PH13 - Visualization – Conversation in The Periodic Table

PH13

Jul 14, 2015 01:55

hi :)

Brian

Heyo!

Sorry, was reading the chat history

Poor bacteria.

PH13

^^

I don't know if they still live

Brian

Well, nice thing about 'em, is there's almost always more where those came from.

So, yeah, just about everything I've tried ORCA's Molden files with has not liked them: VMD, CASINO (quantum Monte Carlo: vallico.net/casinoqmc).

PH13

it seems as they were hard to cultivate ... so I would not be too sure about it g

at home I use ORCA and in office the molden files come either from Molcas or from Turbomoles tm2molden

Brian

MultiWFN likes them okay, but only because they wrote a special filter for them

PH13

Jul 14, 2015 01:59

xD

know your enemy

Brian

hehe

VMD claims to support Molden files -- it may or may not be happy with those from Molcas / Turbomole -- only way is to try

But it opens a lot of other formats, too, so I'd try everything those packages are able to export to

I'm in process of expanding my answer on OP, with an actual step-by-step for ORCA .plt files

PH13

just as you posted this answer I looked if VMD was still installed and it was ... but the renderer doesn't work ... so I stopped this project again very fast

that's great

Brian

Ugh - OpenGL problems?

PH13

nope, more the vmd-windows-connection thing ... I guess

there is this line with "explorer %s" which he doesn't like that much

Brian

That reminds me - Gerald Knizia's group has made another viewer, iboview (iboview.org), that uses some fancy OpenGL stuff that my OEM video card doesn't like very much

weird

PH13

Jul 14, 2015 02:02

Iboview is my newest tool ... so fancy

that's how I made my profile picture

Brian

METALLIC ORBITALLLLLSSSS

PH13

one of the most awesome programs ever

Brian

I was sad - I was looking forward to having tool at hand that's implemented the Knizia intrinsic orbitals

PH13

but I still have to figure out the other abilities of iboview ... not just the fancy orbitals

and what I don't really like about it is that it does not support every basis set

Brian

Eek, that is not so good.

Because of basis functions with too-high angular momentum or something?

PH13

Jul 14, 2015 02:07

pople basis sets are bad

def2 are good

I don't remember the reason ... I will have a look

Brian

Neese is HUGE on the Ahlrichs def2's. Considers their polarization coverage to be much more balanced, I think it is.

Can't be high angular momentum, then, if Pople sets are bad -- if anything, they're under-polarized / under-high-momentummed.

PH13

ok ... the error is about something else ... "Sorry, this program does not support Cartesian basis functions. Please use a spherical harmonic basis (e.g. def2-* or cc-pVnZ bases)"

Brian

Ahh, <nod>, that makes sense. Implementation is completely different for both of those

I thought ORCA's bases were laid out in Cartesian, though... maybe that's part of the problem? I'd have to check to be sure.

PH13

no, I can do it with whatever molden file I have

but I mean ... they all come either from orca, turbomole or molcas ^^

Brian

Ah, nope, ORCA's default is spherical harmonics

Manual p283

Are you seeing the 'does not support Cartesian' error if you use Pople bases in ORCA?

(ORCA/Pople --> Molden --> iboView)?

PH13

Jul 14, 2015 02:15

I have to try it. But maybe then there is also a Linux -> Windows problem?

Brian

Oh, very possibly

make sure you run 'unix2dos' on all files before transferring to Windows

PH13

in my molden->avogadro problem nothing changes with changing linebreaks

Brian

<nod>, but it depends on the software

PH13

not directly ... but maybe the error is somewhere between the lines

Brian

for example, if I work up an ORCA input file on Windows, copy to my Linux server, then try to run w/o 'dos2unix'ing it, ORCA chokes badly

PH13

Jul 14, 2015 02:16

I run "unix2dos" manually through notepad++ ... does it depend "where" I convert them?

that's true

Brian

dunno, I've only ever done it on the Linux side

PH13

ok

Brian

It is (or should be) an unambiguous transformation... but who knows.

PH13

^^

Avogadro has also another funny thing ... I drop my first molden file in it, then there are all thos atoms missing. When I throw in another molden file it crashes ^^

Brian

<nod>

PH13

Jul 14, 2015 02:18

I really want to use it, but it fights back as much as it can

Brian

All respect to Geoff & team for putting something together as ambitious as Avogadro, but it definitely has some stability issues

PH13

also when I want to use this pH stuff ... I don't really know how to use it, but pressing the buttons makes it crash

sure ... it has great potential

Brian

I would guess that might only work on things like proteins

PH13

yeah, maybe you're right

maybe Geoff knows more when he answers my e-mail

Brian

I really have made a lot of use of the SMILES fragment input.

Find what I want on Wik, copy the SMILES, dump it in, and boom: the molecule I want.

PH13

Jul 14, 2015 02:22

as you mention it ... the conformational search also fails for bigger molecules and makes everything crash :D

Brian

I've even magicked up some other stuff just out of my own head. SMILES is pretty helpful

hehehe, well, that might just be a 'bigness' problem. :-)

Am curious -- whence 'PH13'? You feel alkaline most of the time

?

PH13

it does a good job until somewhat around 80-90% and that it's enough

I used SMILES only fr conformational searches with RDKit

no ... I couldn't think of a better name ... and my usual name could be found to easy throughout the web so I needed something new to be at least some sort of anonymous }:->

Brian

I debated anonymity, then decided against it. Does mean I probably won't post much on Academia.SE or Workplace.SE, but I'm okay with that.

PH13

I have always big concerns about what I say and I don't like to be wrong. So I don't give my name that easy away that it doesn't make me feel awkward too fast and too easy.

it's awkward enough to be wrong ... xD

then not the whole internet has to know about who was wrong in detail

Brian

:-)

I've decided it's better for my ego to have my being-wrongs plain to see

PH13

Jul 14, 2015 02:29

that's great :)

Brian

Ok, gotta get this VMD step-by-step finished, then get to bed. Good luck wrestling your visualizations!

PH13

do this :)

thanks ^^

and good night

Brian

g'night!

PH13 - Visualization

The Periodic Table

Participants