🚀 go-pugleaf

Hallo Wolfgang!

Tuesday July 04 1995 07:29, Wolfgang Gellerich wrote to All:

 WG> This can not be answered with yes or no. It depends on properties of the
 WG> target architectur, the programmer's skills in writing assembly code and on
 WG> how much optimization the M2 compiler does. E.g., a compiler applying the
 WG> graph coloring algorithm to do register optimization will probably produce
 WG> much better code on a RISC architecture than anyone trying to write RISC
 WG> assembler by hand. But, of course a stupid code generator can produce very
 WG> slow code.

You could of course take the compiled procedure, disassemble it, and optimize it by hand even more. It's the approach that creates the difference you mention, not a problem of assembler. One should always find the best algoritm before starting to program.

 WG> In either case, M2 is MUCH more portable.

In this special case, I doubt that, he's programming in TopSpeed.

Marco  (M.W.T.J.v.d.Voort@stud.tue.nl or MarcoV@stack.urc.tue.nl)