Re: CPU specific/optimized Debian builds ?
On Thu, May 30, 2002 at 11:27:50PM +1200, James Kahn wrote:
> Harder, if not impossible to get real results. Tests would either be
> subjective - "It just *feels* faster" - or have a high signal to noise
> ratio.
Not really. You create a benchmark of actual applications and compare
the results to the non-optimized applications. You run the tests a
number of times and see what happens. This isn't unheard of--a lot of
windows benchmarks, for example, actually run MS office apps, or video
games, or whatever--which is much more useful for most people than
reporting how fast a given machine can do fourier transforms. If the
results are only noise, well, I guess that answers the question. :)
--
Mike Stone
--
To UNSUBSCRIBE, email to debian-devel-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Reply to: