Sign in with
Sign up | Sign in
Your question

Any1 w/ Q6600 OC'd to 3.4++Ghz Stable Air-cooled?

Last response: in Overclocking
Share
June 9, 2007 7:20:37 AM

From the title above.. Does anybody managed to OC a Q6600 stably to 3.4Ghz or above using air cooler?

I roam around the net searching for OC'd Q6600, what i found are a lot of Q6600 just running around 3.000 Ghz stable... I haven't find something like running in the 3.4 Ghz stable, unlike the Q6700 which are common... As what i've read from a lot of reviews on Q6600, running above 3.00Ghz mark are certain to hit some major issues and would render the OC useless as it is unstable..

Anyone with Q6600 able to OC this quad stable above 3Ghz mark?
Any find are greatly appreciated...
June 9, 2007 8:24:57 AM

I can do 3.2 on my system and have it orthos stable. You really need a more aggressive cooling option (i.e. water) if you wanna go much above that number; there are limits to the rate at which air can dissipate heat. Remember that a quad chip is basically just two duals on the same die which you're asking to more or less cool simultaneously with a single HS/F.

Also, unless you live in the North Pole or keep your room at a pretty cool temp via air conditioning, small changes in room temp and can big effects on your load CPU temps. See this thread for details. I made a 7-point plot at different room temperatures and found that every delta °F of room temp. affected the average core temp by about 0.8 °C. For example at 68 °F the load temps averaged for all cores was about 53 °C. If the room temp was about 78 °C, that average shot up to about 60 °C.
June 9, 2007 10:48:52 AM

I was thinking also of the temp. before.. Yeah, i failed to internalized before that quads are heat monster procs... Thanks for that one dude, i'll just be keeping it at that 3Ghz platue with my TR 120 Extreme... Definitely now, i was convinced that it was the heat issue that is to consider..

Too bad my TR 120 can't hold much with a Quad proc...
A water cooler could be on the way, but definitely, not until end of the year.. :roll:
Related resources
June 9, 2007 2:21:07 PM

Dude i have a Quad 6600 not a DC E6600... By theory, i have two E6600 running in a single physical CPU... So heat output is another matter :(  , and its puts up an OC ceiling to me being a non-water cooler as of this time...
December 30, 2007 7:31:23 PM

Yes I run my q6600@3.213 on stock voltage using a zalman 9500 and AS 5 Compound, very stable.
January 3, 2008 5:57:50 AM

i have a q6600 running at 3.6 ghz with 1.5 vcore. that is right in the range of recomended for the processor. if u will look harder you will find alot of ppl have them at 3.6 ghz. so long as they are G0 steeping.
January 3, 2008 2:26:37 PM

Specs:
Q6600 B3 @3.3mhz V1.5
Mother: Maximus formula
Loaded temp: 53c 68/68/68/68 (speedfan and Prime 10hrs w/ round check enable)
Ultra-120 extreme lapped w/noctua1200
Antec900
Arctic MX2 TC
January 4, 2008 6:14:21 PM

I have my q6600 G0 running at 3.33ghz with vcore at 1.35. keep in mind you can tweak some other BIOS settings to improve stability. To run at 3.4ghz I had to use a drastically higher core of around 1.425. Personally I don't really want to go over 1.4V on my processor but I THINK you can safely go to 1.5V. I think those reviews that attest to instability over 3.0ghz are just plain wrong... my system is rock solid and heat is not an issue. I'm using an abit ix38 quadgt and a zalman 9700nt.
January 5, 2008 1:17:01 PM

I can get mine up at 3.6Ghz (9X400).. But it runs pretty hot and sucks up 1.5v to remain prime95 stable... Personally, I dont see benefits over the normal 3.2 I run it at... So, I have mine set at a cool, stable 3.2 as of now.. I dont see the point in risking ur sweet Quad for a few more megahertz!! Its still a mind-blowing processor at 3.2 :) 
January 5, 2008 4:34:52 PM

mm I may just get a nice high CFM fan to put on my old ultima 90 lap the HSF get the best thermal paste and see how high I can clock it.

It would be a bit of fun...
January 5, 2008 10:23:31 PM

3.6 is far from unheard of. There are two factors that would limit arrival at this speed: voltage and temps. Voltage must be kept at or below 1.5 or you run the risk of premature burnout (maybe not right away, but well within the 3ish years a typical performance CPU is run). The hottest core should be kept at or below 75*C or, again, you run the risk of a fry.

