MIME-Version: 1.0
References: <5382FD21.6040600@clear.net.nz>
            <CANQYN6zOq8dd5BUi32nCJGh9QgagZaaoj4UW1Ku3B4XWHi2PNA@mail.gmail.com>
            <A11620FE-AE5E-4CF7-A014-A5C5D4E05121@gmail.com>
            <CANQYN6zPFfu3zaS+Vp_V1Z_e_PNQHrO1xrRkutXAMFuU89yZ-w@mail.gmail.com>
            <53865805.7030602@clear.net.nz>
Content-Type: text/plain; charset=UTF-8
Message-ID:  <CANQYN6yF5he1UJ7z2PwwkUx8JYSEwSgUAVG400paofTBnRQcqQ@mail.gmail.com>
Date:         Sun, 1 Jun 2014 13:40:38 -0400
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
Sender: Mailing list for the LaTeX3 project <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
From: Bruno Le Floch <blflatex@GMAIL.COM>
Subject: Re: l3fp and sin(pi), cos(pi/2)
To: LATEX-L@LISTSERV.UNI-HEIDELBERG.DE
In-Reply-To:  <53865805.7030602@clear.net.nz>
Precedence: list
Envelope-To: <rainer.schoepf@GMX.NET>
Status: R

> I had the experience recently of checking a complicated trigonometric
> expression derived, after an embarrassingly long labour, from another
> complicated trigonometric expression. Had I made a mistake in the
> working? I evaluated the expressions using l3fp. One, for a number of
> different parameter values gave 0, the other for the same parameter
> values gave -0. I concluded, ruefully, that the derivation was correct.

Note that \fp_compare:nTF { 0 = -0 } { true } { false } gives true, so
you could just let l3fp compare the numbers.  I have to say I'm
surprised by that use of l3fp: I'd definitely use a general purpose
programming language for that (e.g., python).  Happy l3fp helps.

> To someone, like me, not versed in numerical analysis, the occurrence of
> signed zero was a surprise. The 0+x trick is good to know and perhaps
> worth mentioning in the documentation.

I'm not keen on mentioning it in the documentation, as it is a trick
indeed.  Instead, I've moved "customized formatting of floating
points" (e.g. with a printf-like syntax) up in my todo list.  I have
some ideas there.

Bruno