| Re: [eigen] 3.0.4 coming soon, testing appreciated |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/eigen Archives
]
- To: eigen@xxxxxxxxxxxxxxxxxxx
- Subject: Re: [eigen] 3.0.4 coming soon, testing appreciated
- From: Gael Guennebaud <gael.guennebaud@xxxxxxxxx>
- Date: Wed, 23 Nov 2011 23:15:30 +0100
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :content-type; bh=2KSgLnFYAirp9m8hoGN2ohEfVotLdMty4nd7aAf/K88=; b=DKHkLbTSNUB6MxidZ1x+ciKc71cKS86Oh00vOm3pcFGwB1LCol5f98un9Gdo2qx0Gy 5uNuzOsmn2+RdoPOY0E669gUWc/Z8ugdZT0E0Up1QSvJBJsYLfB7+fukuASA+6pZwYNJ TNXRK3h8VZVDUI8cgUlrPdaAScTEla+WkOwLM=
Hi,
note that this bug concerned only the *fast and direct* path triggered
by computeDirect(), not the default one. Also, this concerns an
upcoming feature for 3.1 which is not part of 3.0.4. That said, I've
just committed an improved version.
gael
On Tue, Nov 22, 2011 at 6:42 PM, Jitse Niesen <jitse@xxxxxxxxxxxxxxxxx> wrote:
> Now I fixed a bug, I deserve the right to point to another bug which I want
> to have fixed for 3.0.4 :) I'd like us to give correct results for the 3x3
> eigenvalue problem; see messages "3x3 symmetric eigenvalues Problem" of 31
> Oct - 2 Nov. I did not follow this carefully, so I might be wrong, but my
> impression is that we give very wrong results in some simple cases because
> we have some special code for the 3-by-3 case. If this proves difficult to
> fix in time, perhaps we can just use the general code?