All this being said, numbers are numbers. Will you notice a difference between 3.4 and 3.6? Eh...maybe. Will you notice a difference between 3.2 and 3.6? Again, maybe. Right now, games do not bottleneck at the Q6600, even at the stock 2.4. Where you'll really start to notice those speed gains is toward the end of the Q6600's useful life, maybe two or three years down the road from now. At that point, when it becomes the bottleneck, one will probably see notable gains between 2.4 and 3.2, and 3.2 and 3.6. Right now, in 2008, it's just an academic exercise.

All this being said, a good OC mantra is to stay well within specs. The extra 100 or so megahertz is not worth the liability. Keep your vcore at or below 1.45, and keep your hottest core at or below 71*C. Whatever you can get while staying within those numbers (probably around 3.4ish) should be more than fine.
January 7, 2008 2:42:18 PM

try get xeon x3220, they are same as q6600,but some require low voltage, I have a x3210, doing 400*8 with 1.2125 volt(vid),
June 23, 2008 6:10:42 AM

On B3 I get 3.33 or 3.4 on same 1.36v - very stable and idle 32-28 coretemp. 9X370 or 9x378. High 50's under load. Early 60's max in summer. Will post 3.6 easily but eventually crashes after a few days.

8x, 7x etc not stable. i.e. 400 or more regardless of mem timings etc.

Ultra 120 extreme cooling recommended. Quickly drops peak temps almost instantly. Fans on medium only - high nto required.

Nine Hundred case. Side fan had little impact (1-2 deg for 120). Put is on the other PC in the end as didnt need it. 8800GTX at something like 640/1550/1048. 71% on fans all is needed max 55-56 deg. 14K 3DM2006. Room air-cond for summer, ambient temps normal.

Memory had been the biggest issue IMO - Kingston 6400 4/4/4/12 at 800. Seems to have an aversion to 888 plus (378x2.4 say) and loosening timings doesnt appear to work well any higher. From memory I have it on 2x375 at 4/4/4/12, or it could be 5/5/5/15 for current 3.4 o/c. Tweaked voltages to +0.20 for all options under Gigabyte p35 ds3.

SuperPi 1M - approx 15.3seconds. Fine for long term purposes and no need for more IMO. 3.6 gave me 14.7 ish for same test. Can post full info later when Im at home if of interest. If your on GO from my understanding with some decent cooling (Ultra very highly recommended) the above should be very achievable.
July 7, 2008 8:22:03 PM

i run mine at a stable 3.2 vcore 1.35, no real issues as of yet. i did get mine to 3.82 on air (test only) and i had no problems, but i only ran it for about 2 hours (testing/benchmarking, ect) but temps hit 75C. i backed it down to 3.0 and then back to 3.2...temps didn't change too much (maybe 2-3C) now i sit about 46 idle and peak at 68C if pushed...everything runs fine, no issues as of yet...good luck.
July 8, 2008 12:40:47 AM

quick add on....one thing to also look at is not just how your cooling holds up on full load over time...but how fast it recovers once the load is removed. i can make high temps happen...but short of prime95...i never have all the cores loaded at the same time...and even the loaded ones don't stay maxed out for long periods....so looking at a peak temp is one thing...but also think about avg temps and your coolers recovery rate..just a thought
July 8, 2008 1:22:01 AM

Strange. 3.6ghz is by far the most common speed people run q6600 at when air cooled. It's a simple set-and-go oc, takes 1 minute to set up. Heat is hardly an issue with any half decent air cooler. I get 59C full load at 3.6ghz, and 69C at 4ghz. Although at more than 3.6ghz, there is a big jump in vcore required to run prime95 stable. I couldn't get it to pass stress at 4ghz on any less than 1.65v. It's high, but since Intel's "recommended" vcore is up to 1.5 for q6600, not really what you'd consider dangerous. 65nm g0 q6600 have a higher voltage tolerance and requirement than just about any other cpu.
July 9, 2008 4:56:20 PM

I run 400x8=3.2 with temps in the mid 50's and a vcore of 1.375 on a Q6600 with a VID of 1.325.

It is rock solid stable. I have a Xig cooler and Antec 300 case. I have the same setup in a 900 case and there is no difference in CPU or GPU temps.

I can get 3.6 stable but the temps are in the mid 60's fully loaded (on Prime).

