[gmx-developers] gromacs 2016 beta1 bug somewhere in the new simd for bonded/LJ/lincs

[Berk Hess]

Hi,

I just found a bug in the 2016 beta that probably explains your 
observations:
http://redmine.gromacs.org/issues/1968
Fix at:
https://gerrit.gromacs.org/#/c/5885

Cheers,

Berk

On 2016-05-17 09:58, Carlo Camilloni wrote:
> Hi
>
> The x must be identical because I used -rerun to parse the trr file, while
> I don’t know if the velocities are recalculated by rerun.
> What I am sure about is that the differences observed with metadynamics
> are highly significant.
> Right now I don’t have more time to investigate this, I will definitely
> do it as soon as I can, but if you have time i would have a further
> look into it..
>
> Cheers,
> Carlo
>
>
>> Message: 1
>> Date: Fri, 13 May 2016 12:10:52 +0200
>> From: Berk Hess <hess at kth.se>
>> To: gmx-developers at gromacs.org
>> Subject: Re: [gmx-developers] gromacs 2016 beta1 bug somewhere in the
>> 	new simd for bonded/LJ/lincs
>> Message-ID: <5735A82C.7080508 at kth.se>
>> Content-Type: text/plain; charset="windows-1252"; Format="flowed"
>>
>> Hi,
>>
>> In the files you sent I see that x and v are identical for all versions
>> and f has a few components that differ up to 0.002 kJ/mol/nm. This is
>> fully within the variation I would expect from different codes.
>> We need more evidence to judge if there is a bug present here. I can't
>> judge if the differences in your metadynamics results are significant
>> compared to the statistical accuracy.
>>
>> Cheers,
>>
>> Berk
>>
>> On 2016-05-13 11:59, Carlo Camilloni wrote:
>>> Dear GMX developers,
>>>
>>> First of all thanks for your continuos efforts in making gromacs better.
>>>
>>> I have done a very basic calculation with the new gromacs-2016-beta1
>>> and I think there is a bug somewhere
>>> in the force calculation. (all calculations are run using 1 processor,
>>> no mpi, no openmp, no gpus)
>>>
>>> I have run a calculation on alanine dipeptide:
>>>
>>> 1) GMX507: 2 steps using 1 processor (avx2 simd and nothing else, the
>>> tpr is generated with gromacs 507)
>>> 2) rerun with GMX512 on the trr file
>>> 3) rerun with GMX2016 on the trr file
>>> and then I compared forces, velocities and energies, and while they
>>> are essentially the same for the step 0
>>> in the case of GMX507 and 512 they are not the same anymore for GMX2016b1.
>>>
>>> The differences are not huge, even if from a rerun I would expect no
>>> differences at all, in particular on the
>>> forces, so I have run a metadynamics calculation from the same tpr
>>> file and using the three codes
>>> patched with plumed to see the effect on the free energy along the phi
>>> torsion angle, and unfortunately the
>>> difference is quite significant.
>>>
>>> to be sure I repeated the same tests on a linux workstation (the first
>>> test was on my macbook), so different
>>> compilers and so on, but always avx2 simd)
>>>
>>> here you can find everything (all the files I have generated):
>>> https://dl.dropboxusercontent.com/u/1349402/bug-report-aladip-2016.tgz
>>>
>>> I hope this helps.
>>>
>>> Best,
>>> Carlo
>>>
>>>
>>>
>>>
>>>
>>>
>> -------------- next part --------------
>> An HTML attachment was scrubbed...
>> URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20160513/1ad1cd57/attachment-0001.html>
>>
>> ------------------------------
>>
>> -- 
>> Gromacs Developers mailing list
>>
>> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List before posting!
>>
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers or send a mail to gmx-developers-request at gromacs.org.
>>
>> End of gromacs.org_gmx-developers Digest, Vol 145, Issue 17
>> ***********************************************************