When Python outruns C++ ...

[Anders J. Munch]

"Mike Meyer" <mwm at mired.org> skrev:
> "Anders J. Munch" <andersjm at dancontrol.dk> writes:
> > "Mike Meyer" <mwm at mired.org> wrote:
> > > I disagree with that statement. Clearly, *someone* knows all the
> > > coding tricks used by the compiler - because they put it into the
> > > compiler. So they can achieve the same level of performance.
> > And chess program authors can beat their own inventions?  Not so.
>
> Go program authors can beat their own inventions, so clearly this
> depends on the problem domain. In the chess (and go) problem domains,
> the time the human spends on the problem is artificially limited by
> the rules of the game. For programming, limiting the programmer time
> is a goal, not an axiom.

Here's a brilliant algorithm for register allocation that I just stole
from the PSU labs.

1. Generate all possible combinations.
2. Score each of them according to estimated performance.
3. Pick the best scoring combination.

Now try with 8 registers and 10 variables.  1,073,741,824 variations.

No limit on programmer time?

please-share-the-secret-of-eternal-life-with-the-rest-of-us-ly y'rs,
Anders