[mythtv-users] Finally... OBJECTIVE PVR250 Quality measurment

papenfuss at juneau.me.vt.edu papenfuss at juneau.me.vt.edu
Wed Mar 17 15:12:09 EST 2004

On Wed, 17 Mar 2004 steve at nexusuk.org wrote:

> Hash: SHA1
> On Wed, 17 Mar 2004 papenfuss at juneau.me.vt.edu wrote:
> > 	OK folks... I have finally cooked up an *easy* and *objective* way to 
> > verify the quality of video captures.  It's based on the EIA1956 standard B&W 
> > TV test pattern that can be found snooping the 'net.  I made an image that can 
> > be burned to a DVD and played in a set-top player.  The resolution is 720x480, 
> > which is (as good as or) better than anything the PVR-250 alegedly can capture.  
> > If there's some interest, I'll find a place to post it.  The DVD image zipped 
> > up is only 500KB.
> Except the quality of digital video compression systems is mainly only an 
> issue with moving images...
	Yes and no.  If you can't capture a still frame at high resolution, 
though, you certainly can't capture motion at high resolution.  The 
bitrate/resolution thing is an entirely separate issue, and one that's much 
harder to quantify.  The image quality problem I'm seeing looks like simple 
analog lowpassing to me.  Analog video bandwidth, pure and simple. 

	... Although you did raise an interesting point.  Capturing from the 
/dev/yuv (the raw YUV capture buffer, right) would rule out any encoding.  
Might try that... or anyone else feel free.


>  - Steve                                             http://www.nexusuk.org/
>      Servatis a periculum, servatis a maleficum - Whisper, Evanescence
> Version: GnuPG v1.2.3 (GNU/Linux)
> Comment: Public key available at http://www.nexusuk.org/pubkey.txt
> iD8DBQFAWKY05zUOsIV3bqERArnJAJ4y5+swmJ2cm8fbOHyaC7TWzKkhygCcCpEG
> Sf65MTeG2KHHmpi4/zVoops=
> =Nlp7

* The prime directive of Linux:  					*
* 	- learn what you don't know, 					*
* 	- teach what you do.						*
* 						(Just my 20 USm$)	*

More information about the mythtv-users mailing list