comp.lang.ada
 help / color / mirror / Atom feed
From: vincent <vincent.diemunsch@gmail.com>
Subject: Re: Mathpaqs release 30-May-2015
Date: Wed, 3 Jun 2015 08:16:18 -0700 (PDT)
Date: 2015-06-03T08:16:18-07:00	[thread overview]
Message-ID: <a3c6f4a9-943a-459d-9198-bb802dcf4179@googlegroups.com> (raw)
In-Reply-To: <eb391245-c663-4e5f-a62e-8bd191c75d25@googlegroups.com>

Le mardi 2 juin 2015 14:27:48 UTC+2, gautier...@hotmail.com a écrit :
Thanks Gauthier for this interesting work !

I am really interested by your multiprecision integers. I am happy to see that they are implemented in Ada. 

I used to write a binding to GMP and MPFR but I only considered "Unbounded Integers", implemented as controlled types, just like unbounded_strings are implemented. Do you know how fast is your Library compared to GMP ?

If we had an Ada compiler based on LLVM, we could use "portable" inline assembly code to accelerate computations... Would this be useful ? I suppose the most important thing is the algorithm, isn't it ?

What I would like to have is the euclidean division :
 
A div B = if B > 0 then Floor(A,B) else Ceiling (A,B) end if;
Where Floor(A,B) is the floor of the rational number A/B.

A mod B the corresponding remainder, which is "mod" when B > 0.

These are the div and mod operators from Oberon for instance. I think they really lack in Ada. Would it be difficult to have them starting from your implementation ?

Kind regards,

Vincent
 

  reply	other threads:[~2015-06-03 15:16 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-02 12:27 Mathpaqs release 30-May-2015 gautier_niouzes
2015-06-03 15:16 ` vincent [this message]
2015-06-03 15:41 ` vincent.diemunsch
2015-06-03 18:09   ` Jeffrey R. Carter
2015-06-04  7:25     ` Vincent
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox