Is something wrong with my processor?

[flexy]

I am sort of puzzled you worry about temps and all that and you're not even overclocking. I want to have your problems Put 1.25V++ in those CPUs and start overclocking, THEN you have stuff to worry. PS. if you don't even intend to overclock I'd think the Noctua being overkill..but that's your decision

 

[sapped]

@yuriman
Alright so it should just stay around 80 for those heavy stress test tool, got it thanks!
Although which version of Prime did you meant? Since a few mentioned that the latest version are not recommended to run on haswell.

@elfenix
I see, was just hoping to determine whether the components are running as they should in store. It's the load temp which I'm worried about.

@flexy
No I do plan on overclocking mate, but I can't just do it directly in the store,
still need to learn how to do that.
Didn't you mention 4.7GHz might be possible with the devil's canyon? So a good quality cooler has better chance to stabilize at that frequency no?

 

[xthetenth]

I am sort of puzzled you worry about temps and all that and you're not even overclocking. I want to have your problems Put 1.25V++ in those CPUs and start overclocking, THEN you have stuff to worry. PS. if you don't even intend to overclock I'd think the Noctua being overkill..but that's your decision

Hey the Noctua D15 is a great way to quietly keep a 4790K under 40 most of the time and under 50 in games that stress it.

 

[Yuriman]

80c in the latest Prime.

The reason many are not recommending Prime95 for stress testing is because it utilizes AVX instructions, and whenever Haswell chips use AVX a large amount of voltage is automatically added, so it's not really very indicative of whether the CPU will be stable in non-AVX loads. Also, because of the large voltage bump, temperatures will be significantly higher in programs that utilize those instructions.

IMO, it's not a bad idea to do stability and stress testing with both AVX and non-AVX stress tests.

 

[Ketchup]

Hey the Noctua D15 is a great way to quietly keep a 4790K under 40 most of the time and under 50 in games that stress it.

50 in games that stress it? that would be nice, but not this guy. The 4790 is a hottie. Some games can keep the temps in the 60s, but only because they are not stressing the CPU.

sapped, if you want to overclock very much, you need to go water.

 

[flexy]

@yuriman
@flexy
Didn't you mention 4.7GHz might be possible with the devil's canyon? So a good quality cooler has better chance to stabilize at that frequency no?

POSSIBLE since DV *should* not run as hot like the old Haswell (like mine , but of course it's still Haswell....so it WILL get hot...but you might have a better chance to *maybe* get to 4.7. However I would first aim low, 4.5, 4.6 and then see whether it does 4.7. (Of course this might need significant tweaking in BIOS as well, on the other hand you could be lucky and your chip does 4.7 just by increasing VCore..who knows

 

[sapped]

Hey guys, thanks again for all your comments on the thread, appreciate it.

flexy mentioned that the Noctua NH-D15 might be an overkill for my CPU, and I've been
thinking about this for a some time.

How do you guys think the Noctua NH-U12S is?
http://www.noctua.at/main.php?show=productview&products_id=63&lng=en

It's the second highest rated CPU cooler on PCPartPicker, great reviews too while being cheaper.
Another great thing is it's fully compatible with my tall RAM.

Although the NH-D15 is not that bad, since I can remove the top heat spreader from my Trident X,
(will this cause heating problem with my RAM?) like it's shown here:

http://www.tweaktown.com/reviews/6313/noctua-nh-d15-cpu-cooler-review/index5.html

But it still sits at least 20mm taller, with the front fan not blowing directly into the heat pipe.

I'm wondering will this be much of a problem to heat dissipation? Since the heat pipes are where
it's supposed to be the hottest for transferring heat right? So with no direct wind blowing into
it, will this cause much issue?


Also some other general questions regarding the cooler:

1. What if I attach the fans at the middle and back of the heat sink? Will this be just as cool as
attaching to the front and middle? Since I have space at the back side.

2. Will running the NH-D15 in single fan mode (mid only) still be much cooler than the NH-U12S,
or will it be practically the same or even worse?

3. And lastly the important question, will the NH-U12S allow my 4790k to be overclocked to 4.7GHz
and stabilize?

 

[myocardia]

If you're going to buy an aftermarket heatsink, and especially if you're going to spend the money to have someone else install it for you, I would recommend the Noctua NH-15. It's more than double the heatsink of the NH-12S, and will easily fit into your case, since it's 'only' 160mm tall, and your case allows heatsinks up to 180mm. Also, your RAM will be much cooler, if you remove the extra heatsinks from it, and have the fan right over the top of them, pulling cool air over the RAM (with the fan closest to the front of your case, instead of having it pulling, from the rear).

To answer your questions, 1) Putting the 'outside' fan pulling air, instead of pushing it, won't make more than a 1-2°C difference, at most. That all depends on the case it's installed into, of course. 2) Yes, even in single-fan mode, the NH-15 will still blow away the NH-12S. It's more than twice as much heatsink. 3) No heatsink is a guarantee of how high your CPU will overclock. You'll likely be limited by heat, though, if you buy the NH-12S, instead of the NH-15.

 

[myocardia]

Oh, I forgot to tell you about my temps, with my 4790k, and a Noctua NH-D14. When running the latest version of Prime95, and having it run small FFTs with AVX2 instructions enabled, my hottest core is 70°C after 8 or 10 minutes. That is with only the center fan installed on the heatsink.

I have a case with a 200mm fan directly over the CPU, though, and I have the heatsink turned 90°, so it blows upward, toward that huge fan, so the video card keeps me from being able to use the second fan.

 

[Ketchup]

Hello myocardia, what was your top CPU speed when you were running the Prime95 test? Have you ever run OCCT? I would be curious to see if one pushed it hotter than the other.

 

[myocardia]

Hello myocardia, what was your top CPU speed when you were running the Prime95 test? Have you ever run OCCT? I would be curious to see if one pushed it hotter than the other.

Ooops, forgot that info, didn't I? I was running all four cores @ 4.4 Ghz, with HT on. I've never run OCCT on this system. I wasn't actually stress-testing it, I was just checking out how well the NH-D14 did it's job.