dif CPUs?

I believe it is not that easy to make that assumption, you should consider other factors like different CPU quality factor, yes, this is the matter of luck and nothing you can do about it and also different contact quality of the HSF base into the IHS "and etc". :)

That is why as I suggested, the easiest method is to try both CPUs at the same mobo and HSF and try to mount it in a "very" consistent way for both. Yes, it will be quite a hassle. :(

Again, this will not a 100% fool proof either, cause there is a possibility that both CPU don't have the same IHS flatness. :confused:

See ? Its quite pita isn't it ? ;)

Regarding flatness checking on IHS & HSF's base, I suggest you click my sig on "IHS + Ketchup are Tasty !" and use that method if you had a chance.


Also, my Q6600 takes PRESCIOUS little voltage to run stable. FAR less than my X3220. To me, that means that it likely will be a better overclocker. Is this a fair assumption?


Currently imo there is no definitive "proof" yet that cpu with lower voltage is a better overclocker.

Also, I assume you were reading those voltage from "software" right ? To have a fair comparison, a true voltage reading taken from DMM at both rig is a must.

Btw, what are the VIDs for both CPUs ?

[bing]

I would be happy to, but these 2 beautiful rigs I put together (and they ARE sweet!!!) are my first 2 OC attempts (3.6Ghz and 3.5Ghz both stable FTW!!!! but i want more....) so....uvoc I assume is undervolt OCing, but I really dont know how to do it.

Just to save time, I assume both at your Asus P5K mobos at the "lowest" VCore setting in the BIOS is 1.1000 Volt right ?

Make sure you disable C1E & SpeedStep on both and set the VCore to 1.1000 Volt "AND" set everything on "STOCK SPEED".

Also try to set both fan speed at max while you're doing that.

Observe both CPU temps on idle and on load and report back here, it should be easy and I think it needs no more than 15 minutes.

Of course to have a real UVOC, it will need quite some time like this HERE (http://www.ocforums.com/showthread.php?t=520284). :D

[Dawgdoc]

Vid is the same: 1.275v

I think that both the CPUs are the same flatness, since I lapped both. Sure, there could be slight differences, but after 45 mins of sanding from grit 400 to grit 1000 on a flat piece of glass...I think that are prob pretty dang close :)

Im off to give it a go...want to do all this and still OC the Q6600 this evening.

Thanks for the advice so far Bing.

Bbak in a lil while!