----- Original Message -----
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Saturday, October 16, 1999 10:19 AM
Subject: [MP3 ENCODER] Optimization Techniques
> On 15-Oct-99 Richard A. Smith wrote:
> > On Fri, 15 Oct 1999 10:59:39 +0100 (WEST), Patrick De Smet wrote:
> >>Anyway, I got this idea from a recent book discussing all sorts of
optims,
> >>and how (good) recent compilers work, I can look up the ref if you want
> >>it;
> >
> > Yes please send me the ref offlist. In fact if we can probally take
> > this discussion off list unless other are interested in it.
>
> I'd be interested as well.
> Keep it on-list, there's plenty of people working on speed tweaks, and
this
> forum is meant to help us write good & fast code.
>
> I remember seeing an ORA book (I think) which was about writing efficient
code.
> But i didn't have time to browse it :(
>
> As a suggestion keep the subject relevant so that people can filter if
they
> want.
Quite a good page:
http://www.ontek.com/mikey/optimization.html
Books-wise, Michael Abrash ("Zen of..." series) is good, as is a book called
Inner Loops. Both tend towards x86, but there's a lot of good general info
there too.
Re the array thing that started this off, I ran some more tests with it, and
think my result with MSVC must have been a glitch. Overall, with that
summing test, I found
Iterating backwards through the array is usually faster
With MSVC, using explicit pointer code is usually faster
With gcc, using array notation is usually faster (particularly, I found,
when the dataset exceeds the level 2 cache size - couldn't figure out
why...)
If anyone wants to look at the benchmark code, mail me at my work account
([EMAIL PROTECTED]) because that's where the code is. :)
-- Mat.
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
