Q6600 Results so far

[Methious]

Well naturally after a little burn in I had to unleash it a little, so after extensive updating (still have more to do) I went for some over clocking. I still have some fine tuning to do and some testing to do but I ran OCCT (Because it was late, 2 am) and got 3 runs at 30 minutes stable @3.0 GHZ.

I take it that 4 Gig kits take a little more NB Voltage on some boards? I've gotta go back in and shave voltages some but the highest temp I saw was 38C max and that was just for a second or two, hovered at:

#0 37C (full Load 30 minutes) Idle is #0 32C
#1 37C #1 32C
#2 30C #2 22C (is it normal for the cores to vary by 10C on a Quad?)
#3 32C #3 24C

Voltages in line with Grayskys' OC guide at the top of the forum (Thanks Graysky). Running a notch or two higher, suspect I can lower them as adding NB voltage stabilized it.

The Thermal Compound is thinly spread, completely covering, and in full contact with the WB. I used a torque driver from a local shop to evenly torque the screws down. (Those CPU lids that hold down the CPU are a trip first time around,

Is that pretty decent for a Quad Q6600 under water? I redid the whole loop (Thanks for the advice Bio!) to shorten the hoses, took the NB Water block out of the loop because this looks to have better NB heat sink (might add a small fan).

Here's the preliminary results from the top of my head:

At Stock 2.4 I ran AquaMark formerly best OC CPU and GPU on the M3A I got 122FPS Stock on the Q6600 I got 133 no OC GPU or CPU
OC X2 6400 and GPU on LightsMark 2005 I got 400FPS, OC 6600 @ 3.0 gpu @ 702/980 I got 430FPS

I'll get some screens when my eyes get open.

[Apoptosis]

sounds about right... Never ever heard of anyone using a torque driver to tighten the stuff up evenly! Wouldn't hurt anything by doing that, but remember they all have an IHS on top (integrated heat spreader), so you don't have to be exact and can't crack the core... You have to do some serious stuff to ruin one - http://www.legitreviews.com/article/402/1/



I wonder what my temperatures were on that one ;)

[ibleet]

Interesting idea using a torque wrench to make sure that the pressure was applied evenly.

The Q6600 that I recently worked on went all the way to 3.2Ghz on stock voltage, so those generally do 3Ghz without breaking a sweat. They undervolt like mad, like 2.4Ghz @ 1.1v, so you know there is insane overclocking headroom in that chip.

We met our match at 3.6Ghz@1.4v but only because the temps were too high on air. You'll have it over 4Ghz in no time under water.

If I remember correctly, there was quite a difference in the core temps, but I think the difference was maybe 6-7c at idle. I'll check on that for you later today.

One thing I do want to do is to check the chip and the HS for flatness using a razor blade, and lap them if needed. I may only lap the HS as lapping the cpu voids the warranty, but who knows.

You're looking fine so far, so expect a huge OC with that great chip.

[martini161]

thats about how high i could run my q6600 stable before i umm, you know, lets say damaged it you should be able to get higher on a p35 board though what are your temps like?

[Methious]

Maxed out for an hour the highest temp was 38C and quickly reached 38 and never creeped up.

3.0 is running pretty good, I need to run some Orthos/Prime 95 cause I don't trust OCCT fully. There are some Bios options strange to me, guess that's expected since I was on AMD for so long.

Been recording some benches as I go, best I hit on Aquamark was 166, up from 122 with the x2 6400 running @ 3.52 in both cases the GPU was running 699/983.

I kept every thing the same hardware wise except the Mobo and CPU and as soon as I get it all correlated and gathered up in one spot I'll post up.

I don't think I'm going any higher than 3.0 for now, stuff needs to break in then I'm shooting for 3.2 and calling it good there. I've never been a drive it until it pops person. At 2.4 it was out doing the 6400, at 3.2 it'll be smoking it an that's enough for me.

It amazes me that I can change ram speed independent of CPU speed. I've gotten it to 3.0 @ 4-4-4-15 ram running just above 800. Have to play with that a little but I'm already running 2.15v on ram to hold the 4-4-4-15. 2.2v EVP on the ram and I don't like to exceed specs.

[ibleet]

It amazes me that I can change ram speed independent of CPU speed. I've gotten it to 3.0 @ 4-4-4-15 ram running just above 800. Have to play with that a little but I'm already running 2.15v on ram to hold the 4-4-4-15. 2.2v EVP on the ram and I don't like to exceed specs.

Yep, isn't Intel just great?

My OCZ reaper ram holds 4-4-4-12 at 1.89v. Thats a tad tighter than advertised and a lower voltage...and yes it passes memtest...guess I just got lucky. I haven't tried to overclock it yet.

[martini161]

I would shoot higher than 3.2 the average ive seen around the net on a quad g0 water cooled is like 3.6 so i would go for 3.4 to just get an even 1 ghz increase

[Methious]

I'll probably try for 3.6 stable for about 30 minutes after some burn in time. 3.0 - 3.2 24/7 be fine for me though, eventually this will be in some one else's machine (mid 2009) so I kinda take it easy on them.

I got the Hard Drive cloned for now, and just did all the major updates. Initially I did the off the cd drivers. Funny thing though I got the web drivers and went to install them and one of the drivers told me the installed driver was newer than it was. Figuring maybe it was an XP SP3 driver that was newer, I checked and the web driver was the newest (CD/WEB) so unless I have a driver ghost installing stuff it came from SP3.

After a while I'll clone to An external drive and Dual Vista with XP as the primary boot.