thermaltake watercooling build

Soldato
Joined
25 Feb 2004
Posts
9,393
Location
Heckling for change
locutus12 said:
i did it on the cheap with thermaltake (£30 from MM) and shaved ten degrees off my system and cpu temps... hardly a failure now is it... :rolleyes:

I don't know! Maybe you're CPU was at 60c load before? :rolleyes: Maybe your temps are inacurate! It wouldn't be the first time that's happened here!
 
Wise Guy
Soldato
Joined
18 Jan 2003
Posts
5,384
Location
Hatton
weescott said:
I don't know! Maybe you're CPU was at 60c load before? :rolleyes: Maybe your temps are inacurate! It wouldn't be the first time that's happened here!

One again....

[RedIndain]Wee man speak truth![/RedIndian]

It's wasn't uncomman for reported temps to RISE on older socket A mobbos because of the heat build up caused by the reduced airflow around the socket, but still give hugely better overclocks. My old watercooled Abit NF7 used to drop over 10°C if i pointed a fan in the general direction of the socket.

You can only use mobbo reported temps as a guide, i would be my left testicle that most are not reporting anything LIKE accurately.
 
Wise Guy
Soldato
Joined
18 Jan 2003
Posts
5,384
Location
Hatton
Mikey1280 said:
Thermal probes 4tw, the only accurate way of telling whether temps have gone up or down

More accurate than a mobbo yes, but 100% accurate no.

There are to many variables even with thermal probes, where they are mounted, how they are mounted, what TC you used, the amount of airflow over the probe, the quality and accuracy of the probe, the quality and accuracy of the reader etc.

The most accurate method of reading CPU temps is probably from the on die thermal diode, but once again, mobbos can't/don't read them worth a dirty great turd and it's a mighty great pain in the butt cheeks to make and calibrate a setup to read them (although it has been done).
 
Soldato
Joined
25 Feb 2004
Posts
9,393
Location
Heckling for change
It's a bit Anal (am I aloud to say that?) but, the most accurate way to test is with a multi meter tested with ice, and then next to put the probe next to the IHS (not on top), but even then it's not 100% accurate.

I test my thermal probe with body temperature. :eek:
 
Back
Top Bottom