--Supa 4 HeatSink Shoot-out!!--

[sieb]

Well, after spending a day with the AC on in my apartment, I have cooled it down to a brisk 25C. I have finally, after a couple months of waiting, received all of the candidates. We have the big names, Millennium Glaciator, Thermalright SK-6 (w/ YS-Tech), Swiftech SCX-370, and Swiftech MC-462 (W/ Denki). I didn’t want to get deep into the math stuff, we all know the specs on these HSFs by heart now, I strikly wanted to show some real world results since most of us don’t have a controlled lab to run our systems in.

[img="[URL]http://jason.siebert.com/extras/IMAGE114.JPG[/URL]"]
The 462 is on the computer during this taking (kept forgetting to take a family photo).

Test rig is my newly revamped Alienware with all hand chosen parts and I've tried my best to keep clutter to a minimum.

Antec SX-1030 Black
Abit KT7A-Raid
AMD 1.4GHz Tbird AYHJA
512MB Crucial Ram
2-30Gig Maxtor HDDs
DVD, Plextor CD-RW, Zip250
Windows 2000 SP2

[img="[URL]http://jason.siebert.com/extras/IMAGE104.JPG[/URL]"]

And I will be using a Digidoc 5 for the temp monitoring. I have a thermesistor trimmed and sitting next to the Processor touching the die, one wedged into the heat sink (between the fins if possible), and one inside the case monitoring inside temps. I will be using Sisoft Sandra Burn-in to get the temps up. I know there are better ways, but this usually does the job as my temps round out halfway through the test. I will do 10 runs with the burn in. First test will be done after boot up is finished and I am sitting at the login screen. I will give it about 10 minutes to stabilize, after that I will run Sandra, and then give it about 10 minutes to cool down when it’s done. I will time each test to see how long it takes to dissipate the heat and stabilize. Also, although I debated it, I chose to stick with Artic Silver 2 for the test since that’s what most people use now a days. Each heat sink was properly cleaned and AC2 properly applied.

Enough buildup, here is the schtuff…

Glaciator
Idle: 37C
Load: 48C
Cool off time: 3:32

SK-6
Idle: 39C
Load: 48C
Cool off time: 3:44

MCX-370
Idle: 37C
Load: 45C
Cool off time: 3:10

MC-462
Idle: 32C
Load: 42C
Cool off time: 2:50

Well, that’s that. My scores will probably differ from most but I hoped to get a nice comparison of the top 4. The actual heat sink temps were about one to two degrees higher than ambient, and about 4 degrees higher under load. Right now I have the MC-462 on, and probably permanently. Like I said, there are enough reviews of each heat sink and all its specs, and I decided to do a nice quick test of the top 4 contenders. Overall, these are quite nice numbers for a 1.4. I wouldn’t recommend the Swiftechs unless you have the money, or need the cooling of it. The Glaciator and SK-6 stuck close together for the most part, SK-6 w/ a Delta would have probably done better, but I don’t like Deltas even if they do rock. So for the low end, either SK-6 or Glaciator will do fine, but I lean more towards the SK-6 for its nicer construction. Alright class, end of today’s lesson.

 

[Thelemac]

You seem to have forgotten to mention the ambient temps in the stats...those would make this really qute a good study! Thanks for this.

 

[sieb]

Thanks, but in the beginning, I stated the Ambient temp was set to 25C by leaving the AC on all day. (Oy, power bills going to be bad...)

 

[Thelemac]

Oh, sorry, didn't notice that.

Do you have a measurement of the ambient right around the hsfs? Cause it does, after all, vary throughout the room, and over time. Not to be picky, but I would really like to see that!

 

[k_lined4lyf]

Do you have a measurement of the ambient right around the hsfs? Cause it does, after all, vary throughout the room

Yeah, I'm curious about this too. Sieb, you said you have "it" cooled down to 25C. By "it," do you mean your apartment room temp (I did notice you said you have a temp probe in your case, but I seek only a little clarity)? If so, the ambient temp IMHO, should be the temp of the air that makes contact with the HSF. After all, the temp of the air at the HSF intake is what (in part) affects the efficency of heat transfer.

Thanks for shelling out the cash and getting some real world results =]