On Tuesday, March 23, 2004, at 01:34 PM, Glen Bojsza wrote:
This is a substantial difference...can anyone give me some pointers on how I should optimize repeat loops. I thought that doing the one big loop should be as fast as the multiple loops?
'line n' takes time proportional to n. Looping like this is proportional to n^2. Breaking this up keeps n down. All that extra time is in counting lines.
Use 'repeat for each line' if you can. This is crucial. You will greatly increase your speed.
Accumulating with 'put ... after ...' is efficient. Continue with that. I'm assuming that stimeX is global.
On some systems 'convert' will affect the value for the long seconds, so you might not be able to time this very well if it gets very fast.
Dar Scott
_______________________________________________ use-revolution mailing list [EMAIL PROTECTED] http://lists.runrev.com/mailman/listinfo/use-revolution
