Should I reapply thermal paste?

[Gofspar]

These are my temps after 4mins of AIDA 64.

 

For a 3770k should I reapply or are they just that hot of a chip?

[Gofspar]

THOUSAND POSTS HYPE

[nj4ck]

Yes. 

[Gofspar]

Yes. 

Yes on reapply?

[nj4ck]

Looks fine to me. Under load, that's about what mine is at (playing TF2 and streaming to Twitch)

you get 90°C under gaming load? Doesn't seem normal to me...

[nj4ck]

Yes on reapply?

Well how long has your current TIM been on there? Usually reapplying is never a bad thing, and if you're getting into the 90s with a Kraken X61, that's definitely not normal.

[Gofspar]

Anyone else, opinions are what im basing my decision off of.

[Gofspar]

Well how long has your current TIM been on there? Usually reapplying is never a bad thing, and if you're getting into the 90s with a Kraken X61, that's definitely not normal.

month ago?

 

NT-H1, I might of slid the block around too much mounting the fucker.

[deVIZtation]

you get 90°C under gaming load? Doesn't seem normal to me...

Yeah. What cooler do you have on it though? My temps for my 3570K never go past 60C, but I have a beefy cooler.

 

Kraken X61, hmmm yeah... reapply.

[Gofspar]

Also I used the line method, should I try the Pea method?

[nj4ck]

While encoding on the CPU at the same time. Not just gaming.

What cooler do you have? My 4670k @1.3V never exceeds 70°C even under Prime95, although it is delidded and watercooled.

[nj4ck]

month ago?

 

NT-H1, I might of slid the block around too much mounting the fucker.

Ok, then its definitely not the thermal paste. You either have a bad mount or the TIM between the die and the IHS has dried out, which is typical for Ivy Bridge and Haswell chips. There's a delta of more than 7°C between your cores though, so I'd guess you probably have a bad mount. 

[Najuno]

Well the idle temps of each individual core look okay. What CPU cooler do you use? But the load after 4 mins over 90° Celsius and that with this low Voltage? I´d say that´s not okay. I can tell you this much with my 8core monster here, to be fair I run a custom loop on the extremer side, I get with 4.7GHz @1.45V in 2hours Aida64 about 72° Celsius on the hottest core.

So you really want to check the proper seat of the heatsink or buy something more powerful to cool that CPU because even with an acceptable Voltage you run a higher OC. What thermal compound do you use?

[nj4ck]

212 Evo with a modest application of OC diamond. I'm at 4.3 Ghz on 1.116 - 1.25 volts. Mind you that when I'm streaming my door is closed and its in a room with two OC'd 580s, another system, and a dual CPU server running.

well ok, that explains it. I personally still wouldn't be comfortable running 90°C over extended periods of time...

[nj4ck]

I'd like to de-lid it to see how big of a difference it would make.

When I delidded mine I got a 40°C temp drop. Mind you, that is quite drastic and it turns out the TIM under my IHS had literally turned to stone (thanks, intel). I think the average drop people get is about 10-12°C.

[Gofspar]

When I delidded mine I got a 40°C temp drop. Mind you, that is quite drastic and it turns out the TIM under my IHS had literally turned to stone (thanks, intel). I think the average drop people get is about 10-12°C.

I used a bb sized dot in the middle this time and temps went up gg, can you give me an example of exactly how much I should use

[nj4ck]

I used a bb sized dot in the middle this time and temps went up gg, can you give me an example of exactly how much I should use

 

If you're getting a bad mount, it doesn't have much to do with the thermal paste though. 

[Gofspar]

 

If you're getting a bad mount, it doesn't have much to do with the thermal paste though. 

brb trying again, how do you suggest I mount it, theres 4 nuts on the side and the block slides onto these little screw things and you tighten a nut on it.

[nj4ck]

brb trying again, how do you suggest I mount it, theres 4 nuts on the side and the block slides onto these little screw things and you tighten a nut on it.

The line method works better for Haswell and Ivy Bridge, because of the long-ish dies, that's probably why your temps were going up with the pea method. When mounting, make sure you tighten the nuts in a cross pattern and with exactly the same amount of pressure on all 4 of them. Count the turns, if necessary.

[Gofspar]

The line method works better for Haswell and Ivy Bridge, because of the long-ish dies, that's probably why your temps were going up with the pea method. When mounting, make sure you tighten the nuts in a cross pattern and with exactly the same amount of pressure on all 4 of them. Count the turns, if necessary.

alright brb doing again.

[Gofspar]

The line method works better for Haswell and Ivy Bridge, because of the long-ish dies, that's probably why your temps were going up with the pea method. When mounting, make sure you tighten the nuts in a cross pattern and with exactly the same amount of pressure on all 4 of them. Count the turns, if necessary.

I fucking love you, dropped temp by 30c. I was screwing it down with no fucks given. I counted my rotations and it dropped by a fuck ton.

 

Like im legit teary eyed

[nj4ck]

I fucking love you, dropped temp by 30c. I was screwing it down with no fucks given. I counted my rotations and it dropped by a fuck ton.

glad to hear! 

[Gofspar]

glad to hear! 

Do you have Skype or something? I'd be interested in being your padawan. <3

[nj4ck]

Do you have Skype or something? I'd be interested in being your padawan. <3

whoa whoa, lets not take it over the top here 

[Gofspar]

whoa whoa, lets not take it over the top here 

Honestly though, thank you.