Overheat
Images are sometimes not shown due to bandwidth/network limitations. Refreshing the page usually helps.
You are currently reading a thread in /g/ - Technology
You are currently reading a thread in /g/ - Technology
I have recently had overheating issues with my 6700k, I have an H100i and in prime 95 it can go up to 80C. I have tried cleaning the processor and replacing thermal paste, and cleaning the radiator. What else could be wrong?
4.7 GHZ
1.35V
>>53141637
thermal grease under the heat spreader?
>>53141653
I'm not going to de-lid my CPU
>>53141637
>up to 80C
you what?
Fan problems most likely senpai
>>53141637
>What else could be wrong?
The fact that it's clocked at 4.7GHz at 1.35V you fucking mong
>>53141706
Actually learn what you are talking about faggot, go check silicon lottery. Their 4.7 chips are at 3.393 volts. Fucking retard
>>53141637
>prime 95
I hope you are not using the most recent versions because this could be your problem
>>53141821
Yes I am using the most recent version, what is wrong with the newer versions?
It also gets very hot in Aida64
>>53141852
Just return it as faulty how hard can it be to report that in idle(aide64) it gets those temps
I dont get more than 50C under maximum load for hours on my 3.3ghz 6600
>>53142006
then again, that is 3.3 GHZ
>>53141637
Is it throttling or shutting down? If no then its not overheating.
>>53141769
Stop trying to blow people's CPUs up faggot
>>53141637
80C isn't overheating for that chip.
You don't have any problem aside from autism.
>>53141637
Prime95 isn't a good representation of what temps to expect in "real world" scenarios use other programs ex. OCCT or Cinebench, also I don't recommend going past 1.3V
>>53141637
80C with 1.35v under Prime95 is fine retard, do some research if you're going to overclock
>>53144304
this
unless you live in canada where ambient temps will be 10c or below, 80c is damn good for 4.7ghz.
>> $400 CPU
>> no watercooling
kek
>>53145133
> >>
> kek
Why don't we have 10ghz cpus yet? When will the cores meme end?
>>53141852
Recent versions of Prime95 break your shit. Don't.
>>53145243
2011 according to Intel in 2000
http://www.zdnet.com/article/taking-chips-to-10ghz-and-beyond/
>>53141637
>he bought a water cooler
>he thinks he can buy a 6700k without delidding
>he still tries to overclock
You deserve everything you get. Enjoy your paperweight.
>>53146805
>You can break an Intel CPU by making it find prime numbers
You can't make this shit up. If anyone needs proof that Intel is fucking the market over then they can look right here.
>>53145243
It's all about IPC. That's why AMD chips with 8 cores at nearly 5 GHz can't beat a modern Skylake CPU.
>>53147136
>we have to go back!
80 is fine. if you dont like it, consider lowering voltage
>>53141674
Well then suck it up, because if you don't do that you're fucked. Once you burn your cucked i7 up, maybe you'll wise up and grow out of the childish waste of time that overclocking is.
>>53148197
Cancer: the post
>>53148228
cancer calling cancer cancer calling cancer calling cancer cancer calling cancer calling cancer cancer calling cancer calling cancer cancer calling cancer calling cancer cancer calling cancer calling cancer cancer calling cancer cancer
>>53148421
>>53148197
>>53148228
>>53148421
>>53148439
OP should kill himself because he's cancer calling cancer cancer calling cancer calling cancer cancer calling cancer calling cancer cancer calling cancer calling cancer cancer calling cancer calling cancer cancer calling cancer calling cancer cancer calling cancer nigger
What is going on in this thread?
>>53148523
GET THE FUCK OUT OH GOD
>>53144304
>>53144815
Agreed.
might be a bit off topic but i'm kinda into overclocking. i have a haswell cpu and every guide i read it talks about not going above 1.35v. i thought once the lithography is shrunk like it was with skylake that it can take even less voltage, but it seems like people are pushing even more volts in these than last generation. what gives?
>>53148716
Haswell had the VRM on the cpu itself, which made it get very hot with added voltage, combined with the mediocre thermal paste under the lid, which is why it was recommended to not overvolt the chip
Skylake moved the VRM back onto the motherboard, and has better thermal paste under the lid so it doesn't run as hot and can overclock better.
While using less voltage is useful in overall efficiency with smaller die sets, when overclocking it requires more and more voltage the higher the clock speed because physics. Skylake can generally take less voltage for a given speed than Haswell, but that depends more on the silicon lottery than anything.
>>53148817
The VRM off the die makes sense. The crap paste that the haswell refresh was supposed to help correct didn't seem to do as much as i thought.
Thanks for the reply.
>>53148920
>The crap paste that the haswell refresh was supposed to help correct didn't seem to do as much as i thought
I mostly meant 1st gen Haswell.
Devil's Canyon had much better thermal paste and a more refined process, so users were able to get on average 200-300mhz more at a given voltage with cooler temps vs 1st gen.
4.4ghz-4.5ghz was considered about the max for Haswell 1st gen without exotic cooling, while 4.6ghz-4.7ghz was readily achievable on Devil's canyon.
>>53149064
My 4770K runs at 4.6 GHz and 1.26 volts. I am cooling it with Noctua D14 and the temps are ~75 degrees when running AIDA64 stress test. Nothing exotic here really.
>>53149235
Again, silicon lottery. Sounds like you got a pretty good chip. My 4770k runs at 4.4ghz with 1.25 volts and hits about the same temps
>>53148482
>(You)
>>53149274
Not sure if it's a good chip. I have been trying to get it running at 4.7 GHz, but gave up at 1.35 volts. Shit wasn't stable.
>>53149356
It's above average. I'm not surprised at all that it can't do 4.7, 4.6 is still very good
>>53149381
Sure.
I may try playing around with the BCLK straps, to squeeze an extra MHz or two out of it.
>>53148482
(You) (You) (You) (You) CANCER CANCER CANCER CANCER MEMES EVERWHERE BUZZ!!!
This is normal.
Prime95 uses a form of raw AVX instructions that make intel chips ever since generation 1 Haswell heat up like mad. Use OCCT to stress intel stuff..
>>53141637
That's what you get for not buying the superior 5820K.
>>53149235
I run an 4.7GHz 4790K with a bequiet air cooler. In stress tests cores are ~65-70 except for one core that is at 80. Maybe I should redo the thermal paste? Can't be arsed to delid it though.
>memelake
>cormeme
flush yourself
>>53154257
It's fine as long as it stays below 90. And real world loads won't even heat it as much as stress tests.
