Re: [Maxima-discuss] gcl-2.6.13 released [stable]

[Raymond Toy]

On Sat, Jan 14, 2023 at 3:40 PM Richard Fateman <fateman@gmail.com> wrote:

It would be interesting to hear of a few benchmarks. What observations to make?

 WASM advocates seem to hold that WASM code

executes fast and efficient.  

From what I remember people at work saying, wasm was about 2-3 times slower than the equivalent C/C++ code.  Things may have changed since then.

  How long would one expect (under realistic constraints) a computer system to take to download a complete

Common Lisp system + Maxima (compiled from Lisp to WASM?) into (say) a Chrome browser?

Hard to know now what the size would be, but a stripped ecl binary for maxima is about 13 MB.  That wouldn't be terrible to download, and it would be cached so you don't have to download it every time.

  There are many possible benchmarks to compare different features, like bignum arithmetic, polynomial arithmetic,

input/output, the Maxima testsuite.

On Sat, Jan 14, 2023 at 11:17 AM Henry Baker <hbaker1@pipeline.com> wrote:

Re: emscripten version of ECL:

You can attempt to build it yourself:

https://gitlab.com/embeddable-common-lisp/ecl/-/tree/emscripten

-----Original Message-----
From: Marius Gerbershagen <marius.gerbershagen@gmail.com>
Sent: Jan 14, 2023 9:39 AM
To: <maxima-discuss@lists.sourceforge.net>
Cc: <gcl-devel@gnu.org>
Subject: Re: [Maxima-discuss] gcl-2.6.13 released [stable]

It's perhaps worth noting that ECL can be made to run in this way in a browser and will officially support WASM in the next release (https://gitlab.com/embeddable-common-lisp/ecl/-/merge_requests/277).



<snip>

_______________________________________________
Maxima-discuss mailing list
Maxima-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/maxima-discuss

--

Ray