|Posted by C. Dunn||PermalinkReply|
> Your code must do a lot of 64b computation!
Yes. Lots of 64-bit integer shifts of pointer values. The only other difference is that 32-bit pointer addresses are closer together. The result: The exact same code compiled for 32- or 64-bit has a 2x speed difference.
> When the AMD64 was released benchmarks made showed at best a 20% improvement, which is quite a lot to get with a simple recompilation assuming the code is 64b clean of course but is nowhere 2x.
Yes. 20% is a good rule of thumb. That's what we usually see, for general object-oriented programming. System-level programming is different.
I'll try gdc someday....