<html>
<head>
<style>
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
</style>
</head>
<body class='hmmessage'>
Hi,<br><br>That I have thought about and it would avoid a lot of trouble.<br>But I have not done that, because that would lead to different run results<br>every time you rerun grompp, which can be misleading when you are<br>trying to assess the effects of other parameters.<br><br>But maybe the first issue is more important than the second?<br><br>Berk<br><br><hr id="stopSpelling">Date: Thu, 5 Feb 2009 10:43:22 +0100<br>From: r.friedman@bioc.uzh.ch<br>To: gmx-users@gromacs.org<br>Subject: Re: [gmx-users] micelle disaggregated in serial, but not parallel,        runs using sd integrator<br><br><title></title>
Hi,<br>
Maybe it's a good idea to have ld-seed=-1 as a default if that's not
already the case.<br>
Ran.<br>
<br>
Berk Hess wrote:
<blockquote cite="midBLU134-W513D4AD5766AECD691F1AC8EC00@phx.gbl">
<style>
.ExternalClass .EC_hmmessage P
{padding:0px;}
.ExternalClass body.EC_hmmessage
{font-size:10pt;font-family:Verdana;}
</style>Hi,<br>
<br>
I don't know why I did not add checks for ld-seed before.<br>
Now grompp gives a note when continutation=yes and ld-seed!=-1.<br>
<br>
tpbconv will now generate a new ld-seed when reading a trajectory<br>
(but you should not use tpbconv, use a checkpoint file instead).<br>
<br>
But yesterday I forgot to tell that there is a bug in the checkpointing<br>
of mdrun in 4.0 - 4.0.3. Without domain decomposition the initial box<br>
size would always be stored in the checkpoint file, which causes
problems<br>
with NPT simulations. NPT simulations with domain decomposition and<br>
all NVT simulations were fine.<br>
<br>
Gromacs 4.0.4 will all bugs fixed and extra checks should be released
today.<br>
<br>
Berk<br>
<br>
> Date: Wed, 4 Feb 2009 19:30:47 -0500<br>
> From: <a class="EC_moz-txt-link-abbreviated" href="mailto:chris.neale@utoronto.ca">chris.neale@utoronto.ca</a><br>
> To: <a class="EC_moz-txt-link-abbreviated" href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a><br>
> Subject: [gmx-users] micelle disaggregated in serial, but not
parallel, runs using sd integrator<br>
> <br>
> It appears as if you were correct Berk. I will report on the
results of my 24h test tomorrow, but I also set up another system <br>
> that used ld_seed=1993 and ran in 20 ps segments instead of the
200 ps segments that I was previously using. This system shows <br>
> signs of disaggregation on the 200 ps time-scale as opposed to the
2 ns time-scale that I observed for 200 ps segments.<br>
> <br>
> I don't know how you figured that one out, but I am very grateful.<br>
> <br>
> Now that I see the trajectories, it does make sense that any net
movement applied to an individual molecule by the noise will <br>
> lead to directed movement over many separate segments.<br>
> <br>
> I think this is probably worth a note in the grompp output for sd
runs when a user sets ld_seed to something other than -1 and <br>
> utilizes the -t option (or some other indication that this is
intended as a continuation).<br>
> <br>
> Chris.<br>
> <br>
> -- original message --<br>
> <br>
> Thank you Berk,<br>
> <br>
> I will repeat my runs using the checkpoint file and report my
findings back to this list. Thank you for this advice.<br>
> <br>
> Chris.<br>
> <br>
> -- original message --<br>
> <br>
> Hi,<br>
> <br>
> In this manner you use the same random seed and thus noise for all
parts.<br>
> In most cases this will not lead to serious artifacts with SD,<br>
> but you can never be sure.<br>
> When checkpoints are used, you do not repeat random numbers.<br>
> This also gives a difference between serial and parallel in 4.0.<br>
> With serial you get exactly the same noise per atom, in parallel
not,<br>
> since atoms migrate from one node to another (with domain
decompostion).<br>
> <br>
> If you do not use checkpoints, use ld_seed=-1 and do not use
tpbconv.<br>
> <br>
> Berk<br>
> <br>
> <br>
> _______________________________________________<br>
> gmx-users mailing list <a class="EC_moz-txt-link-abbreviated" href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a><br>
> <a class="EC_moz-txt-link-freetext" href="http://www.gromacs.org/mailman/listinfo/gmx-users">http://www.gromacs.org/mailman/listinfo/gmx-users</a><br>
> Please search the archive at <a class="EC_moz-txt-link-freetext" href="http://www.gromacs.org/search">http://www.gromacs.org/search</a> before
posting!<br>
> Please don't post (un)subscribe requests to the list. Use the <br>
> www interface or send it to <a class="EC_moz-txt-link-abbreviated" href="mailto:gmx-users-request@gromacs.org">gmx-users-request@gromacs.org</a>.<br>
> Can't post? Read <a class="EC_moz-txt-link-freetext" href="http://www.gromacs.org/mailing_lists/users.php">http://www.gromacs.org/mailing_lists/users.php</a><br>
<br>
<hr>Express yourself instantly with MSN Messenger! <a href="http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/">MSN Messenger</a>
<pre><hr size="4" width="90%"><br>_______________________________________________<br>gmx-users mailing list <a class="EC_moz-txt-link-abbreviated" href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a><br><a class="EC_moz-txt-link-freetext" href="http://www.gromacs.org/mailman/listinfo/gmx-users">http://www.gromacs.org/mailman/listinfo/gmx-users</a><br>Please search the archive at <a class="EC_moz-txt-link-freetext" href="http://www.gromacs.org/search">http://www.gromacs.org/search</a> before posting!<br>Please don't post (un)subscribe requests to the list. Use the <br>www interface or send it to <a class="EC_moz-txt-link-abbreviated" href="mailto:gmx-users-request@gromacs.org">gmx-users-request@gromacs.org</a>.<br>Can't post? Read <a class="EC_moz-txt-link-freetext" href="http://www.gromacs.org/mailing_lists/users.php">http://www.gromacs.org/mailing_lists/users.php</a></pre>
</blockquote>
<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>