sozykinsa / GUI4dft

GUI4dft - first free SIESTA oriented GUI
MIT License
21 stars 6 forks source link

Transiesta electrode colocation #2

Open cdelv opened 3 years ago

cdelv commented 3 years ago

Hi, I'm writing this with advice. You really have to make the transiesta interface better. Adding the electrodes is a pain. You can only set the distance to the scattering region but you should be able to move the electrodes in x and y directions in order to center de molecule. The same with the molecule, you should be able to move it and rotated in any axis. As it is is very difficult to set up correctly the system. Thus more, for some reason, the fdf file generated after editing the system has this weird layout

`5 0.00000000 0.00000000 0.00000000 1 1 1

5 0.00000000 2.08565000 2.08565000 1 1 1

5 2.08565000 0.00000000 2.08565000 1 1 1

5 10.42825000 10.42825000 0.00000000 1 1 `

That prompts me errors all over the place. It should be something like this `%block AtomicCoordinatesAndAtomicSpecies

-1.44143472 0.00010000 0.34942222 2

-1.36513472 0.00020000 -0.99077778 2

1.00796528 0.00030000 0.48742222 2`

I hope you keep up improving this project, it would be great.

sozykinsa commented 3 years ago

Hello! Thank you for your interest in my program. Your suggestion for the ability to move the parts is a great thought. I will definitely implement it in the near future. Regarding the last part of your post. Obviously, the desired format is incorrect. Refer to the settings: Settings->Prefered coordinates->Fractional

cdelv commented 3 years ago

Ah thanks, dint thought about the settings. The ability to move parts is important, not only for transiesta but also to build combinations of molecules. That's why being able to select a big portion of atoms and moving, editing them, or deleting them would be also nice. I'm thinking that another quality of life perk would be to have a menu where one can select the settings of the fdf, regarding each aspect of the program, structure, MD rutins types of solvers, etc. Right now everyone that could use this program will know how to use SIESTA but in the future, it may not be so obvious. I notice that there's a strange thing with the visualizations of bonds between different types of atoms. I'm working with cytosine and there are no bonds between nitrogen and carbon atoms.

sozykinsa commented 3 years ago

I made changes to move all the pieces and rotate the scattering region. Over time, I will implement your other suggestions.

sozykinsa commented 3 years ago

Thank you for your comment on bonds rendering! I changed the logic for determining the lengths of bonds that should be drawn. Such a "standard" parameter for each pair of atoms can be viewed on the Settings->View->Bonds tab. If any of the values used seem to be incorrect, write to me. I will change the default settings.