Re: [PATCH] Re: Bignum performance

[Emanuel Berg]

Simon Leinen wrote:

> Emacs also has `most-negative-fixnum' and
> `most-positive-fixnum' (borrowed from Common Lisp but now
> part of the core).
>
> On my 64-bit system (GNU Emacs 30.0.50 on
> aarch64-apple-darwin22.5.0):
>
> (= (- (expt 2 61)) most-negative-fixnum) → t
> (= (1- (expt 2 61)) most-positive-fixnum) → t
>
> (Same as on Emanuel's.)

At least that computation was correct then!

Now that Gerd has solved the little mystery why Fibonacci was
seemingly so much faster on Common Lisp and we also got that
performance gain patch from Ihor I don't know how much sense
it makes to continue translating the emacs-benchmarks to
Common Lisp, or rather if anyone is motivated enough to do it,
but this is how far I got:

  https://dataswamp.org/~incal/cl/bench/

-- 
underground experts united
https://dataswamp.org/~incal