Cooler Article

nightshade

Distinguished
Apr 5, 2002
7
0
18,510
I love the comparison articles for cpu coolers, they're very useful for choosing an important part of your system.

That being said, I noticed a couple problems with Harald Thon's article.

First, the caption for the Swiftech MCX C370 says "MCXC C370: its cooling performance matches that of its predecessor, the MCX370-0A."
However, the graphs are clearly very different, and on the high end processors, the MCX370-0A model is over 20 degrees cooler. Hardly a match.

The second issue was on the page for the Titan TTC-D5TB. In it, he says:

"A buzzer is integrated in the cord, which will sound an alarm if the fan breaks down. This feature won't do you much good if you're using an AMD CPU, because nobody in the world can shut down a computer before the CPU burns up."

I think he may be confused. Awhile back, THG did an article with a movie that showed AMD processors burning up after complete removal of the HSF. Remove the whole HSF assembly and... well, big huge surprise, the CPU dies. (On a side note, I was very impressed by how Intel cpu's handled the same situation).

This is a very different scenario from a fan failure. If the heat sink is still attached and the fan stops, you should have quite a bit of time to shut the system down once you hear the alarm. How much time? A minute or so? Just an educated guess. Heck, it may run for days without a fan depending on the airflow in your case. At any rate, it shoudn't be the instant death alluded to in this article.

ns
 
Ok, I felt bad not backing myself up, so I opened my case and stopped the fan with my finger. I held it there for a minute ten seconds before letting it go.

MBM went up one degree (56 - 57)and then to 58 a little after I let it start up, now back to 56.

I would have held it longer but I was more worried about damage to the fan itself :) Well, I don't have a backup cpu either.

Anyway, no instant death, and that buzzer would be useful afterall (although redundant, as your bios may warn you on fan stoppage and/or temp rise when it's slow)

Oh, by the way I am using a 266 fsb 1.33 ghz athlon, and a thermaltake volcano 6 cu (which I love, as it's both quiet and effective), ambient temp around 24 degrees C.
 
If I stop the fan the temp will jump from 30 to 65 in 1-2 minutes.
Ough!

PS. When the temp hit the 65° I enabled the fan again. Was gettin' to hot for me.

If I buy the boxed version of the "hammer", will I get a box of nails for free??<P ID="edit"><FONT SIZE=-1><EM>Edited by zeexen on 09/16/02 05:35 PM.</EM></FONT></P>
 
Wow, I never could get near 30, it always hovers around 55 or so, but rarely over 60, which isn't unusual for these cpus. I mean, if you look at the number in the article, they hit 70 quite often and call it "good". But heck, half the time this room is around 30 itself :) I used to use this vanta HSF with a screaming delta fan at 7k rpm, and I couldn't take the noise, and it still wasn't 30, even after lapping the bottom of it. So I commend you for your HSF-fu.
My dual celeron 300's oc'd to 450 run at 30ish though 😀

Anyway, that's my point, if you have 1-2 minutes, you can close everything down, write an email to your friends telling them you're having computer problems, play a couple games of minesweeper (on the easy setting) and shut down nicely. Far from the claim of "nobody in the world" being able to shut down before the cpu goes up in smoke.

Fan stop != HSF removal.
 
I mean, if you look at the number in the article, they hit 70 quite often and call it "good"
Read the article again. They werent using actual processors. So the 70 isn't a real world processor temperature. The article was meant to benchmark coolers under a stable/constant enviroment. You can't do that w/ a processor because the heat production fluctuates w/ its usage and loads. Load fluct. could give one cooler an advantage over another. blah blah blah....

[-peep-].
 
You guys are damn brave to actually touch and stop the CPU fan, as well as let one, which in your case being a HOT Tbird, run afterwards, despite the HS being on!

Anyways, to LED, I could not grasp what the CPU simulator was, but thanks for making it clear, I was unsure about why would they be running CPUs at over 70ºC and state that they are ok!
For a second I thought the actual AMD temps are this high, compared to on-die temps of P4s being in the low 40s!

--
Where did your THGC username come from and why did you choose it? <A HREF="http://forumz.tomshardware.com/community/modules.php?name=Forums&file=viewtopic&p=19957#19957" target="_new">Tell here!</A>
 
LED, ohh you are right, for some dumb reason I skipped that page. Thanks for the clarification.

Eden, actually come to think of it, I've broken an awful lot of stuff in the past... ;D