DVD2AVIdg/MPEG2DEC3dg 1.1.0 requirements

[neuron2]

Now is the time to specify the feature list for version 1.1.0.

Here is the tentative list, presented for your review:

1. Correct the version string reported in the About box.

2. CLI support for selection of 1.76 vs. dg style project files.

3. Support for transport streams and HD.

4. Improved random navigation performance.

5. Automatic D2V file generation for MPEG2DEC3dg.

6. Generate D2F files.

7. Record aspect ratio information in the D2V file.

8. Fix case presentation for CLI file names (don't map to all
upper case).

Please note that this proposed release continues the plan to merge diverging versions. By supporting transport and HD streams, we continue the merger. Functional improvements can come when the initial goals are reached: correct MPEG decoding and merger of existing variants of the tools into a single variant.

[Tuning]

This may be difficult,but can DVD2AVIdg automatically select the field operations force film or swap field order all by itself,according to the values in the information window.If this is a useless question ...please ...neglect this post.


-Tuning

[len0x]

@Neuron2

About number 7 - does this mean that structure of d2v file has to change? It can break lots of applications. May be if this info can be put say after FINISHED (like % of film is put there now) it would do ?

@Tuning

About FF: this feature would be usefull only when having 100% FILM. And I've heard that even in this case it can be wrong. So in no way this should be made compulsory. But having it as an option (may be even with percentage above which FF should be activated) can be nice. It just a matter of changing two lines of output d2v...
I don't think swap field order can be autodetected, or can it be ?

[neuron2]

Originally posted by len0x
About number 7 - does this mean that structure of d2v file has to change? It can break lots of applications. I contemplate revising the D2V file in several non-compatible ways to support some new features. E.g., the picture type will be added to the D2V file flags to allow for more efficient random navigation. The option to save/read the old compatible format will remain.

[len0x]

Originally posted by neuron2
I contemplate revising the D2V file in several non-compatible ways to support some new features.

Good to know. Atm I'm actually doing some parsing of d2v files myself. So can you please give me a shout when format's changed, so that I can be prepared (if I'm not following this forum, I might forget about it)? :)

[abman]

With HDTV/ATSC, will it support the automatic PID selection found in DVD2AVIt3? The autodetection was nice but could use a little work still, as about half the time it gets it wrong or uses an old incorrect value. Also CLI PID selection would be nice.

[neuron2]

Originally posted by abman
With HDTV/ATSC, will it support the automatic PID selection found in DVD2AVIt3? The autodetection was nice but could use a little work still, as about half the time it gets it wrong or uses an old incorrect value. Also CLI PID selection would be nice. When we have agreement on the gross features, I'll spec out each one in more detail as work begins on it. But yes, I plan to support the auto PID detection.

[Cyberia]

Would adding MPEG-1 decoding be a new feature or too big of a project at this point in time? (Given that you are adding HDTV streams)

However, there are already a lot of things on the list, maybe it's best to wait.

[neuron2]

Originally posted by Cyberia
Would adding MPEG-1 decoding be a new feature or too big of a project at this point in time? That's 2-3 versions away.

[neuron2]

Originally posted by Tuning
This may be difficult,but can DVD2AVIdg automatically select the field operations force film or swap field order all by itself,according to the values in the information window. If this is a useless question ...please ...neglect this post. I agree that it would be possible to run an analysis pass over the clip and then use the data to produce reasonable (good enough) specifications for all the required settings. Usually humans can produce better quality settings than an algorithm can. But ideally, an automatic mode would be used by default and experts would override that for full control.

Some of our esteemed colleagues here have been working on solutions for automatic feature setting. I assume we will see great things from these people.

[Tuning]

I agree that humans are far more superior than these algorithms.But if you think of the newbies the automatic mode can be helpful.Even though there exists probability of wrong calculation on video source.

Thanks Donald for the information.

Bye

-Tuning

[jfcarbel]

Neuron- from his post in another thread:
"VOBs have special data in the stream, such as "navigation packs", that DVD players use. But not all MPEG2 files are VOBs, so pure MPEG2 decoders don't have VOB navigation. It would be interesting to consider enhancing DVD2AVI, et al, to use VOB navigation when it is present in the stream"

DVDShrink - "dvd IFO files contain an index table for each vobu (=GOP) in the VOBS: the current preview engine uses this table for seeking, I don't build a separate index file"

It would be nice to see this option so we could quickly generate an index file since d2v does take a significant time to build. It would be useful when using DVD2AVI as a VOB and audio cutter rather than a frameserver.

I also vote for adding audio to DVD2AVI as others above requested.

[jonny]

@neuron2:

Hi, this is a little OT.

I have a question regarding mpeg2dec.

Do you think it would be possible to add a function, really similar to the actual SelectRangeEvery, but that extract slices in a GOP aligned way (for example driven by SelectRangeEveryGOP(%_of_the_clip_to_extract, range) instead of the actual SelectRangeEvery(period, range))?

I'm asking this mainly because this could improve the speed of the compressibility test encodes.

This isn't a feature request, since i'm not sure i need something like this, but i'd like to know what you think about this (if it would be easy/hard to implement).

Regards

PS: IIRC the very first mpeg2dec version had an undocumented function similar to this one.

[stax76]

this "automatically" option sounds great :)

[SILICON]

Please, add this feature to the list:

- Command line parameters or command line with configuration file.

Thanks.

[KpeX]

Originally posted by SILICON
Please, add this feature to the list:

- Command line parameters or command line with configuration file.

Thanks. It's already there. Read the included documentation!

[cweb]

A new suggestion:
could a cutter feature be added to cut segments of an mpeg2 file? I'm thinking of DVB-S mpeg2 files mostly. Something was done by Dark Avenger in his DVD2AVI adaptation called MPEG2CUT, yet it isn't bug free and suffers from the problems you've been fixing (thanks again) in DVD2AVIdg.

The cutter feature would enable you to select a segment and save it to a standalone mpeg2 file.

[jfcarbel]

cweb- I like the cutter idea and am all for this.

We really don't have alot of nice user friendly choices to cut up MPEG2 clips without recoding and still keeping the audio sync.

For example,

DVDShrink has offered a start/stop feature, but it has been reported that there is audio sync problems with it at times.

Nero Recode offers the same start/stop functionality but it still does a re-encode and is not a true GOP cutter.

The only real solution that I have heard of that works without a hitch is Womble MPEG2VCR. Many people who have a WinPVR card use it to remove commercials. It always keeps audio sync and does no recoding so it is very quick.

Now I know that with vStrip GUI you could do it if you knew the LBA values, but something with a nice UI that allows you to scan thru the movie and play its audio to make sure you are at the exact points where you want to cut would be much nicer to use.

I think with minor modifications to DVD2AVI, audio playback could be added and GOP cutting added. Then we would have a great UI for getting clips from our DVDs. I know it gets away from what DVD2AVI is targeted for, but I think that the changes could benefit alot of us.