[AG-TECH] AG tools on SPARC/SOlaris platforms?

[Ivan R. Judson]

> Well I'm sitting at a Sun desktop at the moment so I'm 
> interested but its difficult to justify too much effort in 
> this direction.

How come it's hard to justify the effort? I'd be interested in the
reasoning...
 
> > There seem to be several technical issues:
> > 
> > 1) The Python-based venue management software *ought* to be 
> portable to
> >    Suns -- what are the likely technical hurdles?
> > 2) The principal video capture device used previously on 
> Suns, the Osprey
> >    1500, is no longer made (though the company says it 
> would issue some
> >    on special order if there were enough interest).  Newer Suns have
> >    USB connections, and I've seen mention on the Web of cameras that
> >    supposedly operate on Sun USB ports, but have no direct 
> experience.
> >    Has anyone had experience using USB-based video cameras on Suns,
> >    and how difficult would it be to support such devices within vic?
> 
> I also came across a port of the bttv drivers for solaris 
> http://bt848x.sourceforge.net/ that broaden the range of 
> possible capture cards though the VIC source may need 
> tweaking i've no direct experience.
> 
> > 3) Given that vic and rat were born on beige boxes, it ought to be
> >    straightforward to build AG versions (with the AES 
> encryption and any other
> >    AG enhancements) on Suns -- are there hidden technical 
> trip-ups lurking
> >    here? 
> 
> I have used the UCL sparc versions to monitor AG sessions 
> from my desk The one issue I have seen is some Solaris sound 
> capture setups seem to impose a very long time delay. Its 
> probably worth checking manually run vic/rat works ok before 
> pushing ahead with the python stuff.

It's a good idea to get the basic C/C++ based media tools verified, but that
last sentence seems to imply there's a performance relationship between the
media tools and python layer of the toolkit which is quite untrue. The most
interaction that occurs is equivalent (if not exactly) forking and killing
the media tool processes, which should transfer any performance issues from
the python layer to the tools.

Cheers,

--Ivan