• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

FRONTPAGE Intel i7 4790K Devil's Canyon CPU Review

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
29C idle 87C small

I wouldnt delid with those temps...your tim/gap is much better than most.



on another note....5 ghz prime (does 10 mins count :D)

this is just with large ffts prime 27.9...temps are fine since delidded and not hitting any small ffts in first 10 mins.
 

Attachments

  • 5ghz_prime10mins.jpg
    5ghz_prime10mins.jpg
    310.8 KB · Views: 114
Im holding off on delidding for a bit.

Tonight it ran at 4.9 all the way through 3dmark 11 and I popped off a 22,138.

Last night it would not do it, she is breaking in a bit.

Good news is it is at 1.39v
 
Highest I've gotten with mine is 5.1ghz @ 1.48vcore. Not sure how much higher I want to go. Cooling with tap water but I'm going to switch to an ice bucket. After that, I might try delidding.

52ghz4790khwbotandpi_zps9d80af4f.jpg~original
 
Last edited:
Temp are great, but 1.48vcore? :eek:

I'd be more worried of that chip having a very short lifespan.
 
Highest I've gotten with mine is 5.1ghz @ 1.48vcore. Not sure how much higher I want to go. Cooling with tap water but I'm going to switch to an ice bucket. After that, I might try delidding.

Mine maxes out at roughly same, prior to delidding I maxed out at 5.1 ghz for spi 32m, and 5ghz at CB r15 with 1.47vcore

The only thing delidding helped with is benching at around 5ghz, after delid, I only needed 1.38V for CB r15 at 5ghz and could do CB 5050 though back at higher 1.48v, and I could do spi 32m at 5150, upping bclk to 101. well that and the 23C temp drop with prime 28.5 small ffts.

But vvcore required for 4.6 and 4.7 prime 28.5 stable for 10-24hrs was exactly same before and after delid.

but if you do the ice bucket, let us know how it goes, interesting to see.

my max after delidding spi 32m and CB, still needed 1.48v CB at higher 5.05ghz.
5150spm32m.jpg
CB5050.jpg
 
This is as high as I'm gonna get it, 5.2ghz @ 1.4vcore

52ghz4790khwbotandpi_zps9d80af4f.jpg~original


Once I get some cool Labs metal Tim I will delid my g3258 and see how it goes.
 
Last edited:
yep, I agree, that was a nice increase with ice spi on 5.2ghz.... 5.2 on mine is suicide only and that requires fast hands to get saved before bsod.

your making me want to try ice now, especially since im delidded...dang internal rads though....would have to go through lot of work to get it out...if winter were here I could do outside thing...but 90F here outside.
 
thanks, i tried for 5.3ghz and went up to 1.5vcore but didn't want to push my luck. i've got my loop rebuilt, just getting the last of the air out. i'm running 4.8ghz @ 1.255vcore using prime95 small fft with a max temp of 71F. one of the cores is 10F lower then the other 3 which are only a degree or 2 off so it might be a good candidate for delidding.

if i try the ice bucket again i think i will add some salt to get the temps even lower and insulate the water block. i had some moisture forming because i had a fan blowing on the VRM heatsinks.

2014-07-13160536_zps20a17739.jpg~original
 
if i try the ice bucket again i think i will add some salt to get the temps even lower and insulate the water block. i had some moisture forming because i had a fan blowing on the VRM heatsinks.

Actually I recommend to use more fans to get rid of the moisture. I had all tubes wet when I was benching on water with ice. Idle temps were ~0-5*C but it was long time ago, for sure not on haswell.
I was using about 2kg bricks of ice ( frozen water in plastic bottles ) and not much water in the bucket. In this way it was melting slower and keeping lower loop temps. I was also switching water from time to time as water in the bucket was heating up by ~10*C.
 
Are you sure you are getting good contact off the delid? At times there isn't enough pressure after a delid as a lot of coolers don't have the flexibility to reach that much further down.

I doubt it's a problem with haswell... Wouldn't all of them have that problem? Pretty blanket statement there.
 
I doubt it's a problem with haswell... Wouldn't all of them have that problem? Pretty blanket statement there.
Ya it's a pretty bad generalization, but regulating tens of amps on dye means you get a fair amount of heat from the small drop, and I bet the thermal diode is right next to that.

I wish someone put a big chunk of something over the dye which is transparent to IR and reasonably thermally conductive so the hot spots would be perfectly clear.

Gallium TIM would probably help a lot.
 
Back