[gmx-users] Wrong LJ energies on Apple Mac OS X with Myrinet

[Goran Krilov]

The issues comes up whenever I use the myrinet linked binaries,
even if I don't run on multiple processors.  Also, the result seems
to be independent on how many processors I split the parallel run
over, I get the same (incorrect) result. I have the tpr file so I  
will file
a bugzilla.

Thanks!

Regards,

------------------------------------------------------------------------ 
----
Dr. Goran Krilov                    Tel. (617) 552-8927
Assistant Professor               Fax (617) 552-2705
Department of Chemistry     E-mail: goran.krilov at bc.edu
Boston College
Merkert Center, 2609 Beacon St
Chestnut Hill, MA 02467
------------------------------------------------------------------------ 
-----


On Mar 28, 2006, at 4:48 PM, David van der Spoel wrote:

> Goran Krilov wrote:
>> Hi,
>> I have an  issue with GROMACS on Apple PPC that I am trying to
>> resolve. Careful analysis of our data has shown that one particular
>> term of the energy function (LJ short range) gives a different  
>> result,
>> depending on which version of GROMACS we use.
>> Namely, if I use the binary precompiled by GROMACS developers on   
>> darwin 8.2.0,
>> I get the result 9.99750e+03. However if I use the binaries built   
>> with the
>> Myrinet MX and Apple 4.0 compiler on 8.3.0, that were built by us
>>  we get  1.00229e+04 for the same term. That is a significant
>> difference. Especially so, because all the other energy terms are   
>> exactly the
>> same for the two versions.
>> Moreover, when we use GROMACS 3.3 and GROMACS 3.2.1 on Linux, we get
>> 9.99750e+03. Also, independent calculation with a different  
>> package  (TINKER)
>> gives us .99750e+03. In all cases we used the same input and data   
>> files, and outputs
>> are identical except for the difference in the energy term noted.
>> So it would seem our Myrinet built GROMACS gives the wrong  
>> answer.  Any ideas what
>> could be going on?
>
> A difficult problem, although there is only one deviation. You say  
> you have myrinet stuff linked in, but do you use it in this case?  
> If this is not a parallel simulation, then the myrinet library will  
> not affect the results. If this is a parallel simulation you could  
> try to use the tcp communication, if this gives the right results  
> your myrinet library might be faulty, although it seems very  
> unlikely. I would  rather suspect the compiler. But first things  
> first: if you can provide a tpr file that works in 3.2.1 and that  
> produces the right values in 3.2.1 we can test it. Please file a  
> bugzilla in that case.
>
>> Thanks!
>> Regards,
>> --------------------------------------------------------------------- 
>> --- ----
>> Dr. Goran Krilov                    Tel. (617) 552-8927
>> Assistant Professor               Fax (617) 552-2705
>> Department of Chemistry     E-mail: goran.krilov at bc.edu
>> Boston College
>> Merkert Center, 2609 Beacon St
>> Chestnut Hill, MA 02467
>> --------------------------------------------------------------------- 
>> --- -----
>> _______________________________________________
>> gmx-users mailing list    gmx-users at gromacs.org
>> http://www.gromacs.org/mailman/listinfo/gmx-users
>> Please don't post (un)subscribe requests to the list. Use the www  
>> interface or send it to gmx-users-request at gromacs.org.
>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>
>
> -- 
> David.
> ______________________________________________________________________ 
> __
> David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
> Dept. of Cell and Molecular Biology, Uppsala University.
> Husargatan 3, Box 596,  	75124 Uppsala, Sweden
> phone:	46 18 471 4205		fax: 46 18 511 755
> spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 
> ++++
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the www  
> interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php