From: "John R. Strohm" <strohm@airmail.net>
Subject: Re: Is ther any sense in *= and matrices?
Date: Mon, 9 Jun 2003 00:17:24 -0500
Date: 2003-06-09T00:17:24-05:00 [thread overview]
Message-ID: <bc15fh$ccu@library2.airnews.net> (raw)
In-Reply-To: bebbba07.0306082027.2c50eb6b@posting.google.com
X-A-Notice: References line has been trimed due to 512 byte limitation
Abuse-Reports-To: abuse at airmail.net to report improper postings
NNTP-Proxy-Relay: library2.airnews.net
NNTP-Posting-Time: Mon, 09 Jun 2003 00:20:49 -0500 (CDT)
NNTP-Posting-Host: !\p8l1k-V\n`(V3 (Encoded at Airnews!)
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2800.1106
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106
"Russ" <18k11tm001@sneakemail.com> wrote in message
news:bebbba07.0306082027.2c50eb6b@posting.google.com...
> "Bobby D. Bryant" <bdbryant@mail.utexas.edu> wrote in message
news:<pan.2003.06.08.22.56.37.297066@mail.utexas.edu>...
> > On Sun, 08 Jun 2003 11:51:31 -0700, Russ wrote:
> >
> > > "John R. Strohm" <strohm@airmail.net> wrote in message
> > > news:<bbumef$fnq@library2.airnews.net>...
> >
> > >> With all due respect, ladies and gentlemen, it has been known for a
> > >> very long time that the difference in "efficiency" between A := A + B
> > >> and A += B is lost in the noise floor compared to the improvements
that
> > >> can be gotten by improving the algorithms involved.
> > >
> > > Oh, really? I just did a test in C++ with 3x3 matrices. I added them
> > > together 10,000,000 times using "+", then "+=". The "+=" version took
> > > about 19 seconds, and the "+" version took about 55 seconds. That's
just
> > > shy of a factor of 3, folks. If that's your "noise floor," I can't
help
> > > wonder what kind of "algorithms" you are dealing with!
> >
> > I'm just curious why the compiler didn't generate the same code for both
> > versions.
>
> The compiler didn't generate the code. I did.
Why?
If you are hand-generating assembly code for a PowerPC processor, I would
bet a fair number of doughnuts that your cache utilization is shaky at best,
and THAT is what is generating your factor of 3 difference.
next prev parent reply other threads:[~2003-06-09 5:17 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-06-05 10:45 Is ther any sense in *= and matrices? Preben Randhol
2003-06-05 11:13 ` Vinzent Hoefler
2003-06-05 11:28 ` Preben Randhol
2003-06-05 11:53 ` Vinzent Hoefler
2003-06-05 15:27 ` Preben Randhol
2003-06-05 15:40 ` Vinzent Hoefler
2003-06-05 15:47 ` Preben Randhol
2003-06-05 16:38 ` Vinzent Hoefler
2003-06-05 17:16 ` Preben Randhol
2003-06-05 17:17 ` Preben Randhol
2003-06-05 17:59 ` Vinzent Hoefler
2003-06-07 19:38 ` Russ
2003-06-08 6:46 ` John R. Strohm
2003-06-08 18:51 ` Russ
2003-06-08 20:52 ` tmoran
2003-06-09 4:24 ` Russ
2003-06-09 5:13 ` John R. Strohm
2003-06-10 9:38 ` Ole-Hjalmar Kristensen
2003-06-10 16:11 ` Wesley Groleau
2003-06-10 19:24 ` Ole Kristensen
2003-06-10 18:33 ` Russ
2003-06-10 23:16 ` John R. Strohm
2003-06-09 6:58 ` tmoran
2003-06-08 22:23 ` John R. Strohm
2003-06-09 6:06 ` Russ
2003-06-09 10:06 ` Mango Jones
2003-06-08 22:56 ` Bobby D. Bryant
2003-06-09 4:27 ` Russ
2003-06-09 5:17 ` John R. Strohm [this message]
2003-06-09 14:53 ` Bobby D. Bryant
2003-06-09 17:46 ` Russ
2003-06-10 9:57 ` Ole-Hjalmar Kristensen
2003-06-05 12:33 ` John R. Strohm
2003-06-05 19:25 ` Wesley Groleau
2003-06-05 20:17 ` David C. Hoos
2003-06-05 20:52 ` Wesley Groleau
-- strict thread matches above, loose matches on Subject: below --
2003-06-10 19:00 tmoran
2003-06-10 19:37 ` Ole Kristensen
2003-06-10 19:37 ` Ole Kristensen
2003-06-10 19:48 ` Ole Kristensen
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox