DVD authoring issues - any ideas?

[ArmyOfQuad]

So, for a while now, I've been using dvd-lab pro 2 to author the video portion of my dvd-a discs.

I've been trying to find the simplest solutions and methods, and eliminate redundancies and waste.

The process used to entail creating an mpg of a slide show for title slides, encoding the audio to dolby digital and dts, and importing all of that into a movie and creating a menu.

At some point I decided, this was more than I really need for what is primarily a music disc. I remember when sacd and dvd-a came out, as much as I liked the bells and whistles of dvd-a, at the end of the day I felt sacd had the solid audio format, operating just like a CD. And so I moved away from menus and title slides, which reduced a lot of time in the process, I didn't need to create a bunch of slides, and lay out a menu anymore. I would make an mpg of a single still slide of the album cover art, just to throw something on the screen. I also did away with dolby digital and kept dts only on the dvd-v side, as I really see no reason to have double the audio on the disc, and I don't really know of a receiver out there that is only doing dolby digital and not dts. I know that a dts only soundtrack isn't necessarily to spec, but....I really don't care. It's superior to dolby digital, I don't need the same audio content on a disc twice, it's wasteful and unnecessary.

And then I made a recent discovery, that dvd lab pro 2 has an option to add to a disc called audio title, which only requires adding audio. You can add a still to display while playing the audio content. This was exactly what I had been looking for, and it was right there in front of me. Rendering an entire mpg out of one still picture was just so wasteful.

However....I seem to have uncovered a problem by getting to this point. DVD-lab pro seems to have this annoying habit of not reading the timing of a dts program correctly, thinking it's over a minute shorter than it really is. This isn't a problem when you're laying it out along with an mpg that is timed properly, as it will make the program the length of the mpg and continue playing the audio past the time it thinks it cuts off at. But, now that I'm laying out dts only in an audio title, it decides to end the program at the point that it thinks the dts program ends, which results in the disc ending over a minute too soon.

Now, I suppose one may argue if I were authoring my dvds to spec, this wouldn't be an issue.

But, that's a cop out. The specs have just been hiding this issue.

What I can't figure out is, just why is dvd lab pro 2 not getting the proper time off the dts?

Has anyone else run into anything like this?

Is there anything I can do to get dvd lab pro to continue past the point it thinks is the cut off point? I tried creating a last chapter at the end point, but that hasn't worked.


I suppose I could add a few minutes of silence to the end. Sure, the disc would continue on for a bit maybe, but is that much of an issue?


I dunno, hopefully someone else can chime in with some insight.

 

[schultz1010]

So I'm not the only one being driven crazy by DLP. Audio only titles are not, to me, implemented correctly - nice idea but the developer lost interest before ironing out the kinks (there's a discussion of this over in another thread "Modern DVD-A authoring programs".

Neil Wilkes did mention this in the Modern thread (#40) "Encode your AC3 and DTS streams, make sure you use DTS Padded with DLP or you may get issues as it will accept .cpt but is not happy about it."

are you using the CPT (packed) or DTS (padded) variant for the DTS encoded files? Seems DLP prefers the padded version. No idea if this is your problem but might not be too hard of an experiment to rule it out.

 

[neil wilkes]

That could easily do it - .cpt files do not get properly parsed but .dts ones certainly do (they are the same length as the PCM stereo whereas .cpt will be wrong)
You also have to be careful with timecode in an audio-only title.
It uses the HH:MM:SS component of 29.97 NonDrop timecode

 

[schultz1010]

The thought just hit me that you may, like many on QQ, be decoding quadraphonic records (perhaps even using the Involve SM/Vinyl I found on another forum/thread on QQ) and creating DVDs. If that's the case I'm not as sure now as I was a while back that DVD-A is the way to go. It's 100% spec legal to put 4 channels of 48K/24b PCM audio on a DVD-V and you can get 2+ hrs of playing time. No need for any lossy compression be it AC3 or DTS. And you can use all the Abstraction Layer stuff you want Yeah, if you've got access to studio quality master material, more than 4 channels, believe 96K is a good thing , etc then DVD-A is needed but for quad conversions I'm almost ready to let DVDSP do it

 

[neil wilkes]

