From: Ulrich Drepper <drepper@gmail.com>
To: "libstdc++@gcc.gnu.org" <libstdc++@gcc.gnu.org>
Cc: Jonathan Wakely <jwakely@redhat.com>,
GCC Patches <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH] libstdc++: add uniform on sphere distribution
Date: Sat, 09 Aug 2014 18:00:00 -0000 [thread overview]
Message-ID: <CAOPLpQfZnWxkRj6fV3kTGEXEJkg_MCujVqPUwZnrUtdd11REFw@mail.gmail.com> (raw)
In-Reply-To: <alpine.DEB.2.11.1408091921450.10998@stedding.saclay.inria.fr>
On Sat, Aug 9, 2014 at 1:40 PM, Marc Glisse <marc.glisse@inria.fr> wrote:
> __x = result_type(2.0) * __aurng() - 1.0;
You're right, we of course need the negatives as well.
> Assuming the 2 coordinates are obtained through a rescaling x->2*x-1, if
> __sq is not exactly 0, it must be between 2^-103 and 1 (for ieee
> double), so I am not sure hypot gains that much (at least in my mind
> hypot was mostly a gain close to 0 or infinity, but maybe it has more
> advantages). It can only hurt speed though, so not a big issue.
Depending on how similar in size the two values are, not using hypot()
can drop quite a few bits. Especially with the scaling through
division this error can be noticeable. Better be sure. Maybe at some
point I have time to investigate the worst case scenario for the
numbers in question but until this shows hypot isn't needed it's best
to leave it in.
I've committed the patch.
next prev parent reply other threads:[~2014-08-09 18:00 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-07-13 2:01 Ulrich Drepper
2014-07-13 2:06 ` Ulrich Drepper
2014-07-13 9:28 ` Paolo Carlini
2014-07-13 10:04 ` Ulrich Drepper
2014-07-13 10:40 ` Paolo Carlini
2014-07-23 10:05 ` Marc Glisse
2014-07-23 10:44 ` Jonathan Wakely
2014-07-24 22:14 ` Ulrich Drepper
2014-07-24 23:19 ` Jonathan Wakely
2014-08-08 23:24 ` Ulrich Drepper
2014-08-09 7:15 ` Marc Glisse
2014-08-09 11:55 ` Ulrich Drepper
2014-08-09 12:34 ` Marc Glisse
2014-08-09 15:24 ` Ulrich Drepper
2014-08-09 15:33 ` Marc Glisse
2014-08-09 16:55 ` Ed Smith-Rowland
2014-08-09 16:56 ` Ulrich Drepper
2014-08-09 17:40 ` Marc Glisse
2014-08-09 18:00 ` Ulrich Drepper [this message]
2014-08-13 7:36 ` Bin.Cheng
2014-08-13 7:43 ` Bin.Cheng
2014-08-13 8:41 ` Paolo Carlini
2014-08-13 11:25 ` Paolo Carlini
2014-08-13 11:32 ` Marc Glisse
2014-08-13 11:40 ` Paolo Carlini
2014-07-13 13:55 Ed Smith-Rowland
2014-07-13 14:12 ` Ulrich Drepper
2014-07-13 14:48 ` Ed Smith-Rowland
2014-07-13 15:47 ` Paolo Carlini
2014-07-13 16:04 ` Ulrich Drepper
2014-07-13 16:12 ` Paolo Carlini
2014-07-13 16:18 ` Ulrich Drepper
2014-07-13 16:29 ` Paolo Carlini
2014-07-13 16:45 ` Ulrich Drepper
2014-07-13 16:56 ` Paolo Carlini
2014-07-14 7:58 ` Andreas Schwab
2014-07-14 8:18 ` Paolo Carlini
2014-07-16 11:30 ` Ed Smith-Rowland
2014-07-16 15:03 ` Paolo Carlini
2014-07-16 17:38 ` Ulrich Drepper
2014-07-14 10:31 Dominique Dhumieres
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAOPLpQfZnWxkRj6fV3kTGEXEJkg_MCujVqPUwZnrUtdd11REFw@mail.gmail.com \
--to=drepper@gmail.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=jwakely@redhat.com \
--cc=libstdc++@gcc.gnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).