Re: -O3 vs. -Os
Re: -O3 vs. -Os
- Subject: Re: -O3 vs. -Os
- From: Rustam Muginov <email@hidden>
- Date: Tue, 17 Jan 2006 15:49:50 +0300
On Jan 17, 2006, at 12:06 PM, Steve Checkoway wrote:
How about a game? Still, I would say that this uses fairly
computationally intensive mathematics. Matrix multiplications left and
right for graphics work, resampling sounds for audio output, etc. I
have used Shark and there is no single hot spot clearly visible in the
game.
Do you using MacOS X audio resampler or your own?
Resampling, if it has to be performed with high quality, is a heavy
computation task.
the same question form matrix math - is it "in-game" code or code from
Accelerate.framework?
Still, I wonder on what you were basing your assertion that only
"manually-tuned high-performance libraries" benefit from -O3. I
wouldn't claim that this game is manually tuned at all. Plus, I would
think that such code would benefit the least from optimization if it
is carefully done by hand to begin with.
I think in this case it is pretty clear that more highly optimized
code that is actually smaller than the less optimized code is likely
to run faster. Based on framerates (turning vsync off, of course),
that is the case.
I'm curious if this is normal.
Where is "size/speed" tradeof.
If you do have some math-intensive code, for example, where is a
"threshhold" - then this chunk of code grown large enough to not fit to
stack, performance drops drammaticaly. So _generaly_, its prefered to
minimize code size.
If, from the other side, you had rearranged your algorithm to make a
streamlined instructions flow, then it could be more productive to
"unroll" code, to exclude branches e.t.c.
--
Sincerely,
Rustam Muginov
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Xcode-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden