<html>
<head>
<style>
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
</style>
</head>
<body class='hmmessage'>
Hi David,<br><br>I just checked with shake tolerance 1e-10 I really get deviations of 1e-10.<br>First I forgot to turn on Shake and got the default Lincs,<br>which with the default settings gave me roughly the accuracy you reported.<br>(you did not by coincidence make the same mistake?)<br><br>Berk<br><br><hr id="stopSpelling">From: gmx3@hotmail.com<br>To: gmx-users@gromacs.org<br>Subject: RE: [gmx-users] shake for water<br>Date: Thu, 5 Feb 2009 23:02:31 +0100<br><br>
<style>
.ExternalClass .EC_hmmessage P
{padding:0px;}
.ExternalClass body.EC_hmmessage
{font-size:10pt;font-family:Verdana;}
</style>
Hi,<br><br>I don't agree.<br>It uses the small 1+a approximation for the square.<br>Also mdrun prints the rmsd determined with independent code,<br>which is consistent with the (correct) tolerance.<br><br>Berk<br><br>> Date: Thu, 5 Feb 2009 22:41:47 +0100<br>> From: spoel@xray.bmc.uu.se<br>> To: gmx-users@gromacs.org<br>> Subject: Re: [gmx-users] shake for water<br>> <br>> Berk Hess wrote:<br>> > <br>> > <br>> > > Date: Thu, 5 Feb 2009 19:35:09 +0100<br>> > > From: spoel@xray.bmc.uu.se<br>> > > To: gmx-users@gromacs.org<br>> > > Subject: Re: [gmx-users] shake for water<br>> > ><br>> > > David Mobley wrote:<br>> > > > All,<br>> > > ><br>> > > > A quick question on constraints... I'm using TIP4P-Ew in gromacs 3.3.2<br>> > > > and am concerned with reproducing energies from another code very<br>> > > > precisely for several specific snapshots. I am doing a zero-step mdrun<br>> > > > of a setup with one small molecule and two tip4p-ew water molecules.<br>> > > ><br>> > > > Anyway, I have set the shake tolerance to 1e-12 in the mdp file, but<br>> > > > to my surprise the internal water distances are good only to 1e-06 and<br>> > > > 1e-07. Is this expected behavior? Note that I am running in double<br>> > > > precision. I assumed that, er, the distances should converge to the<br>> > > > shake tolerance.<br>> > > Well, the documentation might be lacking, but the code tells the truth.<br>> > > It seems that the tolerance is used on the distance squared, which is<br>> > > consistent with your observation of a precision of 1e-6. So try 1e-24.<br>> > <br>> > No, it is not the square.<br>> > The code does a small 'a' approximation: (1+a)^2=1+2a+a^2 is approx 1+2a.<br>> > I have also tested/benchmark shake in gromacs for my first lincs paper<br>> > and the plincs paper and it always behaved the way I thought it would.<br>> <br>> <br>> First we compute the inverse square of the shake distance dA in tt[ll] <br>> for each shake pair:<br>> <br>> if (bFEP)<br>> toler = sqr(L1*ip[type].shake.dA + lambda*ip[type].shake.dB);<br>> else<br>> toler = sqr(ip[type].shake.dA);<br>> dist2[ll] = toler;<br>> tt[ll] = 1.0/(toler*tol2);<br>> }<br>> <br>> Then in the shake iteration we compute the difference between the <br>> squared distances (variable diff below):<br>> <br>> tx = xp[ix]-xp[jx];<br>> ty = xp[iy]-xp[jy];<br>> tz = xp[iz]-xp[jz];<br>> rpij2 = tx*tx+ty*ty+tz*tz;<br>> toler = dist2[ll];<br>> diff = toler-rpij2;<br>> <br>> Now we multiply the diff with tt[ll], in other words we get<br>> iconv = (1-rpij2/dist2)/(2 tol)<br>> <br>> /* iconv is zero when the error is smaller than a bound */<br>> iconv = fabs(diff)*tt[ll];<br>> <br>> In other words, the tolerance operates on the squared distance.<br>> <br>> <br>> <br>> <br>> > <br>> > The problem is not that you need more iterations than 1000?<br>> > <br>> > And why not use settle that does it right at full precision at once?<br>> > <br>> > Berk<br>> > <br>> > <br>> > ><br>> > > ><br>> > > > Thanks,<br>> > > > David<br>> > > > _______________________________________________<br>> > > > gmx-users mailing list gmx-users@gromacs.org<br>> > > > http://www.gromacs.org/mailman/listinfo/gmx-users<br>> > > > Please search the archive at http://www.gromacs.org/search before <br>> > posting!<br>> > > > Please don't post (un)subscribe requests to the list. Use the<br>> > > > www interface or send it to gmx-users-request@gromacs.org.<br>> > > > Can't post? Read http://www.gromacs.org/mailing_lists/users.php<br>> > ><br>> > ><br>> > > --<br>> > > David van der Spoel, Ph.D., Professor of Biology<br>> > > Molec. Biophys. group, Dept. of Cell & Molec. Biol., Uppsala University.<br>> > > Box 596, 75124 Uppsala, Sweden. Phone: +46184714205. Fax: +4618511755.<br>> > > spoel@xray.bmc.uu.se spoel@gromacs.org http://folding.bmc.uu.se<br>> > > _______________________________________________<br>> > > gmx-users mailing list gmx-users@gromacs.org<br>> > > http://www.gromacs.org/mailman/listinfo/gmx-users<br>> > > Please search the archive at http://www.gromacs.org/search before <br>> > posting!<br>> > > Please don't post (un)subscribe requests to the list. Use the<br>> > > www interface or send it to gmx-users-request@gromacs.org.<br>> > > Can't post? Read http://www.gromacs.org/mailing_lists/users.php<br>> > <br>> > ------------------------------------------------------------------------<br>> > Express yourself instantly with MSN Messenger! MSN Messenger <br>> > <http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/><br>> > <br>> > <br>> > ------------------------------------------------------------------------<br>> > <br>> > _______________________________________________<br>> > gmx-users mailing list gmx-users@gromacs.org<br>> > http://www.gromacs.org/mailman/listinfo/gmx-users<br>> > Please search the archive at http://www.gromacs.org/search before posting!<br>> > Please don't post (un)subscribe requests to the list. Use the <br>> > www interface or send it to gmx-users-request@gromacs.org.<br>> > Can't post? Read http://www.gromacs.org/mailing_lists/users.php<br>> <br>> <br>> -- <br>> David van der Spoel, Ph.D., Professor of Biology<br>> Molec. Biophys. group, Dept. of Cell & Molec. Biol., Uppsala University.<br>> Box 596, 75124 Uppsala, Sweden. Phone:        +46184714205. Fax: +4618511755.<br>> spoel@xray.bmc.uu.se        spoel@gromacs.org http://folding.bmc.uu.se<br>> _______________________________________________<br>> gmx-users mailing list gmx-users@gromacs.org<br>> http://www.gromacs.org/mailman/listinfo/gmx-users<br>> Please search the archive at http://www.gromacs.org/search before posting!<br>> Please don't post (un)subscribe requests to the list. Use the <br>> www interface or send it to gmx-users-request@gromacs.org.<br>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php<br><br><hr>Express yourself instantly with MSN Messenger! <a href="http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/">MSN Messenger</a><br /><hr />See all the ways you can stay connected <a href='http://www.microsoft.com/windows/windowslive/default.aspx' target='_new'>to friends and family</a></body>
</html>