The third one is another example of dsvdc failure. It can be fixed by replacing dsvdc with dgesvd in force calculation. The patch has been sent to Jason to apply.
Thanks,
Qin
On Mar 28, 2012, at 9:11 AM, Daniel Roe wrote:
> Hi,
>
> On Wed, Mar 28, 2012 at 10:42 AM, David A. Case
> <case.biomaps.rutgers.edu> wrote:
>> Can you say which test case(s) you are talking about? Hopefully provide a
>> patch (I guess with new outputs) that implements the proposed change?
>
> It relates to the first two tests I mentioned in this thread,
> pbsa_aug_c2 and sas2test. I provided a cygwin-specific patch
> (switching the SVD method used), but I think the consensus is that it
> should be changed in general since the general SVD method seems more
> stable. Changing the method did not change the test results; in fact,
> it was necessary to get them to even pass on cygwin.
>
> -Dan
>
> _______________________________________________
> AMBER-Developers mailing list
> AMBER-Developers.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber-developers
>
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Wed Mar 28 2012 - 18:30:04 PDT
