Hi David
Thank you for your response.
>>1. Did the mdout file offer any hints? Did the minimization conclude correctly?
Yes, the mdout file indeed showed some results, as shown in the attached file, the minimization "RESTARTED DUE TO LINMIN FAILURE"
As I searched from the forum, the discussion indicates that that this message is not a fatal error and would be expected.
>>2. I'd double-check all your filenames: is it possible you are using the same name twice?
I've attached the submission file, may you help me to check whether it is legit?
>>3. Running it on the CPU mode
I'm not sure how to identify whether the gradient is large or not, may you please give me more information on it?
Much appreciated.
Kind regards
Yi Ren
-----Original Message-----
From: David A Case <david.case.rutgers.edu>
Sent: Sunday, January 9, 2022 12:43 AM
To: AMBER Mailing List <amber.ambermd.org>
Subject: Re: [AMBER] Amber20 minimization, the rendered .rst file is corrupted and unreadable
On Sat, Jan 08, 2022, Yi Ren wrote:
>
>I encountered a rare problem while using the minimization function, I
>was trying to perform the minimization by using pmemd.cuda But the
>rendered .rst file is corrupted and unreadable, which stopped me from
>the subsequent works.
>
>May anyone help me to have a look at the attached file and identify the problem?
What you attached is clearly not a restart file. Some questions:
1. Did the mdout file offer any hints? Did the minimization conclude correctly?
2. I'd double-check all your filenames: is it possible you are using the same name twice?
3. Run a (shortish) minimization with a CPU version. If you have large initial gradients, the GPU version can fail. (But maybe this is not the problem, depending on the answer to question 1 above.)
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Jan 08 2022 - 17:00:02 PST
