https://issues.dlang.org/show_bug.cgi?id=10550
Zach Bjornson changed:
What|Removed |Added
CC||zbbjorn...@gmail.com
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #15 from Joseph Rushton Wakeling joseph.wakel...@webdrake.net
2013-07-09 01:04:56 PDT ---
Masahiro -- thanks so much for your fast attention to this. :-)
--
Configure issuemail:
http://d.puremagic.com/issues/show_bug.cgi?id=10550
Joseph Rushton Wakeling joseph.wakel...@webdrake.net changed:
What|Removed |Added
Status|NEW
http://d.puremagic.com/issues/show_bug.cgi?id=10550
bearophile_h...@eml.cc changed:
What|Removed |Added
CC||bearophile_h...@eml.cc
---
http://d.puremagic.com/issues/show_bug.cgi?id=10550
Joseph Rushton Wakeling joseph.wakel...@webdrake.net changed:
What|Removed |Added
Severity|normal
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #7 from Joseph Rushton Wakeling joseph.wakel...@webdrake.net
2013-07-08 05:58:49 PDT ---
What I'd really like is to have a source for the checking values for Xorshift
used in the unittests. Masahiro, do you recall how you
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #8 from Masahiro Nakagawa repeate...@gmail.com 2013-07-08
07:28:26 PDT ---
(In reply to comment #7)
What I'd really like is to have a source for the checking values for Xorshift
used in the unittests. Masahiro, do you recall
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #9 from Joseph Rushton Wakeling joseph.wakel...@webdrake.net
2013-07-08 07:51:04 PDT ---
(In reply to comment #8)
I remember correctly, I generated test cases from paper based C
implementation.
I implemented C version first
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #10 from Masahiro Nakagawa repeate...@gmail.com 2013-07-08
09:26:03 PDT ---
(In reply to comment #9)
...
In summary, I think we can proceed as follows:
* confirm with experts in the field the typos in the paper
*
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #11 from Joseph Rushton Wakeling joseph.wakel...@webdrake.net
2013-07-08 09:47:04 PDT ---
(In reply to comment #10)
I rechecked the paper and I agree with you (y^=, a c and 160's )
So I think we can fix this paper derived bugs.
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #12 from Joseph Rushton Wakeling joseph.wakel...@webdrake.net
2013-07-08 12:49:09 PDT ---
(In reply to comment #11)
(In reply to comment #10)
I rechecked the paper and I agree with you (y^=, a c and 160's )
So I think we can
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #13 from Joseph Rushton Wakeling joseph.wakel...@webdrake.net
2013-07-08 14:31:53 PDT ---
Pull request submitted:
https://github.com/D-Programming-Language/phobos/pull/1403
--
Configure issuemail:
/e1504f70b8cfdf4cd1082408fe2452a71e299ab3
Merge pull request #1403 from WebDrake/xorshift
Fix Issue 10550 - Xorshift32 and Xorshift160 do not generate
uniformly-distributed random numbers
--
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #3 from Joseph Rushton Wakeling joseph.wakel...@webdrake.net
2013-07-05 07:38:56 PDT ---
(In reply to comment #2)
I also think the choice of a, b, c values may be in error: currently we have
alias XorshiftEngine!(uint, 32,
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #2 from Joseph Rushton Wakeling joseph.wakel...@webdrake.net
2013-07-05 07:35:38 PDT ---
I also think the choice of a, b, c values may be in error: currently we have
alias XorshiftEngine!(uint, 32, 13, 17, 5) Xorshift32;
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #1 from Joseph Rushton Wakeling joseph.wakel...@webdrake.net
2013-07-05 07:35:07 PDT ---
The Xorshift32 non-uniformity can be fixed by correcting the update rule in
popFront(), from:
static if (bits == 32)
{
http://d.puremagic.com/issues/show_bug.cgi?id=10550
--- Comment #4 from Joseph Rushton Wakeling joseph.wakel...@webdrake.net
2013-07-05 08:12:24 PDT ---
Uniformity in Xorshift160 can be restored by tweaking the update rules:
else static if (bits == 160)
{
temp
17 matches
Mail list logo