Whilst it is indeed absolutely spec legal to use 4 channels of 24/48 PCM (you can actually use 8 channels at 16/48) there has never been a player built that can read such discs. Multichannel PCM in DVD-Video is theoretically possible though, if you can hack the player firmware and/or the authorting software to accept such files (no abstraction layer tool will allow this) although Scenarist supports 48KHz and 96KHz PCM audio at 16-, 20-, and 24-bit resolutions up to 6 channels for 48 kHz but only 2 channels in 96 kHz. The problem is not that you cannot do this, only that no player I am aware of supports this.
There are other problems too - you will lose all the browseability, as DVD-Video is graphically dominant whereas DVD-A is audio dominant - the dominance refers to the primary part of the stream, and in DVD-V all audio is tied to a visual - change this & the audio will glitch.

 

[neil wilkes]

Yeah, if you've got access to studio quality master material, more than 4 channels, believe 96K is a good thing , etc then DVD-A is needed but for quad conversions I'm almost ready to let DVDSP do it

Have fun with that. DVDSP was Spruce Maestro until Apple screwed it up into DVDSP.
They also added a hell of a lot of spec fudges that are not strict spec, leading to potentially interesting usage issues.
Also with 96kHz audio personal belief systems do not enter into it at all - there are very valid reasons to use a 96kHz sample rate, especially with old material that may need restoration work (it has absolutely nothing to do with frequency response) and anybody who honestly cannot hear a difference between 96k and 48k is either using the wrong monitors, has a serious issue somewhere in the reproduction chain (this can often be a resampling sound card, such as a Creative Labs one) or needs to retrain their ears. Seriously. The reasons to use 96k or 48k in a mix session are likewise very little, if anything at all to do with extended frequency response either but we use a different process to decide a project sample rate in exactly the same manner you choose a tape speed (would those who say there is no difference between 48 or 96k also say there is no difference between 15 and 30IPS?) and for very similar reasons.

 

[ArmyOfQuad]

That could easily do it - .cpt files do not get properly parsed but .dts ones certainly do (they are the same length as the PCM stereo whereas .cpt will be wrong)
You also have to be careful with timecode in an audio-only title.
It uses the HH:MM:SS component of 29.97 NonDrop timecode

It seems .cpt is the only option I have with the current dts encoder that does dts 96 24....is there a way or an encoder that will do dts 96 24 to a .dts file?

 

[neil wilkes]

The only ones I know of are either the Nuendo DDE (plugin for Nuendo, no longer available & does not work properly with anything newer than 5.5) or the old DTS-Pro series encoder (the Java based one).
The DTS-HD MAS system only encodes to .cpt (and likewise only decodes from it too) - although renaming the .dts file as .cpt works for the STreamPlayer to decode it so maybe changing the extension from .cpt to .dts would work?
I have never tried this though.........

 

[schultz1010]

I've done it - created a DVD-V disc with 4 channels of 24/48 PCM in DVDSP and it played without a problem on Oppo 103 player. Average bit rate was only about 4.5Mb/s and no glitches or interruptions were noticed.

It started out as a slide show - much like what's been discussed in other threads - think "Audio only" track in DLP.

We're talking about (or I thought we were) audio only DVD-V - any video is nothing more than a still image with the menu buttons or perhaps the title of the song - couple hundred kbits/sec max. known as an "Audio title" in DLP there's no video - can even be a black background but usually has a still with the album and title of the song.

In DVDSP you create a slide show, set the still times to your exact desired time (better than 1s resolution), add the audio, then convert to a track. Now you've got a video track using timed cells, seamless playback and multi-channel PCM audio. Main reason I went and tried to create a DVD-A disc was I thought the Active Menus in Chrome would work and provide more interactivity during playback. Not so - big disappointment. But now that I think about it after conversion from a slideshow to a track in DVDSP I could go back and add BOV and voila - active menus. Might give that a try if DAC gives me too much grief.

