frustrations in de-slowifying code.
sqrmax at comcast.net
Sat Mar 20 09:00:21 UTC 2004
First, I'd like to tell you that yes it can probably done. A long
time ago I optimized my finite state model startup code from 119
seconds down to 0.4 seconds. I remember the final figure was 230x
faster than the original. And no primitives!
Friday, March 19, 2004, 11:36:57 PM, you wrote:
AG> 1. making revisions that would seem to combine the jobs of a 4 second
AG> method with a 5 second method removing one of the loops and the creation
AG> of a large array in the process produced a 15 second method... =\
The first thing that comes to mind is if you are creating a ton of
small objects. I made my function plotters up to 72% faster by
reusing points while scaling (I implemented x:y: in Point to avoid
creating new instances - Behavior>>new *takes time*!!!).
AG> 2. The run-times are _EXTREMELY_ variable. said 15 second method
AG> sometimes takes as long as 5 _MINUTES_.
Hmmmmmmmmmm garbage collect?
More information about the Squeak-dev