On Thu, 17 Aug 2000, Shridhar Daithankar wrote:
> I was shocked to see that the CPU fan was not connected. With your methods, I
> immediately rebooted the machine to test CPU temp. It was 60. I put machine off
> for 20 minutes to pull it below 35. And then it was much faster than original.
>
> with CPU fan and cabinet open. I had to put external ventilation.
>
> After 4.5 hours of compilation, not a single Sig11. and machine is way too fast
> for it's earlier performance. I wonder how I missed that. Anyway thanks a much.
> I,m posting this to LIH also.
yeah even i was surprised when i got mine to work this way. But no
external ventillation for me. And just look at the SPEED IMPROVEMENT oh
my!!!
In the day time it takes about 11 minutes to compile the kernel, and at
night just 5.5 minutes. I just cant believe this. Even the mmmx-checksum
speeds get a big boost.
Which prompts me to conjecture that the k6-2 500s are possibly
overclocked k6-2 300s.
Though my work around of (a) keeping cabinet open (b) compiling at night
works...it is not the best of the ways.
So could anyone suggest better cooling methods. I have a cpu fan which
blows air into the cpu...i doubt whether it should be the other way. And
any suggestion on availability of thermal conducting paste etc would be
very welcome.
And Shridhar isnt this a BIG relief. With the kernel not compiling
before, i was having nightmares...now sweet dreams only...would become
sweeter if i can cool the cpu better
sreangsu
p.s. A sadist that i am i bungled into an even more stringent hardware
health test than kernel compilations....making the comp play both sides of
gnuchess. In that case u have to wait much less for a sig11....if at all
----------------------------------------------
Find out more about this and other Linux India
mailing lists at http://lists.linux-india.org/
