On 13.02.2005, at 22:43, Sven Neumann wrote:
Doing benchmarks is exactly what would have to be done at this point. The main problem is probably what to use as a benchmark. One could write a script-fu and run that using batch-mode.
I've been using the coffee stain effect fu in the past but this is *way* to random to deliver sensible results for timing tests; it is somewhat useful for profiling though which is not what we would want to do in this case, though.
It would be interesting to have numbers that show the effect of the "num-processors" gimprc option and it would be interesting to see how the value of the TILES_PER_THREAD #define influences the performance. So far I can only guess that the cost of thread creation is perhaps a problem.
If anyone can deliver some good benchmark I'd be happy to run it on a few machines with the CVS version....
Description: This is a digitally signed message part