Aren't they all compiled into assembly/machine code? Wouldn't that make them all the same speed?
Printable View
Aren't they all compiled into assembly/machine code? Wouldn't that make them all the same speed?
Is it not the case that language a has ten megs of assembly/machine code, where as language b has 15. Therefore the memory footprint is bigger. At least thats how i would logicly view it!
No. See compiled vs. interpreted vs. JIT-compiled.Quote:
Originally Posted by Fromethius
All models are trade-offs of varying degrees.
Memory footprint has nothing to do with speed. In fact optimisation often increases memory usage.Quote:
Originally Posted by CodedFire
Moved from Chit Chat
Ive spent the whole day reading up on cil (formerly msil?) and just in time compilation. Its a pretty good system. Although from what ive read it can be at times faster than native code due to the fact that it optimises the code it compiles to suit the chipset family, Surely this cant be right? after all it is still an extra compilation step vs straight native code?
It depends on what your code does, and the features that are provided in the compiler. Presumably it can make use of things like MMX, SSE, and threading on multiple processor cores - all of which can dramatically improve speed.