• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Help with temps 4790k

Associate
Joined
21 Jul 2010
Posts
829
Hey all

i would post in 4790k thread but my issues have been saturated a little with other replies

Ok so

my spec is 4790k (stock) / gigabyte gaming 7 / H60 cooler / nzxt Phantom case / 750W PSU

i am running into low 90s temps when stress testing meaning no room for overclock (not that i mind i am happy it running at 4.4 if the temps are sensible when cpu is pushed)


what i've done

1.set vcore to 1.2163 (seems to stress test ok on this without one of the workers erroring)
2. set fan profile for the pump from full speed and even tried manual to full power
3. i do have XMP enabled and seems to run flawlessly (temps where the same before)
4.reseated the h60 and even moved the pumb control to another connector just incase

so basically

i honestly dont know what else to do! - could it be i have a hot running chip? or the h60 isnt doing an adequate job or is my board over volting in some way shape or form :(
 
Last edited:
Hi mate ,

Does seem a little strange. As most clock up ok. First thing I thought of was cooler seating ? But you said you have checked that.

What are your idle temps like ? And are you using push pull fan setup and with which fans ?
 
Hi mate ,

Does seem a little strange. As most clock up ok. First thing I thought of was cooler seating ? But you said you have checked that.

What are your idle temps like ? And are you using push pull fan setup and with which fans ?

ahh now thats the only next step i can think of i am using corsair fan supplied with cooler and another fan that isnt like noctua or anything just an nzxt branded

idle temps when it clocks to 800mhz is 27c when it goes full power to 120watts i think it shows at 4.4ghz thats where it goes to 94c
 
ahh now thats the only next step i can think of i am using corsair fan supplied with cooler and another fan that isnt like noctua or anything just an nzxt branded

idle temps when it clocks to 800mhz is 27c when it goes full power to 120watts i think it shows at 4.4ghz thats where it goes to 94c

Hmmmm 27 dregs is good, as expected. :)

So seems seated correctly. Which thermal paste did you use when installing.

The fans won't make massive difference, maybe bring under 90 though.

I saw a h60 review with a 3770k clocked at 4.6ghz. It reached 83 on hottest core. Ivy is much cooler than haswell. It may just be the h60 is struggling when stressing. Thus is one of the reasons I don't use stress test program's.
 
Hmmmm 27 dregs is good, as expected. :)

So seems seated correctly. Which thermal paste did you use when installing.

The fans won't make massive difference, maybe bring under 90 though.

I saw a h60 review with a 3770k clocked at 4.6ghz. It reached 83 on hottest core. Ivy is much cooler than haswell. It may just be the h60 is struggling when stressing. Thus is one of the reasons I don't use stress test program's.

used some arctic silver 5

i am not a fan of stress tests myself, and i know that under normal conditions the cpu wont be pushed that far, well if i have checked everything that logically can be a potential issue then there isnt a lot more i can do

perhaps i was just expecting a little better temps with the h60 and some room for overclocking in the future
 
It will be down to the stress test program using AVX, I get into the 90's as well at stock if I use the latest version of p95. Haswell chips simply cant cope with this type of test due to the way the die is pasted instead of soldered.
 
It will be down to the stress test program using AVX, I get into the 90's as well at stock if I use the latest version of p95. Haswell chips simply cant cope with this type of test due to the way the die is pasted instead of soldered.

ahh see thats interesting because it is prime95 i was using
 
You don't want to use the latest version of prime. Apparently it has some new CPU instructions that ridiculously max out the cores. It will take it into the max temperature range. But it uses un realistic instructions that no program now days will ever use. If you want to use prime use the older version.

Try running a cinebench r15 benchmark and asus real bench and see how temps do then.
 
well..

i feel stupid but if i had of known about latest version of prime95 T_T

so i ran cinebench r15 and it held at 76c

so i think that resolves my issue..


thanks for all the help and advice
 
I see you mention reseating the cooler but have you checked the backplate is fitted the correct way round?

I have a thread here which highlights my issue and resolution of high temps, albeit with a 4670k overclocked from 3.4Ghz to 4.2Ghz.

I've now been able to bench Prime95v279 at 69 degrees on the hottest core. Yesterday that same bench was 86 degrees, and it was hotter here when I ran today's test.

Tried the Haswell toaster Prime95v285 and the highest core was 79 degrees.

And running Cinebench R15 the highest temperature reached was 60 degrees on one core, the other three remained sub 60.

All on the 4.2Ghz overclock and effectively improving my Prime benches by 17 degrees.

Don't know if that helps your particular situation any, other than to say I hope it is something that another look at the cooler seating can address.
 
well..

i feel stupid but if i had of known about latest version of prime95 T_T

so i ran cinebench r15 and it held at 76c

so i think that resolves my issue..


thanks for all the help and advice

That's a good start. :)

Try asus realbench, it's more strenuous and if you pass that to me your real world stable.

Regards
 
I see you mention reseating the cooler but have you checked the backplate is fitted the correct way round?

I have a thread here which highlights my issue and resolution of high temps, albeit with a 4670k overclocked from 3.4Ghz to 4.2Ghz.

I've now been able to bench Prime95v279 at 69 degrees on the hottest core. Yesterday that same bench was 86 degrees, and it was hotter here when I ran today's test.

Tried the Haswell toaster Prime95v285 and the highest core was 79 degrees.

And running Cinebench R15 the highest temperature reached was 60 degrees on one core, the other three remained sub 60.

All on the 4.2Ghz overclock and effectively improving my Prime benches by 17 degrees.

Don't know if that helps your particular situation any, other than to say I hope it is something that another look at the cooler seating can address.


This is a good point too. Had a similar issue when fitting my old h80 and my new h100i cooler. Seating it correctly has a massive effect on cooling temp. Mine now idles at 22-28 degs :p

Though it really boils down to the silicon lottery again. The better yours is done , the more you'll be able to torture your CPU with lower temps. Not all will be this lucky.

Mct
 
Well yes some of these chips just run HOT at stock. My L33 batch will go 95-100 with a Hyper 212 EVO (not a great cooler but this shouldnt happen @ stock settings at 1.152vcore imo.)

I've reseated it multiple times always using MX-4. Probably not the greatest thermal paste around but that's what I have lying around. I will get a replacement chip somewhere next week and gonna fit a Be Quiet Dark Rock Pro 3 on it this time. I hope I will be able to get somewhere with the other chip and the giant CPU cooler. :)
 
well i certainly feel like a noob you was correct in querying the installation of the bracket neostuey i have now corrected and showing much better temps

i blame the rubbish instructions not clearly showing the mounting position haha

maximum under load is 74c lowest 24c!

thanks for all the help and advice guys :) !
 
Last edited:
well i certainly feel like a noob you was correct in querying the installation of the bracket neostuey i have now corrected and showing much better temps

i blame the rubbish instructions not clearly showing the mounting position haha

maximum under load is 74c lowest 24c!

thanks for all the help and advice guys :) !

No problem, I'm glad it worked out for you. It seems like it's definitely something that can be overlooked and easy to forget to recheck.
 
Back
Top Bottom