Question about cell structures

[rcubed]

Hi,
I've been messing with re encoding back ups of DVDs with DVD Rebuilder, and an issue concerning whether or not a DVD cell can contain only Nav Pack Data. See the thread "Error? Message when rebuilding" under the DVD Rebuilder discussion group for the details of what has been going on. :confused: - sort of. The info I've gotten there has been most helpful. My biggest problem is how to identify them easily.

It would be interesting to see what the "experts" have to say? I put myself in your hands ;)

Thanks for any advice etc.

VobBlanker is a life saver and most useful. My favorite way of getting rid of unnecessary cell, etc. :D

rcubed

I can supply additional data if needed (logs ifos etc).

[blutach]

While we can not allow IFOs to be posted, this thread (http://forum.doom9.org/showthread.php?t=132108) is pertinent.

Regards

[linx05]

That is the thread he started blutach. Also he mentioned it in the post. ;)

Too many beers and BBQs? :P

[Sir Didymus]

...whether or not a DVD cell can contain only Nav Pack Data...


I would say no (and I would leave the place for the real experts to comment on the matter)...

I have personally no definitive arguments on the matter:

From one side, the definition itself of a cell, quoting DVD-Video Demystified, which is "... a unit of video with a duration that is anywhere from a fraction of a second to several hours long..."

However a cell is a sequence of VOBUs, and I am quite sure that, inside a cell, VOBUs made of single navpacks are compliant...

Also, I am sure that every VOB should contain video, and the first VOBU of a VOB should contain video too (I can quote some related errors reported by the Philips Verifier, to confirm this, if needed)...

In the scenario reported by the referring thread (fake cells replaced by rippers), I have to say I fully agree with blutach: since very frequently the cells we are talking about are at the beginning of the VOB, it is always better putting some blank video in it...

Cheers,
SD

@rcubed - In any case, why to post the question in this forum ? it seems there are no reasons to discuss the matter both here and in the original thread, in the DVD-RB forum...

[blutach]

@SD

Given the direction things were headed in the original thread in the DVDRB forum, I actually think it best to have this discussion here. However, I am a bit loathe to transfer the original thread outside DVDRB forum as the original question had to do with a problem with Rebuilder and jdobbs may wish to opine on it there. Otherwise, subject to mpucoder and bigotti5 chiming in, I am with you on this.

@linx05 - I was simply providing a link and reminding folks not to post IFO files as per forum rules.

Regards

[rcubed]

I would say no (and I would leave the place for the real experts to comment on the matter)...

I have personally no definitive arguments on the matter:

From one side, the definition itself of a cell, quoting DVD-Video Demystified, which is "... a unit of video with a duration that is anywhere from a fraction of a second to several hours long..."

However a cell is a sequence of VOBUs, and I am quite sure that, inside a cell, VOBUs made of single navpacks are compliant...

Also, I am sure that every VOB should contain video, and the first VOBU of a VOB should contain video too (I can quote some related errors reported by the Philips Verifier, to confirm this, if needed)...

In the scenario reported by the referring thread (fake cells replaced by rippers), I have to say I fully agree with blutach: since very frequently the cells we are talking about are at the beginning of the VOB, it is always better putting some blank video in it...

Cheers,
SD

@rcubed - In any case, why to post the question in this forum ? it seems there are no reasons to discuss the matter both here and in the original thread, in the DVD-RB forum...

Sir Didymus,
At the risk of kind of repeating what Blutach had to say, but to get it from me directly, in light of the fact that it was a problem that will probably have to be handled in VobBlanker or PGCEdit, or other editing program I thought I would consult the "experts" directly in this forum and for the most part stop adding to the post in the other thread.

I had hoped Jdobbs would comment since it at 1st involved DVD-RB, but now I see it is a structure problem so to carry on the discussion here seems to be more appropriate. If the expert consensus is that it is a legal structure, then maybe Jdobbs might want to consider seeing if there is something that can be done in DVD-RB. I could have posted some of the info from that thread here, but since it was all ready documented there I just decided to reference it here. Besides I'm a really bad typer (thank goodness for spell check). :o

I had not intended to post any IFOs but ment I could email them to someone if that is permitted?

rcubed

[Sir Didymus]

Hi. OK. The reasons for this "fork" in the IFO/BOB Editors and DVD-RB forums are clear now...

