On Sat, 20 Sep 2003, Jaromir Dolecek wrote:

> Simon Burge wrote:
> > 	1 CPU -j2 4056.016u  858.126s 1:19:57.97 102.4%
> > 	2 CPU -j2 4245.934u 1283.933s   56:41.13 162.5%
>
> 25 minutes difference is cool, IMHO. 30% gain looks fine.
>
> The difference would probably be even higher if the
> machine would run some CPU-intensive userland-mostly
> stuff, like some number-crunching.

That's "-j2" with two CPU's vs. "-j2" with one CPU. You
expect THAT to buy you something. The hyperthreading doesn't
seem to make much of a difference for build.sh, but yes, I'd
like to see what it does for CPU intensive stuff. I'll take
some timings next time I do some video transcoding.

Frederick