Re: Why are intepreters so slow today

Stefan Monnier (monnier@di.epfl.ch)
18 Apr 1994 19:47:56 +0200

In article <knight.766417773@tina.mrco.carleton.ca>,
Alan Knight <knight@mrco.carleton.ca> wrote:
> I believe it would be roughly accurate to say that a decent Smalltalk
> should be in the range of 6 to 10 times slower than C for small
> samples of "normal" code. This could be much worse for code with lots
> of array accesses (which are always bounds-checked in Smalltalk) or
> floating point arithmetic. People argue that this performance
> difference evens out dramatically on larger code samples, but this is
> much harder to prove.

There's also the Self system. Self is a language derived from
Smalltalk but which uses prototypes+(delegation+cloning) instead of
objects+(inheritance+instanciation). It is "highly dynamic" in that it
offers multiple and dynamic delegation. But still: they claim being
with a factor of 2 to 3 relative to C. Of course, you could argue
it is compiled, but from the user point of view, it is interpreted
(who cares what's really going on, right ?)

And as far as I can tell, SCM (a scheme interpreter written in C and
that doesn't even compile to bytecode or threaded code) is generally
about 10 times slower than C (for integer style code) even though it
offers bignums, array bounds checking and the likes.

Stefan

-- 

----------------------------------------------------- -- On the average, people seem to be acting normal -- -----------------------------------------------------