Custom Template for CCE for DVD encodes

[syzygytec]

@ddogg
Nice of you to post your recommended CCE teplates for svcd. Could you please recommend to me your settings for both Progressive and Interlaced templates for both 2.67 and 2.50 for DVD encode. Your advice would be very much appreciated by many I'm sure, I guessed at them for 2.67 and now was sitting here try to for 2.50 but it is so much different, in light of the recent comparisons between 2.50 and 2.67 image quality I'd like to do a few encodes with 2.50.
Thanks

[DDogg]

It is the same as rb's 2.67 template except some things are moved around.

CCE 2.5
Progressive -
Quality setting - approx 17 IQP. No anti noise filter
Video - Check Add sequence end code, Progressive frames, Zigzag Scan

Interlace -
Same as above except no progressive and no zigzag checked

[r6d2]

@DDogg, never done DVD output execpt for testing, but are those DVD compliant stuff important?

[syzygytec]

*Edit
thanks for your help ddogg, I think I have decided to stick with 2.67, to me it looks better than 2.50 and I'm not seeing any of the problem as in the comparison topic.

[DDogg]

r6d2, you know more about that stuff than I ever will, plus you actually read the manual. I have not thought about it much, but I would assume the way it forces DVD resolution would foul up FACAR? Also, I have been using an 8500 Max. If memory serves it forces that also. Let me know :) I just don't feel like reading anything technical tonight. I've been doing it all day.

Speaking of PITA, try figuring out the proper script to recode a live capture to DVD from the capture FAQ. Bring three aspirins and a bottle of rum.

syzygytec, IIRC, close all gops is used when you wish to edit or when you are creating branching segments. See the CCE FAQ for timecode and upper field first and some of the advanced authoring references for the close all GOPS. In SVCD we don't use any of 'em. UFF coresponds to putting a 1 in offset line in 2.67. The only time you do that is when the source is interlaced *AND* BFF. It is damn confusing and does not do what it sounds like. RB covered it very well in the CCE faq. He probably needed a bottle of rum after that also :)

[jsoto]

What about IQP=5? I've read some threads recommending this for DVD output (with medium-high bitrate). Also, this is the value stated in Doom9's Guide (http://www.doom9.org/mpg/cce-advanced.htm)

I've done a simple test and I've not been able to find a big difference to my eyes (in my PC, not in a big TV). May be I'm not a quality guy :) or (I prefer this one) I've to evaluate more test clip samples.

jsoto

Edit: sorry, grammar corrections

[bobwillis]

Hi, this is my opinion. I found in very dark scenes that IPQ 17 is preferable due to the fact that it seems to give better (less quantized) shades of darkness across 'flat' areas. With an IQP of 5, over a large area, everything tended to be the same shade. Chapter 9 of Private Ryan, 1 hr and 11 mins is a good example of this; look at Cpt. Miller's face.

Since this test, I've always used IQP of 17. According to the manual, this should give increased 'mosquito' noise, but I haven't noticed it.

Regards,
Bob