And there is no need to post or to send to anybody any piece of IFOs in addition: IMHO your question is very clear:
"Can a cell of a DVD-VIDEO contain just Nav Pack Data ?"...

Let's see the opinions of other contributors; my point of view is "no: at least some presentation data (video or audio) should be present"...

AFAIK no authoring applications allow to build up cells without some actual presentation content... and some ones use directly the term "scene" instead of "cell"...

[mpucoder]

I'd say no also, the video establishes the duration of the cell. Without video the duration is zero making it impossible to construct the following VOBUs with legal timing values. And no authoring app I know of will allow you to make such a cell.

[rcubed]

I'd say no also, the video establishes the duration of the cell. Without video the duration is zero making it impossible to construct the following VOBUs with legal timing values. And no authoring app I know of will allow you to make such a cell.

Thanks for taking the time to respond! Your opinion makes sense about the time being zero. I guess it is unreasonable to expect DVD-RB to be able to encode those cells.

rcubed

[blutach]

Thanks for the info and the usual precision, mpucoder.

A bit strange that the verifier does not complain. Maybe no-one at Philips ever thought anyone would make a cell without video :D

Regards

[Sir Didymus]

Hi blutach!
Who say that the Philips Verifier does not complain ?

Apparently, by a quick reading of the manual, no direct errors exist on the subject...

BUT... Maybe there are tons of indirect errors, reporting some unpleasant consequences of the issue...

[blutach]

Fengtao said it here (http://club.cdfreaks.com/f116/dvdfab-platinum-ripped-files-incompatible-other-programs-232255/#post1942148) (he is the author of this scheme).

Regards

[Sir Didymus]

I see. Thanks for the link to the discussion at CD freaks...

As the minimum, I simply can not see how the following errors /oddities can be missed by the Philips Verifier...

>>> [DVD] ERROR 3015 (ref. DVD-3 Table 5.1-1):
ERR_DVD_1ST_VOBU_VID
The first VOBU of a VOB should have the video data

>>> [DVD] ODDITY 3020:
ERR_DVD_VOBU_EMPTY
The previous VOBU does not contain any data!


Edit - OK. In the next weeks (Christmas holidays ?) I will do some testing on my own and report...

[blutach]

The first VOBU of a VOB should have the video data
I saw that, too but the word "the" confused me a bit. Because, nowhere did it actually say you had to have video data. Given these recent threads, I read "the" video data as "video data, if there was any".

And, as for the oddity, who takes notice of them? :D

Regards

[Sir Didymus]

Hey, blutach, you are taking me on a level of linguistic detail which is very slippery, considering English is not my mothertongue... :scared:

Indeed, it seems to me your interpretation about the "the" is very elastic...

Let's try another one... Do you agree this other reference (page 139 of the manual) is a plain statement about video data should actually be present in the first VOBu of a VOB ? :sly:

http://img248.imageshack.us/img248/8510/image1yz2.th.gif (http://img248.imageshack.us/my.php?image=image1yz2.gif)

[blutach]

SD,

I admit I was very much trying to "see the other side's point of view" and give Fengtao the benefit of the doubt and was perhaps trying to read too much into it.

However, your pic makes it totally unambiguous (we would never have doubted mpucoder anyway) :)

Regards

[rcubed]

To all,
Just for your information. Fengtao just released version 4.0.2.0 Beta of DVDFab and from what I can tell looking at the VTSs in VobBlanker inserted 8192 bytes of data into the cells that were causing DVD-RB to have "heartburn". I was able to rip and re encode the problem DVD with DVD-RB w/o problems. :)

rcubed

[Surf]

rcubed,

Even without the new version, the issue can be solved with PGCedit. Requires a lot of work though!! Involving blanking pain-in-the-butts cells at the start of each PGC and removal of cell commands. Then finally run the whole movie with VOBblanker to remove them 8192byte cells.

I too have this movie.

My approach is extremely simple: using DVDshrink chop-chop method. The audience are the ones with greasy popcorns, candies, chocolates...:p Personally I find backing up a cartoon with DVD-RB is an overkill. What can DVDshrink possibly degrade? Less vivid colors?