Today, I've become a believer with Heatkiller rev 3.0

[ontariotl]

Alright, I've been watching all this hub bub with this Heatkiller rev 3.0 and its dramatic drops in temperature compared to my trusty Swiftech Apogee GTZ. I fell for the hype when I upgraded from the GTX to GTZ. I didn't see that much improvement to even be at least satisfied with the upgrade.

I was quite happy with the GTZ when I upgraded to the Core i7 recently, and didnt think twice about upgrading. Then it hit, I come on the forums and find that Dazmode.com has them (HK 3.0) back in stock. I thought do I really want to spend another hundred bucks to upgrade? I keep hearing good things about it. A few days later, my wallet opened up and today I recieved my new toy (btw, thanks for the sale Daz!!).



While I was waiting for it to arrive, I though what the hell, lets do a test with my GTZ vs. HK 3.0. Others have done it, but I haven't seen too many screenshots with results to compare the two. Just a bunch of comments that say my temps dropped from 2 to 10 degrees etc. So I bought myself a digital thermometer to measure ambient temp. in case it changes during my removal of the GTZ and install of the HK 3.0. I want to make sure the results aren't skewed by a few degrees. I wanted to prove to myself this was worth the upgrade. Although, it could backfire too and have no improvements whatsoever. Either way, I wanted to share.

Ok, so as for my watercooling setup minus the cpu waterblock. I have the Thermochill PA 120.3 with push/pull config with recent upgraded (me and my upgrades) Scythe Ultra Kaze 2000RPM fans.
The rest of my setup include Swiftech MCP-655-B pump and Danger Den 4870x2 block.





Here is the GTZ before removal. Bye Bye.



Install of my new toy...I used a digital caliper to be precise with the recommended 19.6mm setting for the screws.











Alright, enough of the lollygaggin....on to the results.

I just put on Linx on for 20 runs as that gives a good run on the CPU for heat. I wasn't going for stress testing but just a quick look at temps. Don't need to run too long to find out how the temps are going to be once the CPU is 100% load. I used Real Temp to record the highest temp observed during the run. BTW, ambient temps were 26 degrees Celcius for all testing, so there are no skewed results. I know, my room is a little warm.

Here is the GTZ. Core i7 920@ 4.2Ghz, 1.28 vcore. HT on

76 74 72 68 I thought this wasn't bad for the GTZ and running 4.2Ghz

Ok, now the HK 3.0. Core i7 920@ 4.2Ghz, 1.28 vcore. HT on

68 66 67 62 I actually got giddy when I saw the results.

So now its time to push my Core i7 some more. I can run 4.5 easily with the GTZ but I wasn't too happy with the temps during stress testing. Of course maybe it had to do something with my warm room.

Alright, GTZ. Core i7 920@ 4.5Ghz, 1.38 vcore. HT on

92 89 88 83 Nothing to say really

Drumroll......HK 3.0 Core i7 920@ 4.5Ghz, 1.38 vcore. HT on

81 78 78 75 Holly crap, this block actually works.

Sorry for the long winded post, but I had to share what I experienced with the Heatkiller 3.0. I was a doubter, being disappointed about products before, but this thing dropped my jaw with the results. The GTZ is still a good waterblock, it cools the i7 beast pretty good. Just when you get over that 4.2 hump, it starts to struggle.

Looks like I have to prepare my sale of some older waterblocks including my GTZ both in 775 and 1366 flavors. Damn I even still have my GTX. Time for another long post

[miggs78]

Wow.. thanks for the results, it really shows why the Heatkiller block is worth the extra $$..

Now see that board was meant to be for the Heatkiller, that GTZ looked ugly on it.. :P

[JMCD]

That's one hell of an overclock too. You should see about pushing it even higher :)

[MpG]

Interesting, although I'm going to say that I'm 99% certain that you've got some mount variance at work there - while I've been finding the HK3 to be ahead of the GTZ as well, the blocks aren't THAT far apart in performance.

[ontariotl]

Originally Posted by MpG

Interesting, although I'm going to say that I'm 99% certain that you've got some mount variance at work there - while I've been finding the HK3 to be ahead of the GTZ as well, the blocks aren't THAT far apart in performance.

I thought so too, but don't forget the mount for the GTZ is easy to use and can only go so far eliminating the possiblity for too tight or too loose senario. Even the same amount of MX-2 paste was applied. Also for the record my cpu isn't lapped either.
At idle temps, they both seem to be about even, but at load, the HK3 pulls away from the GTZ.

[JMCD]

Originally Posted by ontariotl

I thought so too, but don't forget the mount for the GTZ is easy to use and can only go so far eliminating the possiblity for too tight or too loose senario. Even the same amount of MX-2 paste was applied. Also for the record my cpu isn't lapped either.
At idle temps, they both seem to be about even, but at load, the HK3 pulls away from the GTZ.

It looks like it pulls away even more as the temps get higher.

[gingerbee]

run it for 12 hours then tell me what the temps are? ( thank for all the info heat killers looking good trying not to buy one )

[stoanee]

Gah!....pulled the trigger and bought one. Will have a BNIB Swiftech C17 for sale as soon as I receive the Heatkiller. (As well as a Megahalem )

[ontariotl]

Originally Posted by gingerbee

run it for 12 hours then tell me what the temps are? ( thank for all the info heat killers looking good trying not to buy one )

Sure I can run it for 12 and give you the results. Just can't do the GTZ since its now retired.

[Dazmode]

This block look so good installed! I keep hypnotizing myself: "I like my Supreme", but it gets harder and harder for me. Damn you!