<div>Hi Erik and David,</div>
<div> </div>
<div>I really appreciate your emails and that helped me a lot! My work is getting progress now.</div>
<div> </div>
<div>Thank you again</div>
<div>Chenyue<br><br> </div>
<div><span class="gmail_quote">On 9/4/06, <b class="gmail_sendername">Erik Lindahl</b> <<a href="mailto:lindahl@cbr.su.se">lindahl@cbr.su.se</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">HI Chenyue,<br><br>On Sep 3, 2006, at 1:39 AM, Chenyue Xing wrote:<br><br>> Hi David,<br>><br>> Thank you for your quick reply.
<br>> I tried setenv and compared 2 simulations with exactly the same<br>> initial structure and parameters (one w/ NOASSEMBLYLOOPS, one w/o it)<br>> But the results seem to be different. Is this normal?<br>><br>
<br>To compare different loops or architectures, try running a single<br>step and enable output of forces (use gmxcheck to compare them).<br><br>There are multiple reasons for differences. First, the SSE code is<br>performed entirely in 32 bit, modern compilers optimized for Pentium4
<br>and later mix 32 bit SSE and 64 bit SSE2 code, and old portable 387<br>code is 80 bits internally, 64/32 bits externally.<br><br>Depending on the order of operations generated by the compiler the<br>rounding errors will be different, and since MD is intrinsically a
<br>chaotic process the trajectories will diverge exponentially with the<br>number of steps. This is quite OK, since we are reproducing<br>statistical properties and not trajectories of individual atoms.<br><br>><br>> Also, if I have assemblyloops enabled (not setting noassemblyloops
<br>> 1), can I actually find those assembly codes to read?<br>><br><br>src/gmxlib/nonbonded/<architecture><br><br>Cheers,<br><br>Erik<br><br><br><br><br>_______________________________________________<br>gmx-developers mailing list
<br><a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a><br><a href="http://www.gromacs.org/mailman/listinfo/gmx-developers">http://www.gromacs.org/mailman/listinfo/gmx-developers</a><br>Please don't post (un)subscribe requests to the list. Use the
<br>www interface or send it to <a href="mailto:gmx-developers-request@gromacs.org">gmx-developers-request@gromacs.org</a>.<br></blockquote></div><br>