• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Q6600 Temps ?

S

slamcmy

Limp Gawd
Joined
Sep 10, 2007
Messages
241
hi guys right now i've overclocked my q6600 g0 stepping to 2.6 ghz and im running prime95 small ftt's torture test, and its stable and without errors for 1 hour and 45 minutes...
i forgot to tell you that i have the stock Heatsink Fan from intel and a antec nine hunder case with 4 120mm fans and 1 200mm fan on top, also in my motherboard bios all voltage settings are set to automatic....these are my temps which i monitor by using everest ultimate:

----LOAD TEMPS PRIME95 SMALL FTT'S TEST----
Motherboard: 22 C
Cpu: 55 C
Core/1/2/3/4: 62/62/58/58
Cpu Fan: 1688 RPM ( Working at 100% All The Time )
Cpu Core Voltage: 1.18 V

----CPU Z INFO----
Cpu Core Speed: 2603.3 Mhz
Bus Speed: 289.3 Mhz
Rated Fsb: 1157.1 Mhz

with these temps under full load and with the same settings i have on my bios, how far do you guys think i can get with my Q6600 without changing the cpu core voltage or changing the heatsink fan....and by the way what do you guys think about these temps ?

thanks you for your help
 
your temps are fine as to how far you can OC only way to find out is to OC it! you might have to up the vcore some to get to 3ghz+ and you should run prime all night to see if your oc is stable
 
Intels spec is 71c for a Go Q6600.

If you are going to play around with clocking, it will be more enjoyable if you don't run on the ragged edge. I would spring for an aftermarket cooler and some heatsink compound.

My new system (that's supposedly on a fedex local delivery truck) has a Thermaltake V1. Not only does it work well, but it looks amazing. Too bad the CM Cosmos does not have a window.

I am hoping for the low 3s with a mild voltage bump. I prolly won't run the temps much above the low 60s. Wanna leave some headroom for summer.

Don
 
Heh,

Danny... he's actually right I believe, but a complex technical explanation is a bit tricky. As I understand it (and I may be wrong,) the short of it is that the 71C temp marked on the box represents the max sustainable operating temperature measured from the exterior of the chip, directly on top of the cores, but under the IHS. I'm not sure if this is the same temperature read as the CPU temp in Speedfan and several other programs, or if it is from some third reading which can't be read again once the CPU is assembled.

The temperatures read from CoreTemp are the temperatures within the cores themselves, which can be reasonably assumed to be hotter than a temperature read from the top of the cores, under the IHS. The max-operating-temperature from within the Core is 100c, a.k.a. the T-Juncton temp. If the processor reaches this temp, it will begin to throttle down, regardless of whether or not you have Thermal Control enabled in the BIOS.

Many people assume that a CoreTemp reported temperature of 71C is the safe maximum temperature for a Q6600, but this is not the case. As Mithent stated, there is at least an additional 10-15C of headroom before one can logically conclude that the CPU cores are running too hot. My personal preference is to keep temps below 80C (or 85C is ambients are low and I'm purely benchmarking.)
 
Ahh thank you for the clear up Markyip1 and Mithent. Did not know that about the Q6600 temperatures. I was one of those people who assumed that 71C was the max temperature of the CPU without knowing about the 10-15C additional headroom.

Thank you again.
 
i think im gonna buy the thermaltake v1 heatsink fan cause its much easier to install on my mobo cause it doesnt require the motherboard to be removed in order to to be installed...when we are talking about raising the vcore ? does that mean the cpu core voltage ? how should i increase the cpu core voltage in a safely way, so i dont burn my cpu ?
the voltage im running now is 1.18 v

thanks for your help
 
DonDon said:
I would spring for an aftermarket cooler and some heatsink compound.

My new system (that's supposedly on a fedex local delivery truck) has a Thermaltake V1. Not only does it work well, but it looks amazing.
Click to expand...


I have one and used it for months with a quad.Great sink even with a quad,and damn light.Only problem is dust,its a dust magnet more them most after market solutions,and hard to clean even with canned air.

GL !
 
I purchased in G0 stepping a couple of weeks ago, very impressed so far.
Your temps look pretty good. At stock voltages seems like I was hitting 3.3ghz,
slight increase to 1.4v was good to about 3.7ghz. Temps were getting a bit high at that
point under full load for me, 66-67C. Switched to water-cooling to run 24/7 1.4v @3.6ghz
and now it's a folding monster, temps at 50c.
 
