[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ESPResSo-users] Doubt regarding integration steps

From: Nairhita Samanta
Subject: Re: [ESPResSo-users] Doubt regarding integration steps
Date: Mon, 25 Apr 2016 11:38:06 +0530

Hi Peter!
          Thank you for your reply.

I ran a simulation the way you mentioned.
I got these following results
is: 0.021, should_be: 0.021 ratio:1.0

I have understood it now, and initially I was also thinking that total elapsed time of the simulation should be
[expr $int_n_times*$int_steps*$time_step], which is correct. So I am happy.


On Sun, Apr 24, 2016 at 12:05 PM, Peter Košovan <address@hidden> wrote:
Hi Nairitha,

each time step is $time_step long, each call co integrate executes $int_steps integration steps, and integrate is called $int_n_times times.

Try the following modification to your script:

# let's make these small prime numbers, so that their product is unambiguous
set int_n_times 3
set int_steps 7
set time_step 0.1000
thermostat langevin 1.0 1.0

setmd skin 0.400
setmd time_step $time_step
setmd time 0
# start main integration
for {set i 0} { $i < $int_n_times } {incr i} {
     integrate $int_steps
set should_be [expr $int_n_times*$int_steps*$time_step];
set is [setmd time];
puts "is:$is, should_be: $should_be ratio:[expr $is/$should_be]"

Hope that helps,


On Sat, Apr 23, 2016 at 3:53 PM, Nairhita Samanta <address@hidden> wrote:
Hi Konard!
         Thank you for your reply. I am still not very clear about it. Please see the following, that is  the integration details in my script. Then where do I put the count which you have suggested? As I said I have two variables, in each $int_n_times the system is integrated by $int_steps*$dt.

for {set i 0} { $i < $int_n_times } {incr i} {

 integrate $int_steps
  integrate 0

  incr j


With regards,

On Sat, Apr 23, 2016 at 6:38 PM, Konrad Breitsprecher <address@hidden> wrote:
So the timestep is set by
setmd time_step $dt
Now when you call 
integrate $N
the system is propagated by 
The total elapsed simulation time is tracked internally, you get (or set) it via
[setmd time]
So a convenient way of handling time in Espresso is to call 
setmd time 0
before your integration loop and obtain the elapsed time also via [setmd time].
You should also know the conversion to SI time. For example, if your units are 
Length: [Angstrom], Mass: [g/mol], Energy: [kJ/mol], 
then you elapsed time in SI units is [setmd time]*100 fs.

Hope this helped, 

2016-04-23 14:30 GMT+02:00 Nairhita Samanta <address@hidden>:
Hello all!
               I am little confused here understanding the time-steps of integration in Espresso. When we run a script, we generally have three parameters "time steps" (generally very small), "integration time" and "integration steps". Now while integrating, the system is being integrated "integration time" times and each of them has certain number of "integration steps" and and all the steps get integrated by dt="time steps".

Now my question is if I want to report my simulations somewhere then what should be the total time of integration? Is it

("integration time" X "integration steps" X "time steps") in whatever is my unit system.

I feel this is the most logical.


Dr. Peter Košovan

Department of Physical and Macromolecular Chemistry
Faculty of Science, Charles University in Prague, Czech Republic

Katedra fyzikální a makromolekulární chemie
Přírodovědecká fakulta Univerzity Karlovy v Praze
Tel. +420221951290
Fax +420224919752

reply via email to

[Prev in Thread] Current Thread [Next in Thread]