Text 5198, 65 rader
Skriven 2006-02-06 16:27:18 av Maurice Kinal (1:261/38.9)
   Kommentar till text 5197 av Kenneth Parrish (8:8/2)
Ärende: Re: radeon tv-out and movie encoding experiments
========================================================
Hey Kenneth!

Feb 06 02:55 06, Kenneth Parrish wrote to Maurice Kinal:

 KP> [ 24 bpp radeon-framebuffers ]
 MK>> Not yet but I am ready.
 KP> what is the secret?

No secret.  I just turned on the radeon frambuffers in the kernel but I still
need to get a handle on the append line.  I figure with that one I'll just use
an 800x600 16-bit screen and see if I can output that to the tv-out to
something appropriate.  I have an old Commodore monitor that should be able to
handle it.

 MK>> I tested a small 25M mpeg4 formmated 'movie' I made using the same
 KP> how'd you edit it to the shorter length? did it start a capture from 
 KP> the wintv pvr card?

Nope.  It's just a couple minute wide screen animation with a stereo sound
track I threw together using the same parameters used to reformat the DVD
movie.  I figure this one takes less cpu time and can use it for testing
purposes without having to wait half a day to get any testable results.  :-)

Also it afforded me a way to test the flash disk idea without having to buy a
new +1G one.  It worked so I am confident that if needed I could get that idea
off the ground in short order.

 KP> ah. why wouldn't it?  :)

Lack of speed.  Turns out they are fast enough to play high resolution movies. 
Far better then DVDs and without the DVD structure which makes it far more
sensible.  They can be paused, fast forward, rewind, etc.  What more could
anyone want?

 MK>> to my theory about it all should look just as good as the 24-bit
 MK>> jobber on a tft set to display 16-bit graphics.
 KP> sounds reasonable :)

Offhand I doubt anyone could tell the difference just by watching but I'll bet
the targetted computer will have less problems with the trimmed down version. 
However I'll bet you could tell the difference on a high definition TV but that
isn't the goal in this particular exersize.  I think for those something better
will be needed but I think it will be taken care of in the near future.  Myself
I don't care as I doubt it'll ever matter to me personally.

 KP> seems possible. how are the encoding times going at these stages?

The same as before.  Offhand I think the 16:1 (encoding time:play time) is a
good guesstimate.  Thus for a two pass mencoder of my two minute movie (short?)
it will take half an hour plus change to replicate what would happen with a
real movie.

 MK>> What do you think?
 KP> for small, monophonic is usually not too bad, though many stereo 
 KP> recordings don't mix down to monophonic as well as they might -- i've 
 KP> only read a bit about some of the reasons, but it can be interesting 
 KP> stuff.  thanx  :)

We'll see.  I'll let you know as soon as I know.

Life is good,
Maurice

--- Msged/LNX 6.1.2
 * Origin: The Pointy Stick Society Headquarters - Ladysmith BC Ca (1:261/38.9)