tjm_at_Princeton.EDU writes:
>...disregard last message...a reminder that if you are running sander with natoms > 99999
>you need to change the format specification in the subroutine ew_setup.f on line 1855
>
>from
>
>9018 format(i5,5e15.7)
>
>to
>
>9018 format(i6,5e15.7)
>
>(or i7 if you are really ambitious....)
I was wondering, will future versions of AMBER have a better input file
format which reduces the arbitrary restrictions on system sizes? I
know this has been talked about before, but there's been some
reluctance to actually change the file format. Given that 2-4GB of
RAM per node is not unreasonable, and coupled with the increase in
speed of nodes, and interconnect, system sizes of > 100,000 atoms are
quite within the realm of acheivable simulations.
Dave
Received on Mon Sep 04 2000 - 09:03:55 PDT
