ASUS P8Z68V Gen3 + i7 2600k + Artic Freezer 13 Pro

Toggle sidebar Toggle sidebar
Joaquim Pereira

Joaquim Pereira

Reputable
Aug 14, 2015
15
0
4,510
The setup above generates what I would call a problem, but I don't know for sure.

When I load the CPU, temperatures rise immediately (1-2s) to +85 Celcius to a point where the CPU starts to throttle back form my 4.3Ghz overclock.

The problem, as I see it, is the cooler stays... cool (barely warm).

Is that normal? Shouldn't the cooler be much hotter in those circumstances (like 60-70 degrees)?

ps - the cooler was removed and installed several times as appears to be very-very solid with CPU
 
Sort by date Sort by votes
Joaquim Pereira

Joaquim Pereira

Reputable
Aug 14, 2015
15
0
4,510
I think so, clean / degreasing up with alcohol, but I'm by no means an expert (but already made lots of electronics in the past and working with this equipment is not strange to me).

I applied a generic thermal compound I had around as I do not wish to use/buy an expensive one until I found the reason for this, to me, abnormal behavior.
That, alone, shouldn't have such impact, perhaps a few degrees more, but not a behavior that seems to me more like an isolator than a thermal conductor ;) .

I've tried to apply thermal compound 1) one drop in the CPU center, 2) spread in cooler base with credit card.
Same result.
I reinforced the metal plate that screws down to motherboard base so it make a lot more pressure on CPU, same result.

Will try to restrict to non OC configuration (and I'll post the results here), but I guess that alone cannot explain why the CPU reaches 95º is seconds (where it throttles down) and the cooler is barely warm to the touch (after several minutes the heat should had spread across the cooler).
 
Upvote 0 Downvote
Joaquim Pereira

Joaquim Pereira

Reputable
Aug 14, 2015
15
0
4,510
Ok, back to 3.4Ghz I got about 65º (Celcius).
I'm using HeavyLoad.

The behavior is the same as in Turbo mode - there's a quick jump to 60º (85º Turbo) then slowly rises +5º~6º (+6º~8º Turbo).
No load temp is around 40º~45º in these tests.

That's normal?

Since my computer rarely reaches full throttle that's not a major problem, but on the other hand I can encode video/ create After Effects compositions that take a long time to render...
 
Upvote 0 Downvote
Joaquim Pereira

Joaquim Pereira

Reputable
Aug 14, 2015
15
0
4,510
I was using DIGI+ VRM Load-Line Calibration on "Extreme" and CPU voltage with a "+" offset that does about 1.4~1.408 V.
Now I'm trying Load-Line Calibration and CPU Voltage set to standard which gives 0% over normal voltages (I think). That sets 1.304~1.312V full load applied. A lot less heat...

I'm using Open Hardware Monitor and I'm been reading temps from CPU, not from Nuvoton NCT6776F chip.

Should I try to bring voltage down even more?

Here are my 3 minute and 10 minute full throttle temps.

3%20minutes%20mark_zps0wdmaixx.png


10%20minutes%20mark_zpsvdw235az.png



 
Upvote 0 Downvote
Joaquim Pereira

Joaquim Pereira

Reputable
Aug 14, 2015
15
0
4,510
I've had a few crashes recently and I'm trying to find out the reason (may not be OC related, but software instead), so I'll keep my current setup for a few days. Some of these occurred when waking up from sleep mode.

Later, I'll try to run 4.3Ghz with the lower voltages but then, perhaps, there must be some other settings to consider.

BTW, how can I set max.1.25V keeping the voltage free to follow CPU speed (Intel SpeedStep)? With voltage offset mode (with a negative value)?
Or should I keep voltage stable at 1.25V, ignoring clock frequency?
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom