Re: Premature and micro optimization...

Not only are you micro-optimizing, but you're looking, IMO, at the wrong spot. Inlining the code, thus not calling a sub, might produce a lot more savings in execution time, than skipping tr.

And even then, you should consider the bigger picture: how much of your total execution time is spent in thi sub? Not much, that's my guess.

Comment on Re: Premature and micro optimization... Download Code

Replies are listed 'Best First'.
Re^2: Premature and micro optimization... by chargrill (Parson) on Jun 11, 2007 at 12:28 UTC
Right - the purpose was not to micro-optimize. The purpose was that someone suggested that a simple change might have a negative impact on everything (needlessly applying `tr///` on all IDs). I didn't think it would matter, I had time to kill, and I was curious if the impact would be measurable or not. --chargrill `s*lil; $=join'',sort split q; s;.;grr; &&s+(.(.)).+$2$1+; $; = qq-$_-;s,.,ahc,;$,.=chop for split q,,,reverse;print for($,,$;,$,$/)` [download]	[reply] [d/l] [select]
Re^3: Premature and micro optimization... by bart (Canon) on Jun 11, 2007 at 13:29 UTC
Hmm, I must not have expressed myself clearly. It is not you, but your colleague who is guilty of (passive) micro-optimization, and focussing on a wrong thing. I agree that adding to the code and ending up with something that would appear to be faster, is indeed bizarre.	[reply]
Re^4: Premature and micro optimization... by Dominus (Parson) on Jun 12, 2007 at 11:16 UTC
I must not have expressed myself clearly. It is not you... I misunderstood you the same way. Can I suggest that the problem was that you said "you" when you meant "he"? Your original post had: Not only are you* micro-optimizing, but you're looking, IMO, at the wrong spot.* Hope this helps. -- Mark Dominus Perl Paraphernalia	[reply]
Re^5: Premature and micro optimization... by bart (Canon) on Jun 12, 2007 at 11:43 UTC


Problems? Is your data what you think it is?
	PerlMonks