First Review of Intel's new Heatsink for its 6 core CPU

Cooling Processors quietly

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

First Review of Intel's new Heatsink for its 6 core CPU

Post by ces » Sat Apr 17, 2010 6:05 am

First Review of Intel's new Mini-Tower Heatsink for its 6 core CPU

A three way comparison Review of
the new Scythe Yasya cpu heatsink,
Intel's new mini-tower heatsink for its 6 core cpu, and
the Prolimatech Megahalem

http://www.overclockers.com/scythe-yasy ... er-review/

swivelguy2
Posts: 404
Joined: Sun Jul 04, 2004 9:18 pm
Location: Illinois, USA

Post by swivelguy2 » Sat Apr 17, 2010 11:16 am

Except not. The reviewer saw 100C and aborted the test, including no data from the Intel HSF. He could have:

- trusted Intel not to provide processors with dangerously insufficient cooling

- trusted the built-in temperature failsafes to kill the processor before any damage was done

- used a lower load test so that he could actually get a comparison between the stock HSF and the aftermarket options tested

- included the incomplete data set that he did collect from the Intel product

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Post by ces » Sat Apr 17, 2010 2:14 pm

swivelguy2 wrote:Except not. The reviewer saw 100C and aborted the test, including no data from the Intel HSF. He could have:
- trusted Intel not to provide processors with dangerously insufficient cooling
- trusted the built-in temperature failsafes to kill the processor before any damage was done
- used a lower load test so that he could actually get a comparison between the stock HSF and the aftermarket options tested
- included the incomplete data set that he did collect from the Intel product
I think 100C is a sufficiently informative data point for me. It certainly is not what I had expected.

swivelguy2
Posts: 404
Joined: Sun Jul 04, 2004 9:18 pm
Location: Illinois, USA

Post by swivelguy2 » Sat Apr 17, 2010 2:50 pm

Okay, let's say that the final temp for the Intel cooler would have been 110C (which is probably pessimistic). That means it's performing 69% as well as a megahalems (90C over ambient vs 62C over ambient).

That's not bad for something with 75% of the weight, 52% of the volume, and 23% of the price (based on the one currently available for a buy-it-now of $15 on ebay) of the megahalems.

Now I'm not saying it's going to be quiet, or a good idea to use, I'm just saying we can't write it off immediately just because it hit 100C. I'm sure the 980X is binned very selectively to withstand high temps, and also sure Intel won't ship a chip with a heatsink that can't keep it alive. People must run it at 100% cpu with the stock fan, and Intel must account for that.

btw, here's a REAL review including this heatsink, which shows it performing 82% as well as the megahalems when both are using the same fan at full speed:

http://benchmarkreviews.com/index.php?o ... &Itemid=62

ces
Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Post by ces » Sat Apr 17, 2010 3:22 pm

swivelguy2 wrote: btw, here's a REAL review including this heatsink, which shows it performing 82% as well as the megahalems when both are using the same fan at full speed:
http://benchmarkreviews.com/index.php?o ... &Itemid=62
That is a very interesting review of the new stock Intel cooler. It seems to be a decent performer with the stock fan.

My personal observation is that there seems to be a lot of variability in comparative benchmarks on CPU coolers from review to review. But after a number of reviews are done, you can sort of develop a general sense of where a cooler generally ranks among its peers.

This review gives me hope that the Intel stock cooler might end up being decent performer. I certainly like its size.

Post Reply