Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 16 Sep 1997 13:04:36 -0700 (PDT)
From:      lamaster@george.arc.nasa.gov
To:        multimedia@FreeBSD.ORG
Cc:        luigi@labinfo.iet.unipi.it
Subject:   Re: X11 video grabber for vic...
Message-ID:  <199709162004.NAA00339@george.arc.nasa.gov>

next in thread | raw e-mail | index | archive | help

luigi@labinfo.iet.unipi.it wrote:

> right now I am following the sigcomm session which is being
> transmitted with vat&vic ... I must say that I am _not_ impressed
> since 99% of the times vic is used to show slides and there is all
> kind of artifacts in acquiring slides using the camera. Much better
> results, at a fraction of the bandwidth, could be achieved by using
> an X11 grabber as it was available in nv.

If I understand your criticism of the broadcast, vic actually 
already contains the necessary functionality to do this properly.
The video operator may need to stop and consider what needs to be 
done in light of having to work with viewgraphs.

Now, vic does have a "sending slides" option, and I'm not actually
sure what it does.  A cursory view of the documentation doesn't say,
and it isn't obvious to me from switching it off and on.

OTOH, all it takes to transmit slides properly using H.261 
is to reduce the framerate (to, e.g. 1 fps) and increase 
the quality (to, e.g. 1).  Of course, the operator has 
to adjust these as the video changes from speaker to slides
and back again, since 1 fps isn't a very interesting way 
to watch someone speak and move.

And, of course, the best solution is to use wb to transmit
the slides, (or, for small audiences, web pages - but, I 
like wb/multicast) and use the video for the speaker.  But that
presupposes that the slides have been prepared in advance...

Nobody is satisfied with a video camera pointed at a viewgraph
projector, but, since Mbone is usually an afterthought for 
most conferences, it is difficult to get anything better.
But, lack of technology is not the problem here.  If the 
speaker can't give the conference coordinator source,
the only option I know of is the camera on the projected 
images and set the framerate down and the quality up.

> Some time ago I started writing an x11 grabber for vic
> (http://www.iet.unipi.it/~luigi/grabber-x11.cc), but had to give
> up due to my miserable knowledge of X11 and C++ . The code I have
> is partly working but it has some rough edges and probably a lot
> of bugs and inefficiencies.

Do you mean a function that would grab an X11 window as a frozen
frame, or a changing series of frames?  I sure I'm missing something 
here, but, is this a totally separate question to the above, or
somehow related?  I don't remember what nv did, but, if the 
problem is that the presenter doesn't have computer-available 
slides to begin with, I don't understand how an X11 grabber will
help.  What format would the original material be in.  Not postscript,
or ASCII, or, you would just use wb, and most formats can be 
converted to postscript.  e.g. SGI's "showcase" actually does
a pretty good job of formatting its own slides for postscripts.
PS slides with images should be handled one at a time and run
through "lzps".

I think wb is the short-term solution if source material
is available.  The wb version for FreeBSD seems to work
fine.  (Better, actually, than wb on the SGI, which still
occasionally crashes or hangs or somehow ties up the X
server - not that this is criticism of wb.  SGI still seems
to have display-postscript-related bugs even in Irix 6.2, 
though not as bad as in Irix 5.2/5.3).

> Would someone like to help in cleaning up the above code (which is
> partly working) ?


Corrections welcome.

-Hugh LaMaster



  Hugh LaMaster, M/S 258-5,     ASCII Email:  hlamaster@mail.arc.nasa.gov
  NASA Ames Research Center     Or:           lamaster@nas.nasa.gov
  Moffett Field, CA 94035-1000  No Junkmail:  USC 18 section 2701
  Phone:  415/604-1056          Disclaimer:   Unofficial, personal *opinion*.




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199709162004.NAA00339>