That was just one of the many things that urked me about the code (another was pratically no comments and when i mean pratically, I mean 1 comment per 1000 lines of code if that and I cannot forget the all so ever helpful labels like loop1 )

This was a commerical application of a game that did not need all too much processing power. The programmer (I hope you are right in the fact that might have been one poor, inexperienced programmer, CornedBee) did not know his tools enough to make something that was semi-readable. Lastly, at the end, the program was extremely buggy. I came to the conclusion that there was no reason for such effort and waste.

Do not get me wrong. I do support assembler optimizations, and I do support learning assembler, however, if you are going to do a complex windows application that will not need a ton of speed, I see no reason to use pure assembler.