Originally Posted by tw33k
ummm..did you actually click the graph?
The results make perfect sense. The UK 3000s are doing all the work making the push fans redundant
In other words, just push or pull for an H100 is fine. However, on a thicker rad push/pull does have an advantage.
Originally Posted by Markerx
is it wise to get a h100 on sale for 75 now or wait for the coolermaster one to come out?
Lol, you're asking in the Hydro thread? Of course, go with the H100...lol...seriously though, I haven't seen anything on the new CM unit, although one or two people have mentioned it. Do you have a link to share? I have seen some of their older units, and I would easily prefer an H100.
Originally Posted by HemiRoR
Are these good temps for a 4.6 OC? I'm a first timer so I don't know if I used too much voltage as well :S This is an H100 on an i5 2500K running push pull stock corsair fans.
Originally Posted by nonomos
Looks good to me, but at what speed are your fans running?
Heres my clock/temps, at approx. 4,0.
Those temps are super low. What are you guys using to load the system for the temp tests? I don't see anything in your pictures indicating what percentage your CPUs are running at. Without this data your temps are pretty meaningless information.
Include in your screenshots Prime 95 running for a half hour, or even better Intel Burn Test running at half an hour. That software will fully load your CPUs and really give you an idea of how good the temps are. Watch out with Intel Burn Test though because it really tortures the system. Keep your eye on the temps while it's running.
Running Intel Burn Test for a half hour, with my system also running Furmark to add additional heat from the video cards, my H100 cooled i5 running at 4.3 GHz will hit 70c. My target for this processor was to keep the cpu under 73c using this software.
I agree with @TomcatV.
Many people are posting temps and tests, but not giving the relevant data or not using any standard software like Furmark, Prime95 or Intel Burn Test. Without doing this, and providing an ambient temperature and idle temperature, there's really no point posting at all.
When I torture test my system, I run CPU monitoring software, GPU monitoring software, CPUz to show the frequency the CPU is running at, and both (Prime95 and Furmark) or (Intel Burn Test and Frumark) to provide a load for the load test. I should note that I stress test my entire system, so that's why I add Furmark in there. Testing without Furmark is valid for CPU only testing.
As an example, here is a picture of a test I did a while ago. I use two monitors so I can run a full screen Furmark to generate gpu heat, and I can display the temp monitoring and other software on the second monitor:
That image was testing my older Core 2 Quad which was overclocked to 3.3GHz from 2.5GHz. However, that's the way I always stress test my system although now I usually use Intel Burn Test over Prime 95 because IBT really beats the crap
out of your CPU.
The point is to have a maximum load for temperature tests, and an Idle test, and compare the temps between the two making sure the ambient is the same for both tests, or at least taking it into account.
So, unfortunately, without knowing your ambient, and what you were running, I cannot compare your i5 temps to mine.
Originally Posted by HemiRoR
Oh, sorry. Forgot to say that those were my temps with prime running full blast in the background if that's what you are confused about? Here they are at idle, And one of them full load for 8 or so hours. The other two corsair fans are actually from my C70, not sure if they are the same model as those that come with the H100 (Although they look identical) They seem to be doing the job for now. Ultimately I'd like to buy 4xSp120 High performance corsair fans. The pump controller is set to full blast, but, if I change it to the quiet setting my fan speeds don't seem to change at all? No audible differences either. The TIM I used was this guy: http://www.newegg.ca/Product/Product.aspx?Item=N82E16835100017
I burned it in for a 24-hour session a few days ago and all seemed to be quite stable so far. What should the BLCK really be at optimally for a clock like this? (Excuse my noobiness to this :3) Think I could probably bump my voltage down a wee bit then as well? If you'd like I'd be glad to provide a picture of how airflow is setup inside my rig. As for the ambient temp, it can really vary. I'm in the basement, so the fireplace goes on and off occasionally. Temp ranges between 58-71F mostly. The temp for idle levels at about 18-19c once running for a while. The screeny below was after boot-up.
Warning: Spoiler! (Click to show)
That's a lot better indicator. Those temps are quite low for a loaded i5. Try running IBT for a half hour and see if there's a difference. Just FYI, the TIM on the H80/100 is actually excellent TIM. Corsair claims it's as good as Shin Etsu TIM.Edited by Mergatroid - 10/11/12 at 5:21pm