Aliaksei Syrel wrote > Hi > > Did you know that `2 * 10.2` is two times slower than `2 asFloat * 10.2`? > > Here is performance test (inspect result, takes ~10s to complete): > http://ws.stfx.eu/4T0CR23OURZW > > Who knows how performance can be improved with this knowledge... > > Cheers, > Alex
And on pre-Spur VM it's like 6 or 7 times slower. More interesting fact is that (float * int) gets optimized, but (int * float) doesn't. It would be cool to have an up-to-date document explaining preformance characteristics of Pharo code and optimizations done by compiler/JIT. Right now only source of such stuff seems to be the blog of Clément Béra (https://clementbera.wordpress.com/) -- View this message in context: http://forum.world.st/Having-fun-with-Pharo-tp4912104p4912472.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.