Help, over heating i4770k with H80 cooler.

Well mine will hit 80c at stock under a dual tower k2 cooler with x2 2350 rpm fans. The best way to stress theese chips is normal use. Theyre not designed with tests such as p95.

@the op, the 0x124 bsods your getting can be down to lack of vcore, or to low on chipset voltages such as ioa/d and system agent. The 1.7v reading your seeing is likely initial cpu input voltage. This is what the cpu uses when booting into windows. I have mine on 1.88v as any lower and it fails to boot.
 
Last edited:
Right, having played around with voltages, updated bios and performed a small overclock 4ghz with my own voltages etc, all seems stable now. Temps are ok, never going above 60 degrees after a 2 hr gaming session on Wolfenstein, It's a gaming machine so this is ok for me. I've never had so much hassle building a system before lol.

Thanks for the help guys.
 
BS! My 4770k runs around 25 deg on idle (although in Feb) and around 50 on full Prime load (although once again, it was February). I have a Seidon 120V; which I think is quite similar.


That is more like it :cool: I had to re-do my cooling twice when setting up my system. I used the line method, but what did it for me was that before bolting the block, I puched it to the CPU, moved it clockwise and anti-cockwise a bit (spread the TIM) AND WITHOUT RAISING IT (air bubbles can come into play here) I bolted the block down. ANYTHING can interfere with the contact. Fingerprints, dead skin cells, small hair, ANYTHING. I tried not to breathe when applying the TIM so as to reduce all the possible problems (didn't want to do it a 4th time...).

When I first did my cooling, I tried not to tighten the fittings as not to damage the CPU. The fittings have enough clearance that if you screw them all the way down, the CPU is still undamaged. Don't overdo it though.

Are the fans set to intake air or remove air from the case? Whatever they are set to, flip them over (mine are set to remove air from the case, but that is because they are at the top of the case).

EDIT: The reason the core speed changed is because the CPU won't need that much speed. On idle, you expect it to be 800MHz. On load, it will go to whatever it is capable of.

Prime/Intel burn test put excessive load on Haswell cpus causing silly temps in coretemp/hw monitor etc. When Haswell was released the issue was so apparent that reviewers were issued a Haswell testing kit that comprised of several tools such as Aida64. This is not BS it is recommended by Intel And other motherboard manufactures that Prime/IBT is not used to burn in these cpus. I have thoroughly tested over and over again. Even with a H80i with fans full blast the temps were getting up into the 90s even at stock with IBT/Prime.

I am sure that if you called Intel and told them that using a stock cooler on a 4770 and Running prime it gets to over 100 degrees they will tell you not to use Prime/IBT.

This is the primary reason why Asus put money into developing their realbench software.
 
Prime/Intel burn test put excessive load on Haswell cpus causing silly temps in coretemp/hw monitor etc. When Haswell was released the issue was so apparent that reviewers were issued a Haswell testing kit that comprised of several tools such as Aida64. This is not BS it is recommended by Intel And other motherboard manufactures that Prime/IBT is not used to burn in these cpus. I have thoroughly tested over and over again. Even with a H80i with fans full blast the temps were getting up into the 90s even at stock with IBT/Prime.

I am sure that if you called Intel and told them that using a stock cooler on a 4770 and Running prime it gets to over 100 degrees they will tell you not to use Prime/IBT.

This is the primary reason why Asus put money into developing their realbench software.

Its normal to have high temperatures with these chips. But not 90+ with no OC.
 
Yes, when using Prime/IBT. This is because the programs are putting an unrealistic load on the CPU. Most commercial manufacturers of systems no longer use those programs to burn in systems. I was puzzled when I found this out initially but its true. I did also notice that regardless of the voltage is set to these programs can cause a huge spike in voltage and temps. Quite often you will see 90+ then it backs off a little when the cooling kicks in.

Don't get me wrong, these programs are fine on Sandybridge/Sandy-E/Ivy/Ivy-E and do not cause the same issues (Unless the cooling is pants)
 
I'm not going to say you are wrong, because I'm not 100%, but many people like me do nt have this 'problem.' Can it be that if effects some chips only? I did prim yesterday. Went up to mid 60s and I think its like 30 deg in my room.
 
I'm not going to say you are wrong, because I'm not 100%, but many people like me do nt have this 'problem.' Can it be that if effects some chips only? I did prim yesterday. Went up to mid 60s and I think its like 30 deg in my room.

Yeah that's exactly what I thought. Maybe its that some batches are more prone to spiking or that a specific combination of motherboard/cpu batch stops the spiking/overvolting..
 
Also what version of prime is being used. Any of the later ones that use AVX instruction sets really push temps. And if you use adaptive or offset vcore, more voltage gets added on at load. Yes you may get lucky with a chip that hasn't been too badly pasted/lidded but recent batches have been shockingly bad. Hence in swapping to devils canyon 4790k. I could delid my current chip but as its volt hungry its not worth it as it wont clock higher.
 
Back
Top Bottom