[Cbc] Cbc problem

[John Forrest]

It does look as if depth>=0 has problems.

The idea of depth was that for small problems it might be better to 
switch to a simpler branch and bound which would be much faster per 
node.  depth >=0 tries to be more sophisticated and seems to have 
problems.  depth < 0 does depth first and does work.

However for reasonable sized problems it is probably not a good idea to 
try either version.

I would not classify this problem as small.

I will look into the bug - slowly.

John Forrest

On 20/07/15 11:36, See Chuen Teck wrote:
> Hi,
>
> I would like to report a problem with cbc 2.9.5 (downloaded from arch 
> linux repository). I ran the attached file with the following settings:
>
> doH
> feas off
> depth 0
>
> The below was obtained:
>
> Cbc0010I After 1100 nodes, 361 on tree, 0.538163 best solution, best 
> possible 0.2556182 (25.77 seconds)
> Segmentation fault (core dumped).
>
>
>
>
>
> _______________________________________________
> Cbc mailing list
> Cbc at list.coin-or.org
> http://list.coin-or.org/mailman/listinfo/cbc

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/cbc/attachments/20150721/647c2bcb/attachment.html>