DG DV tools HELP ME for new users

[TheProfosist]

i created this for all those new users out that that have just sarted using DG NV tools or people who have just switched over from a different set of tools. Because i am one of you and we all need to ask some stupid easy questions when were all starting out

DG NV tools ERROR report thread: http://forum.doom9.org/showthread.php?t=147945

[Guest]

OK, so what's your question?

[TheProfosist]

now for the first question to all u all knowing expert encoders out there

i just started a new dvd since im waiting on help on Lain. I just picked up Moribito today but i can figure out what to do with it because

MediaInfo says :
Frame rate : 29.970 fps
Standard : NTSC
Colorimetry : 4:2:0
Scan type : Progressive
Scan order : Top Field First

and i stick through DGindexNV and get a film percentage of 63.6%

what does this all mean because progressive film cant be interlaced and how do i intpret the 63.6%

log files from both mediainfo and DGindexNV
http://www.mediafire.com/file/wiidf0..._DGindexNV.txt
http://www.mediafire.com/file/yyygl2..._MediaInfo.log

my guess is that the video clip is roughly 63.6% 23.976 and that i should use fieldop=0 but how can something be progressive and TFF? my current idea is to use
DGDecode("source",fieldop=0)
TFM()
TDecimate(mode=1,hybrid=1)

[Guest]

and i stick through DGindexNV and get a film percentage of 63.6%

what does this all mean because progressive film cant be interlaced and how do i intpret the 63.6% Scan type progressive does not necessarily mean that the content itself is progressive. As has been noted MANY times both here and in the documentation, scan type progressive tells only how the stream was encoded and not what the nature of the content is.

63% of the movie is pure soft 3:2 pulldown, the rest isn't. What is so hard to interpret? The key question is what is the nature of the other 37%? Is it hard telecined 3:2? Is it interlaced video? CGI at 29.97 progressive? You have to find out to be able to process the clip correctly.

i should use fieldop=0 Correct.

but how can something be progressive and TFF? When you pull down progressive content, you have to specify the field order, because it could be done these two ways:

a a b c d #TFF
a b c c d

or

a b c c d #BFF
a a b c d

[TheProfosist]

i now know that if its hard telecined that i have to use external IVTC but i dont know if it is. whats the easiest was to find out what the rest of the material is?

also i though only with interlaced material did you have to specify the field order

[Guest]

i now know that if its hard telecined that i have to use external IVTC but i dont know if it is. whats the easiest was to find out what the rest of the material is? Inspect it by stepping through the fields as described at my web site (see #1 and 2):

http://neuron2.net/faq.html

But you have to know what these video types are and what they look like when you step through. So if you don't already know that stuff you'll have to step back and learn more about video.

Also note that for anime, the duplicate frames can complicate the identification of the field patterns, especially for 3:2 pulldown. That is where your experience becomes important.

also i though only with interlaced material did you have to specify the field order I just explicitly told you it's not true and gave you an example! Apparently you don't know what 3:2 pulldown actually is.

[TheProfosist]

3:2 pulldown is a process used to change 24p material into 30i for broadcast/distribution needed for pre hd era/ old crt tvs. for every 4 frames their is a fifth frame. i know theres different orders that you can do this. and sometimes the orders change within one video.

so with progressive theres no field order a a b b c c d d
but 3:2 pulldown progressive does before you reverse the pulldown ex a a a b b c c c d d
which i already knew

in your faq it also says "Additionally, if the clip is progressive, motion will be correct for both AssumeTFF() and AssumeBFF(), i.e., field order does not matter." but inorder for it to be that way it has to be pure progressive correct?

and from all that im have learned about 3:2 pulldown the only time you would use a deinterlacer is if the video is interlaced but the original content was progressive and of the same frame rate. which seems to be rare

i must say that i know im still a "noob" at this but in only a few monthes time i have improved alot atleast i can tell with my encodes. i was originally using a dvd ripping program to rip strait to original MPEG2 then i would drop it into divx converter it would auto crop and auto anamorphic resize to square pixels. no looking back i see how horrid mthose encodes are comparatively

[Guest]

for every 4 frames their is a fifth frame. That's wrong. For every 4 *fields* there is an extra *field*. Your misunderstanding here is what makes you think even pulled-down progressive doesn't have a field order.

in your faq it also says "Additionally, if the clip is progressive, motion will be correct for both AssumeTFF() and AssumeBFF(), i.e., field order does not matter." but inorder for it to be that way it has to be pure progressive correct? If by pure, you mean without pulldown, then yes. As soon as you pull down (duplicate) a *field* you have to play the result with one correct field order. That's because you will have frames with fields from different temporal moments, just like interlaced video. When you have a frame with different temporal moments, you have to play it in the one correct temporal order.

the only time you would use a deinterlacer is if the video is interlaced but the original content was progressive and of the same frame rate. which seems to be rare Nonsensical. You use a traditional deinterlacer when you have video with temporal moments at the field rate (aka "pure interlaced video") and you want to make a progressive stream (video with temporal moments at the frame rate) from it.

[TheProfosist]

thats was actually what i was trying to say just dont know correct terms but still for what im doing it doesnt seem like ill be using a deinterlacer much

[Guest]

Don't worry. When you need one you'll know it.