<html>
<head>
<style>
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
FONT-SIZE: 10pt;
FONT-FAMILY:Tahoma
}
</style>
</head>
<body class='hmmessage'><div style="text-align: left;"><br></div><br><br><br><hr id="stopSpelling">> Date: Tue, 14 Oct 2008 07:19:45 -0400<br>> From: jalemkul@vt.edu<br>> To: gmx-users@gromacs.org<br>> Subject: Re: [gmx-users] Unexplained crashes with Gromacs-4.0<br>> <br>> <br>> <br>> Berk Hess wrote:<br>> > Hi,<br>> > <br>> > I don't really know what could be the problem here.<br>> > <br>> > 0..180 is (confusingly) not including the end, so 180 is out of range.<br>> > <br>> > The only thing I can think of (except for unknown bugs of coarse),<br>> > is that I corrected the coupling time of the Parrinello-Rahman barostat<br>> > when going from 4.0 RC4 to the official 4.0 release.<br>> > I guess all your old simulations were with older Gromacs releases.<br>> > So to get the same coupling as before, so would need to multiply tau_t <br>> > by 16.6.<br>> > I don't know if this would help.<br>> <br>> I did see this same error under RC2, but noticed that you fixed this bug shortly <br>> after I saw it, and that's what I assumed it was. But now I have seen it under <br>> the official 4.0 release.<br>> <br>> > <br>> > You did not get pressure coupling warnings before the crash?<br>> > <br>> <br>> There are no messages printed before this error pops up. What's really <br>> confusing is that if I simply try the same .tpr file again, it runs without a <br>> problem. I know that's not terribly scientific, but it seemed odd to me that my <br>> systems running for 40+ ns suddenly had a problem.<br>> <br>> Is there any diagnostic information I can provide? This problem seems like it <br>> would be elusive, since even I can't fully reproduce it.<br>> <br>> -Justin<br><br>If you run in parallel, the dynamic load balancing can affective reprodibility.<br>You can turn it off with -dlb no<br><br>Another source of non-reproducibility is FFTW (with PME).<br>By default FFTW times which of different codes is the fastest and uses that.<br>Configuring FFTW with SSE makes all codes faster and the differences smaller,<br>so you can configure Gromacs with --disable-fftw-measure to remove<br>the run time optimization without performance loss.<br>For debugging purposes the new -reprod switch of mdrun will try to turn<br>off all optimizations which might affect reproducibilty.<br><br>Berk<br><br><br /><hr />Express yourself instantly with MSN Messenger! <a href='http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/' target='_new'>MSN Messenger</a></body>
</html>