[opendtv] Re: 625 video quality is good enough....

Craig Birkmaier wrote:
 > It's time to get rid of interlace. PERIOD!
 >
 > There is NO GOOD reason for this archaic compression technique to be
 > concatenated with digital  compression.

Maybe.  But it always seems the major objection to interlace is that it 
makes everything just too darn complicated.  And yet that claim is also 
often directed at computers in general, though most people find a use 
for computers these days.

I know there is also a lot of theoretical support for the idea of an all 
progressive system.   But I still cannot convince myself that the best 
sampling grid places the samples at exactly the same points at each 
moment of time.  There seems to me to be an informational advantage in 
sampling slightly different points on each frame and letting either eye 
motion tracking or motion compensated interpolation gather a bit of 
extra detail.  Though if I was given a choice I might not just pick 
alternating even and odd lines.

But I do believe that some of the problem is that neither the codecs nor 
most fixed pixel displays deal with interlace very well.  Those problems 
may be the limitations of current software and not necessarily a show 
stopper.

- Tom




> At 11:43 AM -0700 10/17/04, Dale Kelly wrote:
> 
>>You might recall a recent posting suggesting that HDTV might be of 
>>little value in countries using the 625 line system since that 
>>system's quality was likely good enough. My position was, that as 
>>good as that image might be relative to other analog based 
>>standards, it could not compete in quality with true HDTV video 
>>displayed on a 720 or 1080 large screen display.
>>I'm revisiting this subject only because I saw a very graphic 
>>illustration supporting my argument last evening when viewing the 
>>BBC production of "Last Night at the Proms", on the Discovery HD 
>>channel. This is an excellent program in every way but it was 
>>produced in the 625 (580? DTV) 16X9 format* and compared to other 
>>HDTV programming on the same network was noticeably softer, 
>>particularly on the medium and long shots which are such an integral 
>>part of that program. Clearly the viewers in Britain would have 
>>noticed and wanted the difference.
>>
> 
> 
> There is an obvious explanation, one that has significant relevance, 
> as the ITU considers whether it should approve 720P for international 
> program exchange, and the inclusion of 720@50P into the international 
> standards.
> 
> I cannot reproduce the submission to the ITU directly, however I can 
> paraphrase the key points that the advocates of 720P have made:
> 
> 
> 
> - Progressive formats make compression work better (one company 
> estimates  a 10%to 30% increase in the required bit rate for MPEG 2 
> compressed interlace video as opposed to progressive scan video), 
> preserving bandwidth and providing the best quality to viewers.
> 
> - Emerging display technologies are progressive friendly and are 
> dominated by 1Mpixel types.
> 
> - 1280x720 is friendly to modern post-production techniques, which 
> often need to de-interlace source for processing, such as spatial 
> scaling and rotational manipulations.
> 
> - Conversion from 720p to any other format is simpler and provides 
> better quality because there is no source de-interlacing involved. 
> This is the key to why Dale saw a "soft" picture. In order to present 
> the content here in the U.S. the original 1080@50i source (thanks to 
> Alan Roberts for this very useful clarification) had to be standards 
> converted from 50i to 60i for broadcast by Discovery Networks. This 
> requires a de-interlacing step, then frame rate conversion, then 
> re-interlacing.
> 
> Welcome to the realities of standards conversion. It does not get 
> better with HDTV, if we are trying to do frame rate conversions on 
> interlaced source. We are trying to create information that was not 
> sampled, using samples that have been compromised by interlaced 
> acquisition. The net result is that to cover all of the artifacts of 
> the standards conversion, we give up significant resolution.
> 
> Contrast this with a 50P to 60P conversion or visa versa. We do not 
> need to de-interlace the source, and we have excellent spatial detail 
> available to do the frame rate conversions. The results are obvious 
> on a progressive display.
> 
> So bottom line, Dale was seeing the "kinder, gentler, softer" side of 1080i.
> 
> It's time to get rid of interlace. PERIOD!
> 
> There is NO GOOD reason for this archaic compression technique to be 
> concatenated with digital  compression. This is equally true for 
> SDTV(525 or 625 line), as better results can be obtained with a high 
> quality de-interlacing system, before the source is subjected to 
> MPEG-2 compression. Expecting a cheap de-interlacing chip in a 
> consumer display to do as good a job as a $75K to 100K deinterlacing 
> system is ludicrous. On the other hand, it is dirt simple to convert 
> progressive source for interlaced display using noting more than a 
> convolution filter to remove the details that would cause offensive 
> artifacts on an interlaced display.
> 
> If we only put progressive source into the DTV channel, the use of 
> interlaced acquisition would
> disappear quickly, in favor of progressive HD and EDTV acquisition.
> 
> Regards
> Craig
>  
>  
> ----------------------------------------------------------------------
> You can UNSUBSCRIBE from the OpenDTV list in two ways:
> 
> - Using the UNSUBSCRIBE command in your user configuration settings at 
> FreeLists.org 
> 
> - By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word 
> unsubscribe in the subject line.
> 
> 
 
 
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:

- Using the UNSUBSCRIBE command in your user configuration settings at 
FreeLists.org 

- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word 
unsubscribe in the subject line.

Other related posts: