Re: [ESPResSo] PSAW

[Mikheil Azatov]

Hi,

I followed the code of polymerC somewhere up to the line 580(i'm looking at the 2.2.0b version). If angle is specified I see that the new positions are found and stored in "pos" at line 550.  As I understood at this point of the code the new positions found are correct.  After that there is  a part of the code where the positions are found if the angle was not specified. Next there are checks to find find if any of the constraints or overalaps are violated. Now if they are violated the code tries to find the new positions. So as I understand because variable "pos" is already changed the vector b at line 550 will be added to "pos" again and again in the next steps of the loop. And that's why it would get overstretched. is it the right understanding of the error? 

I fit is, the way to fix it would be to subtract b again. I'm thinking line 577 after "break" in the same "if" but i'm not sure because I haven't figured out that part of the code well. 

Please let me know if I'm correct:)

Thanks for your help,

Mikheil Azatov

On Fri, Oct 29, 2010 at 10:19 AM, Axel Arnold <address@hidden> wrote:

Hi!

Looks like there is a bug in the polymer command; if an angle is specified,
only the SAW works, since if a position is rejected due to constraints or
overlap, the bond will be overstretched for PSAW and RW. The bug is around
line 552, compare the correct angle-free implementation right below.

Axel

On Friday 29 October 2010 04:20:04 you wrote:
> Hi, I attached the code below. I might have figured out the problem already
> though. When I use the polymer command without angle feature it seems to
> work normally but with angle I receive weird results like i explained in
> previous emails. So, my guess is that RW doesn't work with angle feature in
> polymer command. Just wanted to check if I'm correct or it's something
> else.

--

JP Dr. Axel Arnold Tel: +49 711 685 67609
ICP, Universität Stuttgart      Email: address@hidden
Pfaffenwaldring 27
70569 Stuttgart, Germany