Firefly and PC GAMESS-related discussion club

Learn how to ask questions correctly  
We are NATO-free zone

Re^9: MCSCF for a large system.


>Dear Sanya,
>Your approach to manual writing is very extraordinary: give users to get lost in the ugly documentation first and then fix the places where ones stack the most. Itís the approach if one wona to melt stuff into pure diehard qchemists and a slag but would be a bit repulsive to an average novices.

Did you expect a ready-to-use solution handed on a silver platter? Sorry, we've run out of silver platters.
Actually, science is full of problems for which no ready-to-use solutions exist.

Would you like to join the team? Would you like to spend your time writing manuals, answering questions on forum, and seeing reproaches as an acknowledgment? Do you think we do this because we have nothing else to do? Do you think we're paid for this?

And anyway, experience is the best way to study. I can tell 100500 times that wstate should be chosen once and laid off for the entire study. But you'll never believe me unless you try and get meaningless results.

>Understanding that you had passed through the above Ďmetallurgyí, I am emphasizing the need for a good scientist to stay open-minded (either it is religion in science or wstates balance problem). Itís of course up to you: join the following discussion or mark the problem as a wasting but I am screening the transition of reaction with two wstate variation followed by geometry optimization. The results are expected by Mon/Tue, it will a separate topic. Why not to check how the variation looks like in energy terms of a real problem, are the routes through wstates make any difference or itís just a matter of mkcals etc. In perspective, with geometry statistics from ďa MD on the PESs parameterized by QMĒ we will see what states combination the Lord (or just the protein :) ) prefers at each reaction coordinate.

OK, try and see that I was right. And don't ask me to help with the analysis of your results. I just try to talk you out of wasting computer time for the study you won't be able to publish (the reviewers are not born yesterday).

>P.S. Thanks for the suggestion about reaction force field. Its however maybe not the best way to proceed. We donít need a field for model a chemical reactions as already making the FF which is tuned to accurately model only this particular reaction (this what the overwhelming citation above stands for, isnít it?). Of what we really need is a good exploration of protein configuration space. Thus GROMOS which is about protein-solvent equilibriums would be superior over FFs focused on reaction description.

Actually, I meant the papers about surface hopping dynamics, that is, applying MD to the study of processes in which Born-Oppenheimer approximation does not hold.

[ Previous ] [ Next ] [ Index ]           Fri Oct 3 '14 11:44pm
[ Reply ] [ Edit ] [ Delete ]           This message read 451 times