OC help i7 920 @ 4GHz

Ah so you get a drop just from them alone, so with the other improvements that you are planing, the new top, and 2nd fan, should see the temps drop quite abit more also.

I kinda thought it was'nt the seating issue that some kept saying, im glad the vented covers worked.
 
ok this is gonna sound weird but the lower temps lasted maybe half an hour and now i am up to 85c again.
is the temperature dependent on p95 phase? are some phases more intensive than others? i have never noticed such thing before. My Q6600 hovers at pretty much constant temp under load. So did my athlon 64 before it and my pentium 4 before it.
i am back to square 1. 85c with vented bay covers. waiting for the vented top to come in. that should make a bigger impact since the existing top actually feels hot to the touch, so i am guessing lots of heat is trapped inside.
 
Id have to guess change in ambient, forced the temp up, or the heat built up in the case again forcing the temp up, not to sure really.

Main thing is you saw a drop in temps, all be it little just from vents, so the 2 new top fans will really help i think, also the 2nd fan on the cooler also.

Also bare in mind nothing will stress your cpu day to day to prime levels of heat, unless you encode a lot, or do distributed computing (F@H/Seti).
 
my ambient is constant at 24c :(
yea i do encode every so often when i buy a dvd or music cd.
i also participate in f@h but 90% of the work is done by the ps3.

i will keep tweaking bios settings to see if i can somehow decrease vcore without losing stability or clocks.

at least i got some decent airflow now so i can't really complain.
i should post pics of the pc eventually.
 
Last edited:
I would personally say at least we know what the issue is now, and things are in the works to sort that out, so its just a waiting game really until the stuff you need arrives.
 
aye, i can't do much at the moment apart from tweaking the bios a little bit to save some volts here and there.
i am charging the camera now :D
 
ok, i am updating the status of the issue:
i have moved the new PC into a PC-7FNWX to take advantage of the USB3 ports in the back and also because it has a 140mm at the top right above the CPU. I also installed some vented bay covers on the FNWX. In the back I have a Noctua 800rpm 120mm fan and in front I have the stock 140mm fan. I have not installed the 140mm fan at the top because my order has not arrived yet but I have removed the cover to get some passive dissipation. So far after 20 mins of p95 I am doing 82c which is basically the same as I was doing in the FN with the vented covers.
In the mean time, I reseated the heatsink a dozen times and I always end up with the same load temps after a couple of hours of p95.

In the PC-7FN I put my Q6600 rig which should be quite happy to run without the 140mm fans on top but what the heck? i will install it anyway when it arrives, hopefully tomorrow :D
Meanwhile, my Q6600 found new glory when I finally managed to get it at 3.6ghz stable for 11 hours in p95 and only at 75c. One core failed but the temps allow for more vcore so I will be posting results on that front too.

Oh and I got a stacker 830 for the bin now...
 
it's not normal. 1.275 vcore is hardly allot. I assume it's d0 stepping and not g0?

If the heatsink has been reseated multiple times do make sure it's cleaned properly, and there will be noticable high temps for a few days until it settles. The issue does not lie with the case imo.
 
Yes, it was cleaned up properly every single time. If i had a penny for every time i installed a heatsink... :D It is D0, currently doing 82c after 1 hour p95. All the cores are within 2c of each other.
At this point if there is bad contact then the TRUE does not have a flat surface or the CPU or both. Unless the akasa apache on the TRUE is not what I thought it would be.
I have the same exact heatsink on the Q6600 and it performs great with a standard Lian Li fan on it.
I use standard ceramic TIM so there is no setting period. WYSIWYG with this thing.
 
it's either there's bad contact, or you're jus unlucky to have one of those chips that love high temps =/ it's november, pretty cold. I really doubt ambients (even with 470 SLI) can push a [email protected] over 80C on a true. what are you using to measure these temps?
 
bios is latest on the P6X58D Deluxe. I am using RealTemp and I cross-reference it with coretemp. Asus probe shows something like 10c less, however i know realtemp is right because once it hit 100c (failed oc attemp) and it started throttling.
So temps are reasonably accurate. I know I made no mistakes mounting the heatsink, i know the fan is good and my ambient is 23c.
So why so hot? I got no clue. Maybe this TRUE is a dud :p
Actual vcore is 1.248 under load. I have LLC enabled. The next thing to try is to push vcore higher up and disabled LLC but i highly doubt that will change anything apart from increasing my idle power consumption and temps.
 
Nah the true is not a dud, thats the sort of temps you will be getting with 1 fan, and thats what you are using, i bet you as soon as u strap on a 2nd fan onto the true the temps will drop.

Don't forget all the mid 70's, low 70's, high 60's on air are all dual fan set ups.
 
i do happen to have spare fans but do they have to be the same model?
that's the thing that is not clear to me. all these people who get low temps run dual fans?
 
Yeh they generally have the top top end air coolers, with atleast 2 fans on them. OR are on water cooling.

It is ideally best to use 2 of the same make and speed of fan, but isn't necessary, but if using 2 that are different speeds, have the fastest one pulling air, and the slower one pushing.

Its always worth a quick test if you have spares, and cable ties so u can do a quick ghetto install for testing purposes.
 
true 120 extreme? there's quite a few trues arent there?

I'm on c0 stepping, when I hit 4ghz and use only 1 fan to push it tops out at 75 (+~2 on scorching summer day).

Maybe there isn't anything wrong, could be the cooler/chip then. jus makes no sense d0 is worst off. my c0 is only an average chip.

The cpu fan is set to maximum speeds right? it's not set to rev up and down depending on pwm/voltage ???
 
yes, ThermalRight Ultra 120 Extreme. TRUE, there is only one and it's the extreme. I have revision C of the cooler. And yes, the Apache is at max rpm constantly.
i just installed a 140mm fan at the top. I will report after an hour of p95.
 
You must log in or register to reply here.
Back
Top Bottom