From: Axel Kohlmeyer (
Date: Wed May 20 2009 - 09:38:40 CDT

On Wed, 2009-05-20 at 14:11 +0200, MyLinkka wrote:
> Hi,

hi ting,

> After a tryout of the latest version, I think beta 3 is great. However,
> I still want to
> say some of the 'drawbacks' (or should be better to use 'dislikes') of
> vmd. I hope I would not offend people here.

you should keep one thing in mind: the best way to get changes
implemented in open source software packages is to do it yourself
and share the results. to ask for is only the second best option.

> 1. The default colors. In my very personal opinion, the default colors
> for atoms are 'unusual'. Especially, the cyan for carbon atoms, very special. :) I
> know no color is right for a specific atom type. But most of the molecular
> visualization program will take 'grey' at default color for carbon. So I think there
> is or should be some convention on this.

VMD is highly customizable. please have a closer look at the
documentation and various examples floating around. for example,
you can reset the colors to whatever you want via scripting.
i have in my .vmdrc (take vmd.rc if you a sucker for windows) file:

color Name C green

in mine and to me the carbons look just the way i like it, green ;-)

> 2. Python support on windows. I love python and Windows. (this doesn't mean
> I don't like Linux. ;) ) I don't intend to compare Windows and Linux
> here. But I wish vmd could support Python scripting on windows. I know there are some
> reasons on the website why python is not supported. But as a windows user, I
> know 'thousands' of software supporting python scripting out there. Why
> cannot vmd?

are you volunteering to help?? the problem with python in VMD
seems to be that many people want to use it, but next to nobody
wants to work on implementing, improving, documenting and
maintaining it.

also, you have to keep in mind that VMD was designed initially
with tcl scripting in mind and having two(!) script interpreters
embedded into the same program at the same time _and_ keeping
them consistent with each other is not as easy as it looks.

> 3. Window layout. I know there is some tradition for L(u)nix software
> that many separated and modeless windows belong to one program.
> Normally on windows, in this case a program will create some child
> windows. (if you check what gimpshop project (and gimp) does, probably
> you will understand this. ) Vmd works in the unix way. But one
> annoying thing is, each time when I start vmd, all windows (console,
> gl and main window) are overlapped (at the upper-left corner of my

i don't know about windows, but under linux you can define the window
sizes and placement via environment variables and if i am not totally
mistaken, you should at the very least be able to do it via the tk
scripting interface via "wm geometry". you'll have to dig a little
to find a way to get the window id or window path.

> screen). I have to manually move and rearrange them in order to see
> all of them. (maybe this could be controlled in the startup file?) At
> this aspect, other software like pymol does better, it include almost
> all the functionalities in one window (although it has two, but you
> can input command in the 3d display window too) and the two windows
> are tiled nicely upon startup.

this is a matter of taste. i personally find the pymol GUI not
very appealing. you can control the various dialogs in VMD from
the OpenGL window via hotkeys (and you can define your own, too).
i find this very appealing as this way the GUI stays out of the
way until it need it, and you don't have cluttered and overloaded
interfaces. having separate windows the way VMD has it is
particularly useful with multi-screen setups (you can dedicate
one to the OpenGL window and the other to the controls.

> I'm sure above things are not all my 'complaints' about vmd.:) (I will
> add more gradually if you like.) But just because vmd is one of the
> best molecule visualization program, I wish it could be better.

as i wrote before the best way to make an open source program
like VMD better is to contribute to it. you would be surprised
if you see how few people are actually working on it and how
many other things these people are doing (or have to be doing)
apart from it. to me it is a miracle that VMD it is _that_
capable at all and manifest to the skills of john as a maintainer.

> Recently, there was a vote for the best molecular visualization
> program at
> the results are interesting pymol vs vmd 38% to 22%. I think vmd has

this is IMNSHO a very stupid competition. pymol and VMD address
problems in different ways and thus a best suited for different
purposes. doing these competitions seems to be a good way to
drag people into visiting a certain webserver, though... ;-)

> many advantages but loses in default settings, because most of the end
> users care only about the quality of image and user experiences
> (something different from technique).

when designing software there are two main competing trends:
1) to make it easy and intuitive to use from the beginning;
2) to make it powerful, fast, flexible and versatile.
to make something more easy and intuitive, you usually have
to take away flexibility or limit functionality and the
other way around. so one has to make choices. as VMD may not
be as appealing and "automatic" as other packages, it excels
in many ways and it will keep growing on you, if you keep
experimenting and working with it. so if you are feeling
occasionally frustrated, don't get discouraged and keep trying.
you will be rewarded in the long run.


> ting

Axel Kohlmeyer
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
If you make something idiot-proof, the universe creates a better idiot.