From: Richard Fish <bigfish@asmallpond.org>
To: gentoo-user@lists.gentoo.org
Subject: Re: [gentoo-user] i386 vs amd64
Date: Fri, 21 Oct 2005 12:37:01 -0700 [thread overview]
Message-ID: <4359435D.6040507@asmallpond.org> (raw)
In-Reply-To: <43591F11.9000300@comcast.net>
Rob wrote:
>I don't want to start a 64bit vs 32 bit war, or a Windows versus *nix
>war, but it has been my experience so far that the fastest benchmarks
>for a highly computation intensive program written in Numeric Python
>came on my 3.5Ghz P4 laptop with hyperthreading- on Windows. Also,
>running the same program on an AMD Opteron gave me a slower speed no
>matter what OS I was using.
>
Did you recompile or install a 64-bit version of python for the
Opteron? If not, you are comparing a 32-bit processor doing 64-bit
computations using 32-bit instructions to a 64-bit processor doing
64-bit computations using 32-bit instructions, which is probably not
what you intended.
>I am baffled by the behavior. The only thing I can figure might be
>occuring would be that the *nix 64 bit toolchains are much younger than
>the 32 bit ones. But as the 32 bit Numeric Python on Windows is still
>3x faster than the *nix equivalents, I have asked Activestate, the
>Windows Python provider, if they do anything special when compiling the
>code and they say no. I think they said that they use some ordinary MS
>comiler.
>
>
Well, MS makes _very_ good compilers, from a speed standpoint. It's
difficult to find an objective comparision between the Visual C++
compiler and GCC, but it would not surprise me at all if the VC++
produced code that was 10-30% faster for many cases.
For example, VC++.NET can use "whole program optimization", where much
of the optimization is delayed until the linking step, when data from
all comilation units (.o files) can be used to make decisions. This
results in more inline functions, more unreachable code being deleted,
better function ordering, and so on.
As for being 3x faster on Windows, that seems a bit strange to me. Were
the "*nix" versions of python compiled specifically for the processor?
Running code 'optimized' for a 386 on a modern processor would account
for this difference in performance.
Note that the OS should make very little difference here. You could
probably do a similar comparison using the ActiveState python on Windows
vs the cygwin version of python.
-Richard
--
gentoo-user@gentoo.org mailing list
next prev parent reply other threads:[~2005-10-21 19:52 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-10-21 0:07 [gentoo-user] i386 vs amd64 Sean
2005-10-21 0:19 ` Peter Gordon
2005-10-21 4:19 ` Bob Sanders
2005-10-22 20:58 ` Sean
2005-10-23 2:34 ` Justin Patrin
2005-10-21 12:54 ` Scott Tiret
2005-10-21 15:46 ` Neil Bothwick
2005-10-21 17:02 ` Rob
2005-10-21 19:37 ` Richard Fish [this message]
2005-10-21 20:49 ` [gentoo-user] i386 vs amd64/ forget my Python comparisons Rob
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4359435D.6040507@asmallpond.org \
--to=bigfish@asmallpond.org \
--cc=gentoo-user@lists.gentoo.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox