Amazingly good cooling for Air

Chris19 · 20 Mar 2006 at 00:33

I always thought that voltages up around 1.7 was for phase but i deceded to have a crack at it myself and to my surprise it actully runs fine.

I have only ran prime 95 on Large TFT for about 10mins but still the cpu only gets to 50 max, It cant be an faulty reading as i have it set to auto shut down at 65oC. Would it be ok to keep it as these votlages for air?

Case is an CM Stacker, with an freezer 64 pro

ajgoodfellow · 20 Mar 2006 at 00:36

The problem is not only heat but electron tunnelling

This occurs at high voltages and can kill the chip

1.7V seems rather high for a 2.7GHz overclock IMHO - does it need to be that high?

Chris19 · 20 Mar 2006 at 00:39

It doesnt need to be that high was fine at 1.58 but just wanted to see what happens at higher volts

Explicit · 20 Mar 2006 at 00:43

Drop the volts back to what they were. It's a bit silly to increase the volts just for the sake of it.

Chris19 · 20 Mar 2006 at 00:46

I am gonna up the fsb now to see how high it can go i sorta done it the wrong way round

Explicit · 20 Mar 2006 at 01:01

Chris19 said:
I am gonna up the fsb now to see how high it can go i sorta done it the wrong way round

Better yet: drop the volts back 1.58 > increase the FSB > test for stability and if that fails then, and only then, increase the volts

King_Boru · 20 Mar 2006 at 08:17

I can see OcUK earning another £190~ in a few days when your chip goes kaput.

ted34 · 20 Mar 2006 at 18:50

Whether it will kill his chip or not those temps seem really good for those volts especially seeing its a dual core. It could be faulty temp sensor or maybe thermal throttling (but doesnt that show fluctuations in cpuz)

Mr Mister · 20 Mar 2006 at 18:56

dude if it shows 1.728v in cpuz it will be more like 1.75v actually going to it!

King_Boru · 20 Mar 2006 at 19:21

ted34 said:
Whether it will kill his chip or not those temps seem really good for those volts especially seeing its a dual core. It could be faulty temp sensor or maybe thermal throttling (but doesnt that show fluctuations in cpuz)

chances are the temp sensor is incorrect, you cant really trust em'

sgoaty · 20 Mar 2006 at 20:04

You are slightly mad but im impressed

Jokester · 20 Mar 2006 at 20:07

Run SnM on it.

Disclaimer:- I accept no responsibility if it kills your chip.

Jokester

King_Boru · 20 Mar 2006 at 20:09

haha, I remember Jokester advising me that S&M would really heat test my CPU,... he was right!

ToLearn · 21 Mar 2006 at 03:03

King_Boru said:
haha, I remember Jokester advising me that S&M would really heat test my CPU,... he was right!

someone remind me not to put S&M into google again when looking for this program comes up with all manner of things

Yewen · 21 Mar 2006 at 04:20

ToLearn said:
someone remind me not to put S&M into google again when looking for this program comes up with all manner of things

HAHA!

Classic, stick Stress Test into the string for better results.

Chris19 · 21 Mar 2006 at 08:34

Meh its back down to 1.58 for now... :rolleyes:

xibxang · 21 Mar 2006 at 09:07

I think my S&M takes pity on my system. With all the sliders set to max, my CPU temp doesn't get above 43DegC. I'm willing to accept that mt temp sensor reads a bit low, but it doesn't half mock me when I start to think, "Hmm, maybe I could get another few mhz out of that chip....."

Bennah · 21 Mar 2006 at 09:12

Atlast someone putting some real voltage into there A64 when its on air. I did this all the time when I had my A64's. Havnt killed a cpu to date. Didnt kill the memory controller or nothing...

Get it back up to 1.7v+

megatron · 21 Mar 2006 at 11:27

I run my a64venice/opteron on 1.6v, ranan xp2500mobile for 1.5 year of 2v watercooling.

PinkFloyd · 21 Mar 2006 at 11:40

I've always been afraid of S&M since I ran it on my old XP2100 system, the power supply burnt out, didn't do the motherboard much good. Wasn't overclocked or anything.
I'm paranoid of the thing....