VMD-L Mailing List

From: Daniel Hornburg (Hleinad_at_gmx.de)
Date: Wed Aug 18 2010 - 02:17:49 CDT

That sounds a little bit like what I had initially in mind. Thanks for the
detailed explanation. I will contact the guy who is in charge for the
projectors and see what he suggests. What you suggest about the modification
of video output module sounds nice. Am I right that this would create
separate video output streams that can be used by different projectors?
Sorry I am just beginning to grasp this whole issue;)

I will keep you informed about this projector stuff.

Cheers,
Daniel

-----Ursprüngliche Nachricht-----
Von: owner-vmd-l_at_ks.uiuc.edu [mailto:owner-vmd-l_at_ks.uiuc.edu] Im Auftrag von
John Stone
Gesendet: Dienstag, 17. August 2010 20:48
An: Shore, Jay
Cc: Daniel Hornburg; vmd-l_at_ks.uiuc.edu
Betreff: Re: vmd-l: VMD 3D movies rendering

Hi,
  Yes, if you set up your projection system with a xinerama type arrangement
where both projectors are treated as a single windowing system
desktop/display, you could encode a movie for the full size of the whole
combined desktop/display, and have the left and right eye images tiled
appropriately so they end up on the correct projector.

This would likely only work well for a full-screen full-resolution playback
mode, as any scaling or letterboxing might create issues that a typical
video player can't deal with. So, assuming this approach, the main
remaining challenge is to pick a video codec that can accept such high
resolution material. For an HD stereo projection rig, you'd be encoding
frames that are
3840 x 1080 resolution (e.g. 2x 1920x1080). One can probably get free
MPEG-2 or MPEG-4 encoders to do this with some hammering and duct tape.
The main drawback is that this type of video wouldn't be very portable to
other systems, but that may or may not be a concern for the original poster,
so if they are willing to do it this way, it can probably be made to work.
This has the distinct advantage of not creating new video/audio
synchronization issues. One could also use one of the above/below formats
in the same way with the same limitations.

It isn't all that hard to write a modified video output module for an
existing movie player that would take one of these formats (particularly
above/below, due to its organization in memory) and apply necessary scaling
and letterboxing to the video and play it back on a variety of hardware, but
so far nobody has done this for the free players, as far as I'm aware. I
would volunteer to do it, but this would be some weeks from now at least as
I'm currently swamped.

