[MPlayer-users] RFC: docs update for "how to create a high quality DVD rip"

[Wayde Milas]

On Sun, 2004-06-06 at 23:20, Jason Tackaberry wrote:
Ive been following this thread with ALOT of interest because this is
exactly what I use mplayer/mencoder for. I have a large (multi terabyte)
server that feeds video anywhere I want in my home.

I encode dvd's, but more importantly, the kids elmo, seasame street and
other dvd's. The kids get about 30-45 mins of tv time a day (3 years
old) and I dont want them watching garbage, so I preview the shows. This
is a great way to have an index of shows for them to watch.

The idea is to have near the same quality as the original, (at least to
the naked eye) but compress the stream as small as possible. Ie, high
qulity/high bit encoding with the purpose to still have a smaller than
mpeg2 stream.
> 
> > > money into my home theatre setup, and when I hear crap in my rear
> > > speakers because my receiver is trying to find a channel in the high
> > > frequencies that has been compressed to hell, I go batty.  It's really
> > > annoying. :)
> > 
> > Then disable that "feature"...
> 
> Well sure, I can take my receiver out of Dolby Prologic mode and put it
> into stereo, but then I might as well just put my headphones on and
> listen that way.
> 
> Watching a movie with no rear channel(s) (or a rear channel with audible
> artifacts) just totally ruins the experience for me.
> 
> Now, luckily, you can encode your audio as vorbis, and I can copy my
> audio stream raw, and we can both be happy. :)  I suppose the definition
> of "high quality" is going to depend on your watching environment.

I have to agree here. Watching a feature film on a 5.2 surround system
with high quality gear (Klipsch and Adcom) in Ac3 is NOTHING like
watching a vorbis stereo version.

Its like a 400 stream compared to a 2600 stream. Does it look similar?
yes. Does post processing help? yes. Is the quality difference
noticeable? hell yes.

What we REALLY need is vorbis to compress greated than stereo files.
From what I understand its theoreticly possible to do this, but no tools
exist to do it/play them correctly. Then we could have our cake and eat
it too.

> > Also: don't denoise.
> 
> Well, I do say denoise.  I just denoise very little.  But I updated that
> section (and a few others, based on your comments) to encourage
> experimenting with higher values.  Based on my experience, I'm not
> comfortable recommending the default values for hqdn3d.

This is a new one for me. Well, not new, but the whole noise thing
scares me. From what I understand, noise will take out some of teh
random noisy garbage thats not supposed to be there on flat surfaces,
ect. However, It also removes detail on stuff that IS supposed to be
there.

How safe is this to actually use on high quality archival streams?

Also, when you go to play the file, is it a simple means to add the
noise back in? If so, is teh noise added back in exactly the same as it
was taken out, thus resulting in the same stream, but causing more cpu
processing time, or is the stream "almost" like the original.

Let me qualify that. I know this is lossy. By original I mean an encoded
stream that has no denoise filter applied when encoded.
> 
> > Never scale up. Always scale down or not at all. Personally my pick
> > depends on whether the original was a movie or made-for-tv. I always
> > try to avoid vertical scaling when the original was made for tv (since
> > the line sampling corresponds to the way the content was originally
> > recorded) but that's pretty irrelevant when the original is film.
> 
> Thanks, that's good advice.

The question I think is at what point (maybe a lose rule of thumb.. of
course you need to play with individual movies to see the exact place)
does it make sense to scale the stream down to create a stream that
looks better X bitrate.

I guess what I'm asking here is that if I have stream at X by Y at lets
say 960. Would it look better at X by Y or X/2 by Y/2? I know this is a
bad generalization, but I'm looking for a loose set of parameters to
know at what rate a good, modern, progressive 24fps widescreen dvd
encoded movie should be encoded at at full resulution, and at what
bitrate I should start thinking of scaling the resolution down.

> 
> > Yes, or conversely using insane bitrates lets you get by with being
> > sloppy. And I don't like sloppiness.
> 
> I don't like sloppiness either, and I don't want to offer any advice in
> the guide that encourages it.

Either do I. thats why I'm trying to understand this all.

Something else for high quality encoding. the cmp functions (default is
1 I think) plain suck. Well, they arent cpu intensive but with some more
processsing time you yield dramaticly higher PSNR values on clean signal
originals.

these are: precmp=3:cmp=3:subcmp=3

3 is a bit overkill but can be slightly better than 2 in my opinion. 2
is a diffinate must.

On animated/cartoon stuff 7 seems to work well also. 2 nad 3 yield
higher bsnr values (by ALOT) on jsut about everything ive tried them on.
B&W dont get a smaller psnr boost.

Be prepared for longer encoding times.

Hope this helps some

Wayde
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.mplayerhq.hu/pipermail/mplayer-users/attachments/20040607/21cb9066/attachment.pgp>