[MPlayer-dev-eng] best place to analyse uncompressed frames

Brian J. Murrell 76fc8798a4b8f32826ea7828b87709dc at interlinx.bc.ca
Wed Apr 24 12:12:29 CEST 2002


On Wed, Apr 24, 2002 at 11:29:11AM +0200, Arpi wrote:
> 
> yes.
> but it won't speed up playback (timer) just skip displaying

I see.  OK.

> dunno... maybe vf_rotate

I will look at that one too.  I was looking at crop.  Cropping is a
fairly basic and easily understood operation on WxH type images.

> it will.
> seek does a-v resync after seeking

OK.

> but just skipping all video frames won't keep a-v sync...

But...

> so you need 'seek 0' after skipblack to get a-v resynced (problem with long
> keyframe intervals)

This is what I was thinking.  I thought seek must do the A/V resync,
so I will just piggyback, or duplicate what seek does.

> sorry if you aren't familiar enough with mplayer code

Well, not too familiar with the decoding and playing part yet.  I have
been through the main block and input sub-system briefly when I was
trying to figure the input file layout.  Also I got somewhat familiar
with the libvo when I was hacking on the hardware brightness/contrast
controls for the mga driver.

> and you aren't
> interested to read and understand it,

Oh?  I said that did I?  Or maybe you are reading my mind.  Geez
A'rpi, you really don't know how to foster an Open Source project do
you?  When you find somebody who is interested enough to want to start
hacking on it, you are supposed to give them guidance and support, not
throw the fact that they don't know the entire codebase back in their
face.

> give it up, now.

Well, I guess contrary to most other Open Source projects which are
always trying to bring new hackers on board, you want to always be the
hacker on MPlayer.  You act like you don't want to see a day when
MPlayer lives beyond your interest in hacking on it.  I would think
you would like to see more developers, whittling away the TODO list,
adding new features, chasing bugs, and generally seeing to it that
MPlayer grows beyond your interest in it.

I am quite a capable hacker once I get to know the codebase -- or even
just the portion I am working with.  I have hacked on quite a number
of Open Source packages to varying depths from simply hunting down and
fixing bugs to adding full features like the ad-hoc DDNS (the first
and only as it's time) method support to the ISC DHCP server complete
with parsing syntax to script the addition of RRs and make them
conditional on certain aspects of the DHCP server's state machine.
But Ted also knew how to culture Open Source hackers.

> it's not easy as i said... and you don't even RTFM tech docs

Well, to tell the truth, I did not know there were tech docs.  Thanks
for pointing them out.  But even upon reading the colorspaces.txt
document (thankfully they are .txt format), I don't really know any
more than I got off the web, like how the planar and packed spaces are
laid out into planes arrays.  But no worries, I will read some more
code and docs and figure it out for myself.

b.

-- 
Brian J. Murrell
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/mplayer-dev-eng/attachments/20020424/ee2f21c1/attachment.pgp>


More information about the MPlayer-dev-eng mailing list