Mail Archives: djgpp/1996/12/17/13:03:48
> You are right, but this is not the point, if something quite empty takes 8
> seconds, i extrpolate and for large projects, I doubt that the difference
> would be only the validation of a human tester. In any case, I managed
> everything to let it go faster. Than you should also calculate the times
> you compile something and test if becomes qite impressive i.e. 8x100 and
> 0.5x100 is already a nice amout of time. ;-)
You can't extrapolate from such a small sample (one test). You need
to compile a larger source also to determine what the static overhead
is. For example, Borland might be 0.1 sec/line, but DJGPP might be
0.05 sec/line plus 8 seconds of overhead.
- Raw text -