On Sun, Nov 23, 2003 at 10:10:24PM +0900, Peter O'Gorman wrote:
> | I'd be interested to see some benchmarks to back up these claims.
> |
> I don't have benchmarks, I'll leave it up to you to convince me that
> anything can load an 80428 byte libintl.dylib faster than a 30316 byte
> one :)

...and make the program run noticably faster.  If libintl.dylib takes
2ms to load and you speed that up by a whopping 50% it'll still only be
half a millisecond.

More importantly, is the performance increase worth the significant extra 
complexity of having stripped vs not-stripped versions of packages?

That's why you need some numbers.  If this is only going to speed most
small programs up by, say, 1% its probably not worth it.  If its 10%, maybe.  
If its 50%, sure.  But right now we don't know anything.


-- 
Michael G Schwern        [EMAIL PROTECTED]  http://www.pobox.com/~schwern/
You and your "facts" and your "physics".  Pah, I say.
        http://www.goats.com/archive/981221.html


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
_______________________________________________
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Reply via email to