Standard practice is to see how high you can overclock your chip without raising vCore (which is the CPU core voltage.) Once you've reached an overclock that is unstable, you can attempt to raise vCore until it is stable. Best to do one notch at a time on the vCore, so that you're not needlessly heating the chip. As far as max vCore, that's really up to a) what your cooling solution can handle and b) personal preference.

In most cases with a Conroe / Kentsfield, temps brought on by a higher vCore and OC will overwhelm your cooling well before you've reached your chip's safe voltage limit. However, assuming the chip is paired with very high end cooling, many believe that a safe everyday operational vCore on a Conroe / Kentsfield chip is 1.55v. Anything higher won't ruin the chip in one sitting, but is best left for benchmarking OCs only. Others believe that the voltage can go much higher (up to and including 1.65-1.7v) for everyday operation without ruining the chip over any reasonably useful period of time, but there is still debate on that subject. The debate centers over the idea of electro-migration, which supposedly (and eventually) makes the chip unusable. Since it is hard to determine whether or not electro-migration is a real and active cause of chip failure and/or whether or not over voltage always leads to said electro-migration, some believe that it is basically FUD while others are very cautious not to raise their vCore over what they believe to be a safe limit. Another argument often brought up in the debate is that even if a high vCore reduces the useful life of a chip from say 10 to 8, 7, or 6 years, does it even matter if I don't plan to use it past 2 or 3?

Note that for Woldfale / Yorkfield chips, everything said above still holds true, but the considered safe vCore limit is .1 lower due to the smaller die size. Hence, people are now stating that an everyday use vCore of over 1.45 for a E8XXX or Q9XXX isn't the healthiest thing for the chip.

Finally, note that with your V1, you will likely max out your cooling capacity before exceeding a vCore of 1.55.
 
Hey, thanks for the clear explanation of the core temps.

My system showed up today. Loading OS's now, and will start some clocking tomorrow.

The Thermaltake V1 looks amazing. We'll see how it clocks in a few days.

Vista gave me all 5.9's accept for memory which was 4.7. Not sure why, I got 2 gig of 800 mhz low latency ram. Oh well, we'll see when I get in and start playing with the bios settings.

Don
 
I was having tons of probs with an asus p5n32-e sli board...so I finally gave up on it and bought a xfx 780i at fry's.

OMG.

3600woot.jpg


with a TR U120 extreme and a G0 q6600 I am at 3600 1.42vcore
idle is 27 on 2 cores 32 on 2
prime running hits 54 on 2 cores 60 on the other 2 with a 1200 rpm scythe on the 120

I cant believe how easy it has been with this board. I spent 2 weeks pulling what little hair I have out fightint that dang asus board.

I am gonna run prime overnight for a good 10hr run to make sure everythign is solid, so far this pass as run for about 30 mins with no errors, then I restarted it.
 
The coretemp and its limit explanations etc. in this thread are mistaken/misleading. Too many old posts easily found so I am not going to waste my time.

You are good to around 100C coretemp on a quad but I recommend some headroom be reserved. Use coretemps "show Delta to Tjunction" to determine exact amount of reserve thermal capacity in deg C you have until the processor thinks it is hot. Note that this is not the thermal shutdown temp. With a "show delta to Tjunction" reading of 0, you still have 10-15C until the CPU will turn itself off.

Here is how you can read the CPU DTS register directly yourself. (which is all these programs do and then do the conversion and math. )

http://www.hardforum.com/showpost.php?p=1031080147&postcount=5

Please read the Intel data sheet and the Thermal and Mechanical Guidelines document easily downloaded in PDF format from Intel before spreading more wrong information around the net.
 
"To ensure functionality and reliability, the processor is specified for proper operation
when Tc is maintained at or below the thermal profile as listed in the datasheet."

In theory this is measured with a thermocouple placed on the top of the chip, but this is obviously not done on a normal system - the Tc is estimated as a delta from the hottest core Tjunction.

While it's true that PROCHOT# and THERMTRIP# aren't activated until significantly higher (based on the per-chip DTS calibration), that doesn't mean that it's a good idea to regularly run the chip higher than its thermal profile.
 
You must log in or register to reply here.
Back
Top