<html>
<head>
<style>
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
</style>
</head>
<body class='hmmessage'>
<br><br>> Date: Wed, 17 Jun 2009 15:17:53 +0200<br>> From: erikm@xray.bmc.uu.se<br>> To: gmx-users@gromacs.org<br>> Subject: Re: [gmx-users] programmes to have in double precision besides mdrun_d<br>> <br>> Jussi Lehtola skrev:<br>> > On Wed, 2009-06-17 at 13:45 +0200, Berk Hess wrote:<br>> > <br>> >> Hi,<br>> >><br>> >> Nothing needs to be double precision.<br>> >><br>> >> Why do you want mdrun in double precision?<br>> >> The only common reason for this is normal mode analysis,<br>> >> in which case you need all the tools involved in double precision.<br>> >> For normal MD simulation there is nearly never a need for<br>> >> double precision.<br>> >> <br>> ><br>> > If you want to run accurate NVE simulations, double precision is<br>> > important. If you use thermostats, then there is no need for double<br>> > precision.<br>> ><br>> > <br>> I can testify to that. Without double precision I've had problems with <br>> drifting total energy when doing non-periodic NVE.<br>> <br><br>Indeed, this is correct.<br>But I guess this falls into the category nearly never.<br>On the other hand, there are so many Gromacs users nowadays,<br>there nearly never is nearly never zero users.<br><br>> /Erik<br>> > Also, I have found that double precision can be nice for energy<br>> > minimization since it can handle more pathological cases than single<br>> > precision. Still, usually it doesn't matter much whether the starting<br>> > point was prepared in single or double precision.<br>> > <br><br><br>That's true.<br>It would be nice if we could make the single precision code<br>handle such cases better (force capping?).<br><br>Berk<br><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>