Cheers,
  John Stone
  vmd_at_ks.uiuc.edu

On Tue, Aug 17, 2010 at 01:34:31PM -0500, Shore, Jay wrote:
>
> If you were using a geowall set up (two projectors each with a
> polarizer), couldn't you just make each frame of the movie contain the two
images side-by-side and then play the movie using a standard movie player,
quicktime etc.?
>
> A geowall that I have used split the image in half, sending one
> through one projector and the other half through the other projector.
>
> I apologize if I missed something.
> Jay
>
>
>
>
> ________________________________________
> From: owner-vmd-l_at_ks.uiuc.edu [owner-vmd-l_at_ks.uiuc.edu] On Behalf Of
> Daniel Hornburg [Hleinad_at_gmx.de]
> Sent: Tuesday, August 17, 2010 10:50 AM
> To: 'John Stone'
> Cc: 'Axel Kohlmeyer'; vmd-l_at_ks.uiuc.edu
> Subject: AW: vmd-l: VMD 3D movies rendering
>
> Ok, now I understand the main problem :) Well, I will have a look on
> the problem as soon as I get the projectors.
> Thanks a lot so far!
> Cheers,
> Daniel
>
> -----Ursprüngliche Nachricht-----
> Von: John Stone [mailto:johns_at_ks.uiuc.edu]
> Gesendet: Dienstag, 17. August 2010 16:40
> An: Daniel Hornburg
> Cc: 'Axel Kohlmeyer'; vmd-l_at_ks.uiuc.edu
> Betreff: Re: vmd-l: VMD 3D movies rendering
>
>
> Hi,
> What I think you didn't grasp from my previous email is that none of
> the different methods that I described can be played by freely
> available movie player software. At present, only custom-written
> software can play such movies, and so far as I know, nobody has made
> modifications to the commonly used movie players to enable these
> capabilities yet. Also, of the various methods I described, playing
> two independent movie streams is actually by far the most technically
> challenging due to the need to drive two separate decoders, and to
maintain synchronization, frame counts, etc.
> When you get access to the 3D projector then you can start looking at
> what software is available to you, and you can feel free to contact us
> to ask about rendering from VMD in whatever format your movie
> encoder/player software will support.
>
> Cheers,
> John Stone
> vmd_at_ks.uiuc.edu
>
> On Tue, Aug 17, 2010 at 12:43:35PM +0200, Daniel Hornburg wrote:
> > Hi thanks for your suggestions!
> >
> > I have not yet access to the 3D projectors and the corresponding
software.
> > As far as I know the technology I am going to use will use 2
> > projectors and shutter glasses. Therefore I think two separate video
> > streams each for one projector will be the best solution. Because in
> > the stereo view both perspectives are displayed correctly I was
> > wondering whether one can easily first extract one and then the
> > other perspective of the molecule. If that is not the case would it
> > be sufficient to just turn the molecule for some degrees and that?s
> > it? I am
> not certain about it.
> >
> > Cheers,
> > Daniel
> >
> > -----Ursprüngliche Nachricht-----
> > Von: owner-vmd-l_at_ks.uiuc.edu [mailto:owner-vmd-l_at_ks.uiuc.edu] Im
> > Auftrag von John Stone
> > Gesendet: Montag, 16. August 2010 18:20
> > An: Axel Kohlmeyer
> > Cc: Daniel Hornburg; vmd-l_at_ks.uiuc.edu
> > Betreff: Re: vmd-l: VMD 3D movies rendering
> >
> > Hi,
> > The issue with making movies for 3D projectors is that there
> > currently isn't a low-cost encoding mechanism for this type of media.
> >
> > There are a bunch of competing proprietary solutions, but nothing
> > that you can reliably use with off-the-shelf software.
> > I do occasional consulting for a business that develops various
> > production tools for Hollywood, and one of the tools I helped
> > develop previously was a 3D movie player that works with the
> > standard digital files that the movie industry uses. All of the
> > code I wrote was 100% custom because there still really aren't any
> > widely-adopted standards other than for Blu-Ray. The most common
> > free/low-cost methods are
> currently:
> > - encode 3D as two separate movie bitstreams
> > - encode 3D as a single movie bitstream, using above-below image
> > formatting
> > - encode 3D as a single movie bitstream, using side-by-side image
> > formatting
> > - encode 3D as a single movie bitstream, using frame-sequential
> > formatting
> >
> > All of these require special player software...
> >
> > The new Blu-Ray 3D format uses H.264/AVC with multiview video coding.
> > So far as I know, there are currently no free or inexpensive
> > MVC-capable movie encoders in existence. Most of these things
> > currently cost >= $5,000 last I checked. At present, the only way
> > to do 3D movies is to either use expensive commercial encoders and
> > players, or roll-your-own with your own custom player.
> >
> > Cheers,
> > John Stone
> > vmd_at_ks.uiuc.edu
> >
> > On Mon, Aug 16, 2010 at 09:40:36AM -0400, Axel Kohlmeyer wrote:
> > > hello daniel,
> > >
> > > On Mon, Aug 16, 2010 at 6:07 AM, Daniel Hornburg <Hleinad_at_gmx.de>
wrote:
> > > > Hello everybody!
> > > >
> > > >
> > > >
> > > > I am using vmd for visualization of trajectories. Since the ray
> > > > tracing movies are quite nice and I have the opportunity to use
> > > > a polarized 3D system I would like to produce high resolution
> > > > movies suited for 3D visualization (real time is not an
> > > > opportunity due to slow rendering of my system).
> > > >
> > > > However when I choose stereo views like side by side and start
> > > > rendering I only get pictures of a single view.
> > >
> > > yes. that is how it works. the stereo display settings only affect
> > > the OpenGL output drivers.
> > >
> > > > Does anyone of you have created movies for 3D projectors and may
> > > > help
> > me?
> > >
> > > i have been thinking about it and may give it a shot for a demo at
> > > the
> > > GTC'10 conference (if there is time). the question is not so much
> > > about rendering (you can easily render the left and right eye
> > > images separately), but about compression and playback. what kind
> > > of encoding would you want to use and what player software would
> > > you want
> to use?
> > > there are no widely accepted standards so you'd have to use some
> > > kind of proprietary solution or implement one yourself.
> > >
> > > cheers,
> > > axel.
> > >
> > > >
> > > > Thanks a lot!
> > > >
> > > > Cheers,
> > > > Daniel
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> > >
> > > --
> > > Dr. Axel Kohlmeyer akohlmey_at_gmail.com
> > > http://sites.google.com/site/akohlmey/
> > >
> > > Institute for Computational Molecular Science Temple University,
> > > Philadelphia PA, USA.
> >
> > --
> > NIH Resource for Macromolecular Modeling and Bioinformatics Beckman
> > Institute for Advanced Science and Technology University of
> > Illinois, 405
> N.
> > Mathews Ave, Urbana, IL 61801
> > Email: johns_at_ks.uiuc.edu Phone: 217-244-3349
> > WWW: http://www.ks.uiuc.edu/~johns/ Fax: 217-244-6078
> >
>
> --
> NIH Resource for Macromolecular Modeling and Bioinformatics Beckman
> Institute for Advanced Science and Technology University of Illinois, 405
N.
> Mathews Ave, Urbana, IL 61801
> Email: johns_at_ks.uiuc.edu Phone: 217-244-3349
> WWW: http://www.ks.uiuc.edu/~johns/ Fax: 217-244-6078
>
> 

--
NIH Resource for Macromolecular Modeling and Bioinformatics Beckman
Institute for Advanced Science and Technology University of Illinois, 405 N.
Mathews Ave, Urbana, IL 61801
Email: johns_at_ks.uiuc.edu                 Phone: 217-244-3349
  WWW: http://www.ks.uiuc.edu/~johns/      Fax: 217-244-6078