Just a thought that popped into my head

Richard rgollub at uninet.com.br
Tue May 7 04:13:23 PDT 2002


Ian Molton wrote:
> 
> have you run benchmarks? perhaps you might revert to 2.95.3...

	No, I haven't. As I said, for my intra-net the overall response is
essentially indistinguishable from the set-up compiled with 2.95.3. 
	It is obviously a qualitative judgement. But my feelings do not
register any significant degradation of response time, and, to me that
counts a lot! Believe me, I am _quite_ sensitive to time variations and
delays :)

	And to be fair, I am here comparing, although subjectively, like with
like: same softwares (diff versions of course), same machine, and same
compilation flags (just -O2, nothing else). Considering, for instance,
that latest Qt and KDE are quite resource consuming pieces of software,
the end result, by simplistic extrapolation, tends to point at least to
very similar likely benchmarks. If time differences exist in the tenths
of a second, I simply couldn't care less...

	In short: I am happy (so far), and see no reason to invest time in
checking which compiler does best, as long as, they produce a working
code, limited only by the inherent bugs of the program being compiled.
Even if 3.0.4 has intrinsec problems (and what program does not have
design/implementation flaws) they did not show up in the ordinary
packages I currently am dealing with, under ordinary use and stressing.

	Richard
-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message



More information about the lfs-dev mailing list