From all of my reading (last month or so), I wouldn't say 3.6 is "a simple set-and-go oc" with the recent batches of Q6600's. They typically have higher VID's (1.325 seems common). If you are buying a Q6600, go with an OEM over the retail. It seems the OEM's are more likely to have a lower VID. I have both (ordered within a week of each other and both from newegg) and the OEM had a lower VID
I am more comfortable with 3.2 with mid 50's for everyday use.

Your heat is going to be directly related to the vcore and the amount of vcore required seems (lots of evidence) to be directly related to the CPU VID.
July 17, 2008 5:55:59 AM

jjblanche said:
3.6 is far from unheard of. There are two factors that would limit arrival at this speed: voltage and temps. Voltage must be kept at or below 1.5 or you run the risk of premature burnout (maybe not right away, but well within the 3ish years a typical performance CPU is run). The hottest core should be kept at or below 75*C or, again, you run the risk of a fry.

All this being said, numbers are numbers. Will you notice a difference between 3.4 and 3.6? Eh...maybe. Will you notice a difference between 3.2 and 3.6? Again, maybe. Right now, games do not bottleneck at the Q6600, even at the stock 2.4. Where you'll really start to notice those speed gains is toward the end of the Q6600's useful life, maybe two or three years down the road from now. At that point, when it becomes the bottleneck, one will probably see notable gains between 2.4 and 3.2, and 3.2 and 3.6. Right now, in 2008, it's just an academic exercise.

All this being said, a good OC mantra is to stay well within specs. The extra 100 or so megahertz is not worth the liability. Keep your vcore at or below 1.45, and keep your hottest core at or below 71*C. Whatever you can get while staying within those numbers (probably around 3.4ish) should be more than fine.



I disagree!!! Im running a q6600 at 3.6ghtz on an extensive air cooling system, I was able to get the temperatures to around 55*C under full load which is on the higher end yes, but its not too hot either. Before I made the jump to 3.6 i was running 3.2 for close to a month, and I play a game called EQ2 which is a very CPU intensive game, well, to make a long story short I can see a major difference between 2.4 GHZ and 3.2 GHZ, and i also can see a small difference in game performance from 3.2 GHZ to 3.6 GHZ, so im not saying you are completely wrong there are games that are not so CPU intensive, maybe a good majority. However, there are games that are very CPU intensive, EQ2 (everquest 2) included, and therefore you will see a difference in game performance. Also please keep in mind you "should" only see a difference if the application is 32 bit, any 64 bit application you probably would not see a difference in performance until the time comes when data outmatches the cpu.
December 1, 2008 11:51:02 PM

I don't know why does everyone increase their voltages. I think I can go higher with stock voltage but there is no reason for me :D 

My setup:
Gigabyte P35DS3L v2.0
Q6600 @3.2Ghz Stock Volts 400x8
VID 1.2625V Z-CPU 1.216V w/ Zalman 9700@1000rpm 36C idle/44C ingame; 25C ambient
4GB 4-4-4-9 800Mhz 1:1 XMS2
MSI Geforce GTX 260OC 620/2160 @52C idle/64C ingame
:o  :D  :o  :bounce:  :bounce:  :bounce: 
Anonymous
a b à CPUs
a b K Overclocking
December 2, 2008 1:14:28 AM

I don't have a q6600... but I have a q9550 at 3.85 ghz air 24 hour prime stable
December 2, 2008 2:35:05 AM

You shouldn't have any issues OCing to 3.4+ if you have a G0.
December 27, 2008 1:15:13 AM

Yep, I am running 3.42 on q6600 on air - ocz vendetta - no problem at all :hello: 
December 27, 2008 9:00:32 AM

I also have a Vendetta (unfortunately the first version with the louder fan). My Q6700 has a horrible VID of 1.325v, so I need 1.475v for it to pass torture tests, but I leave it on 1.45v when not testing. This is @ 350 FSB x 10 CPU multplier for 3.5 GHz. I was wondering if high FSB was better or high multplier.
December 31, 2008 9:38:11 PM

Video_freak said:
I also have a Vendetta (unfortunately the first version with the louder fan). My Q6700 has a horrible VID of 1.325v, so I need 1.475v for it to pass torture tests, but I leave it on 1.45v when not testing. This is @ 350 FSB x 10 CPU multplier for 3.5 GHz. I was wondering if high FSB was better or high multplier.


When you increase your FSB you are overclocking CPU and memory
When increasing multiplier just CPU is being overclocked
When doing both: depends on what your final memory timings and frequencies are.
In theory 389 x 9 or 350 x 10 with 800Mhz memory same timings should be identical in performance. the 389 x 9 might be more stable with less Voltage :heink: 

!