Abinit vs SIESTA vs QuantumEspresso  [SOLVED]

External optional components of Abinit (BigDFT, Wannier90, ...) and codes using / providing data from / to Abinit (AtomPAW, EXC, DP, SAX, WanT, Yambo, ...).

Moderator: rangel

Post Reply
hatuey
Posts: 14
Joined: Tue Mar 23, 2010 3:35 pm

Abinit vs SIESTA vs QuantumEspresso

Post by hatuey » Thu Sep 08, 2011 1:22 pm

Hello all,

I am new to periodic system modeling. When we look for codes that can help studying such systems, we find Abinit, Siesta, QuantumEspresso among others.

My questions:

1- Why should I use "this" code or "that" code? (in this forum, "this"code is Abinit :)
2- Specifically, which are the advantages and main differences to use Abinit over Siesta?

Regards,

Hatuey

User avatar
sponce
Posts: 60
Joined: Sat Apr 16, 2011 9:44 am

Re: Abinit vs SIESTA vs QuantumEspresso

Post by sponce » Sat Jan 07, 2012 2:50 pm

Dear Hatuey,

Up to my knowledge SIESTA is more oriented toward quantum transport whereas Abinit and PWSCF are more oriented towards solid bulk properties.

If you have to choose bewteen Abinit and PWSCF I would say that both have their advantages and drawback and it strongly depends on what properties you want to simulate.

I would say that the main advantage of Abinit for a new user would be the very good tutorials.

Hope I help you

PS: Please note that this is a biased answer as I'm mainly working with Abinit...

hatuey
Posts: 14
Joined: Tue Mar 23, 2010 3:35 pm

Re: Abinit vs SIESTA vs QuantumEspresso  [SOLVED]

Post by hatuey » Sat Jan 07, 2012 3:21 pm

Thank you very much!

mverstra
Posts: 655
Joined: Wed Aug 19, 2009 12:01 pm

Re: Abinit vs SIESTA vs QuantumEspresso

Post by mverstra » Tue May 27, 2014 2:05 pm

I would add that SIESTA goes to much larger systems but you have to be very careful with the accuracy, and often need to optimize the pseudopotential _and_ the basis set to make sure you get the physics of the system right. Convergence with QE and abinit is more straightforward.

I would agree with Samuel about the tutorials, and add (purely subjective) that the use of abinit is easier - error messages, inputs, post-processing. The QE package includes a large number of different codes and not all of them interrelate easily. Also the extent of the PAW implementation in certain advanced features is not clear to me, but this is just because I use abinit much more than PWSCF.

Try them! See when you get fed up and which gives you the best results for the lowest effort...
Matthieu Verstraete
University of Liege, Belgium

hatuey
Posts: 14
Joined: Tue Mar 23, 2010 3:35 pm

Re: Abinit vs SIESTA vs QuantumEspresso

Post by hatuey » Tue May 27, 2014 2:48 pm

Thank you very much!

Post Reply