Mike,
That's not really true. Simulations with 0.5M is about as routine today as
20,000 atoms a few years ago. We are talking about protein complexes in
water. So, other people's code/parameter can't work. Naturally, you ask why
people waste time on doing atomic level simulations on such big systems. The
problem is that certain type of questions can be answered only with the
MD-type simulations.
I am not advocating or even suggesting that Bob or you guys work on this for
this release. But this will become an issue very soon, much sooner than
perhaps what we like. This is not visionary or anything like that. It is
reality already, unfortunately.
Indeed, most people run simulations with far smaller systems. I enjoy
simulations with 500 atoms :) and most of my simulations are small (<20K).
Again, there are important biomolecular/biophysical problems that can be
studied only by large-scale simulations.
yong
-----Original Message-----
From: owner-amber-developers.scripps.edu
[mailto:owner-amber-developers.scripps.edu] On Behalf Of Crowley, Michael
Sent: Tuesday, December 04, 2007 8:35 PM
To: amber-developers.scripps.edu
Subject: Re: amber-developers: Fw: How many atoms?
Hi Bob,
For my 2 cents,
We serve nearly all users with far less than 999,999 atoms and get very
little valuable information from runs of 1 million+ atoms at least with the
kind of systems we are running. It does make sense for folks who do
materials science looking at fracture propagation and the like, but there
are already codes that deal with that. I suggest your time is better spent
on features and <1G atom performance, rather than the million atom goal. The
million atom can wait for the next release with almost no loss of clients or
science.
Best wishes,
Mike
Received on Wed Dec 05 2007 - 06:07:39 PST
