[MPlayer-users] Which codec is fast and good?

[D Richard Felker III]

On Thu, Aug 28, 2003 at 07:07:12PM +0200, Matthias Wieser wrote:
> [Automatic answer: RTFM (read DOCS, FAQ), also read DOCS/bugreports.html]
> Am Donnerstag, 28. August 2003 08:12 schrieb Rainer Hantsch:
> 
> > Yesterday I gave pp=tn a chance - because of the hint of Matthias
> > Wieser.
> >
> > o  The first thing I recognized is that it causes much less CPU load.
> > My Celeron-1700 was running at 95-100% with 'denoise3d' but it was
> > somewhere around 50-70% (mostly 60%) with pp=tn.
> >    A good optimization.
> 
> Here too.
> 
> >    Recording 2:05:00 of TV in 640x400 (I recorded StarGate for
> > testing), resulted in 4.8 GB File size !!! This would have been
> > somewhere at 3.5GB with 'denoise3d', I estimate, based on earlier tests
> > with 1 minute recordings.
> 
> Without pp=tn the size should be even larger.
> My experience is, that tn is better than nothing but minimal worse than 
> denoise3d. But tn doesn't need much CPU time.
> 
> Does anybody know, what's the (visible) difference between denoise3d and 
> hq_denoise3d?

pp=tn can and does leave (actually it creates) serious artifacts due
to fundamental design flaw.

denoise3d leaves small artifacts due to low precision computation and
rounding errors.

hqdn3d is very slow due to higher precision arithmetic.

Rich