Author Topic: error  (Read 3781 times)

0 Members and 1 Guest are viewing this topic.

Offline qingfang

  • Heavy QuantumATK user
  • ***
  • Posts: 25
  • Reputation: 0
    • View Profile
error
« on: July 27, 2010, 05:41 »
I am a new user in using atk.I installed the 32bit atk in the linux systerm.but I must open the atk in the terminal window.and after the caculation ,it will appear "Not enough space for matrix"
so now ,how can I open the software in the desktop?
and what's wrong with my script or another things?

thank you very much!

Offline qingfang

  • Heavy QuantumATK user
  • ***
  • Posts: 25
  • Reputation: 0
    • View Profile
Re: error
« Reply #1 on: July 27, 2010, 05:43 »
and  I install the 64 bit atk in the linux systerm.but it will appear this sentence"it can't open the binary file"

Offline zh

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 1141
  • Reputation: 24
    • View Profile
Re: error
« Reply #2 on: July 27, 2010, 14:07 »
I am a new user in using atk.I installed the 32bit atk in the linux systerm.but I must open the atk in the terminal window.and after the caculation ,it will appear "Not enough space for matrix"
 
It clearly indicates that the memory of computer is not enough. This may be due to the possible reasons: i) the physics memory of your computer is indeed not enough; ii) the case under study may be due to too large and it can not be handled with a desktop computer and you had better run your job in a parallel computing way, e.g., in a cluster of computers.

Quote
it will appear this sentence"it can't open the binary file"

Maybe, you use a wrong command or way to run the code. Please refer to the tutorial of ATK to learn how to run this code.

Offline qingfang

  • Heavy QuantumATK user
  • ***
  • Posts: 25
  • Reputation: 0
    • View Profile
Re: error
« Reply #3 on: August 7, 2010, 03:57 »
right now,when i caculate the transmission spectrum,the temperature which i set is 800 k,and k points is (2,2,100).
but it appeared "contour Integration Error:********" in the log file,and it continued  to caculate,and then result.
so ,why?
the result is correct?


Offline Nordland

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 812
  • Reputation: 18
    • View Profile
Re: error
« Reply #4 on: August 7, 2010, 08:39 »
It is not a problem, but gives your a indication of the numerically accuracy for the given choice of parameters for the contour integral.