http://www.hometheaterhifi.com/volume_....00.html

That is the best explanation you can find any ware about this subject.

Another very good discussion right on topic :
http://www.avsforum.com/avs-vb....r+Video

For correctly flagged film based source the HTPC should win... If you then add scaling to the mix (which you are not) the PC should not only win but do so by a considerable margin.

For incorrectly flagged or video sourced material using a SWDVD player the HTPC will lose.


May be this is what happening with our Indian DVD’s barring DEI.

Finally I have done some testing with some of the DVD’s I have. While RP56 doing better with the Video based movies from EROS, VIDEOSOUND and YRF I certainly won’t say it make them progressive. Still the image is soft and does not show minor details, as is not the case with a progressively mastered DVD. I can say the image looks better than my HTPC image by about 15 to 20%.
It must be noted that Faroudja chip based DVD players do better with Video based material than any software DVD players. Probably that is what Michael was experiencing with NTJNH DVD.

I still don’t have the Na Tum Jaano Na Hum Review DVD Michael was referencing, so I can’t comment on that one.



Edited By ganti on Aug. 12 2002 at 14:07

Thanks ganti for info links.

Michel Hefner's NTJNH review is here.
http://www.zulm.net/modules....sid=474


Your statement,
"For correctly flagged film based source the HTPC should win... If you then add scaling to the mix (why not; this is the best way to go; the results are amazing) the PC should not only win but do so by a considerable margin"
sums up the importance of correct flags set in DVDs.

This info also implies that there is another category of DVDs which have progressive frames but the flags haven't been set. If a decoder can identify the original film frames, the video fields can be weaved back equally effectively. It is this category of DVDs that I want to identify which in my opinion are next best to the properly flagged DVDs.

Rana



Edited By rana on Aug. 12 2002 at 16:47

Rana,
From What I understand from the above links, if there are 30 frames drawn from 24 frames irrespective of the flag the decoder try to use 3:2 to draw its original frames if it film based.
So only I can identify the problems with Indian dvd’s with the only and definitive problem like the below:
when the deinterlacer blithely combines together two fields that weren't meant to go together, usually because the 3-2 sequence is interrupted and the deinterlacer doesn't adapt quickly enough.
When this happens, the odd numbered lines of the image are from one moment in time, and the even numbered lines are from a different moment.
This is what I see if I try to use Force Weave on this Indian DVDs’. An artifact called combing. Then I can surely say they are interlaced or wrangly drawn from two different fields.

The article mentioned is very good. Based on it I would
say that the DVD of NTJNH provides a pulldown which the
Faroudja can reinterleave (and not deinterlace with loss
of sharpness) since it does not know about flags, while
the software player PowerDVD on my laptop applies vertical
filtering to 2 frames out of 5, creating the frames that
look like combinations of two film frames.
cheers
Michel Hafner

Still in the absence of NTJNH, I have done few observations with 2 EROS/B4U dvd’s Pardesh and Gupt.
1) My HTPC blows away any DVD player for Film sourced DVD’s or properly flagged DVD’s, period.
2) I can say Rp56 do a reasonable job when dealing with the above class.
3) Now comparing an average DEI dvd with that of an average EROS dvds on my HTPC is like comparing apple’s to oranges. I can say the gap is very wide, some thing like 50%
4) With Rp56 the gap is not that wide as it works better with Improperly flagged or not flagged dvd’s

Thanks to Rana,
For being persistent and open minded. It paid off I guess. I am now enjoying some crap dvd’s which I never thought, could able to enjoy them on big screen.
With out an extra penny I am enjoying the majority of Indian dvd’s with my RP56. The PQ increase I am experiencing varies from 5% to 30%. Your mileage may vary and again it depends on your display. As in close up shots the PQ is as good as DEI but it does not show that much sharpness in long shots.
I am not saying that there are as good as DEI. Not in a long shot. But still I can enjoy watching it on my big screen with a relatively cheap DVD player.
So far my testing is limited to few dvd’s. But I will extend my testing to wide range of dvd’s. I can safely recommend this dvd player or as a matter of fact any Faroudja chip based DVD player for all the Indian dvd fans out there. By mere $200 you are in to a long treat.

I tend to agree with you that by weaving the frames together in properly flagged and made DVDs, we achieve about 50% improvement in PQ instead of 100%. Not 100% improvement as all the info is still there in the interlaced signal as well. The difference is in the presence or absence of interlacing flicker and artifacts.

I also concur with your other assessment of somewhere 5% to 30% improvement for improperly flagged DVDs. These DVDs already have video noise embedded in them.

Compare this to only 33% improvement achieved from proper anamorphic transfer over letterboxed transfer. It is also to note that anamorphic DVDs from letterboxed master don’t provide any improvement whatsoever.

Important conclusion is that PROPER FLAGS AND TRANSFER FROM FILM is more important than “ANAMORPHIC”.



Rana

P.S.
Using simple math tells the following:
Assume a score of 1.0 for interlaced and letterboxed (1.78:1) DVDs (90% of Indian DVDs will score 0.1 to 1.0; majority will score 0.5)

Then a properly flagged and anamorphic (1.78:1) Film DVD will score 2.0

Rana@ Aug. 15 2002,09:59



I agree with you 100%.

ganti wrote:

Rana@ Aug. 15 2002,09:59

Quote:

Assume a score of 1.0 for interlaced and letterboxed (1.78:1) DVDs (90% of Indian DVDs will score 0.1 to 1.0; majority will score 0.5)

Then a properly flagged and anamorphic (1.78:1) Film DVD will score 2.0

I agree with you 100%.

Thanks ganti. You agreed so far. Using same logic/ math, let me include other categories as well:

Inerlaced and anamorphic widescreen from anamorphic master = 1.33
Interlaced anamorphic widescreen and from non-anamorphic master = 1.0
Progressive and letterboxed widescreen = 1.5
Progressive and anamorphic widescreen from non-anamorphic master = 1.5
Progressive and anamorphic from anamorphic master = 2.0
Progressive and Full Screen (4:3) from 4:3 master = 2.0 or 1.5, depends ---.

Rana



Edited By rana on 1029773888

The first link in the first post was updated in August 2002. If anybody intrested.

http://www.hometheaterhifi.com/volume_....00.html

Rana



Edited By rana on 1030393488