No hacking of firmware and I do think DVDSP qualifies as an abstraction layer tool Trick was to create the multi channel Quicktime file with Compressor from individual mono channels as I recall (or perhaps Quicktime 7 Pro which has a Quadraphonic preset for the channel assignment)

Whilst it is indeed absolutely spec legal to use 4 channels of 24/48 PCM (you can actually use 8 channels at 16/48) there has never been a player built that can read such discs. Multichannel PCM in DVD-Video is theoretically possible though, if you can hack the player firmware and/or the authorting software to accept such files (no abstraction layer tool will allow this) although Scenarist supports 48KHz and 96KHz PCM audio at 16-, 20-, and 24-bit resolutions up to 6 channels for 48 kHz but only 2 channels in 96 kHz. The problem is not that you cannot do this, only that no player I am aware of supports this.
There are other problems too - you will lose all the browseability, as DVD-Video is graphically dominant whereas DVD-A is audio dominant - the dominance refers to the primary part of the stream, and in DVD-V all audio is tied to a visual - change this & the audio will glitch.

 

[schultz1010]

Have fun with that. DVDSP was Spruce Maestro until Apple screwed it up into DVDSP.
They also added a hell of a lot of spec fudges that are not strict spec, leading to potentially interesting usage issues.
Also with 96kHz audio personal belief systems do not enter into it at all - there are very valid reasons to use a 96kHz sample rate, especially with old material that may need restoration work (it has absolutely nothing to do with frequency response) and anybody who honestly cannot hear a difference between 96k and 48k is either using the wrong monitors, has a serious issue somewhere in the reproduction chain (this can often be a resampling sound card, such as a Creative Labs one) or needs to retrain their ears. Seriously. The reasons to use 96k or 48k in a mix session are likewise very little, if anything at all to do with extended frequency response either but we use a different process to decide a project sample rate in exactly the same manner you choose a tape speed (would those who say there is no difference between 48 or 96k also say there is no difference between 15 and 30IPS?) and for very similar reasons.

I meant to say, and it came off badly I see, that that 96K isn't needed at the end point playback - I can't hear the difference from a DVD-A between 192K and 96K for example. NOW in the mixing, restoration, editing, effects and so on then OF COURSE the higher rates are useful and necessary. But after all the work's been done then i think the higher rates do become more "mine's bigger than yours" than anything else.

 

[schultz1010]

By the old DTS-Pro series do you mean the one that DTS is now calling "DTS Surround Audio Suite"? For $249 http://www.dts.com/dts-store/store-landing.aspx I'm tempted to give it a try since I already have the iLock - from what I can see the encoder does both .cpt and .dts

If you on a MAC then the MASI DTS encoder is a plugin for Compressor-3 (not compatible with Compressor 4) not a standalone program.

The only ones I know of are either the Nuendo DDE (plugin for Nuendo, no longer available & does not work properly with anything newer than 5.5) or the old DTS-Pro series encoder (the Java based one).
The DTS-HD MAS system only encodes to .cpt (and likewise only decodes from it too) - although renaming the .dts file as .cpt works for the STreamPlayer to decode it so maybe changing the extension from .cpt to .dts would work?
I have never tried this though.........

 

[schultz1010]

It seems .cpt is the only option I have with the current dts encoder that does dts 96 24....is there a way or an encoder that will do dts 96 24 to a .dts file?

The MASI one available for CD/DVD use http://www.minnetonkaaudio.com/index.php?option=com_content&view=article&id=54&Itemid=87&lang=en will do .dts, .cpt or .wav but apparently only at 44.1 and 48 not 96 - the manual's available online at that link - I haven't looked at it closely.

 

[Neb-Maat-Re]

What I can't figure out is, just why is dvd lab pro 2 not getting the proper time off the dts?

Hi AoQ

I've encountered this phenomenon a number of times, and while I am by no means the expert in DVD creation in my view this is just a "quirk" in how DLP2 reads the DTS file and does not actually affect playback. I've been so confident this is a quirk rather than a glitch that I am in the habit of ignoring it.

To make certain I have been right about this I just created a new disc. The source file was 37:28 long. When I loaded the Dolby file into DLP2 it correctly reported as 37:28. When I loaded the DTS (*.dts) file it reported as 36:49.

I then entered the chapter points in line with the original 37:28 file length and created the VIDEO_TS. I then opened the VIDEO_TS in DVDShrink and unticked the Dolby track. The DTS track reported as 37:28.

I then created a DVD-A/V ISO in DiscWelder and burned a DVD-RW disc. I played the disc in my Panasonic DVD recorder (the only machine I have left that plays the DVD-V layer when a DVD-A layer is present) with the audio set to DTS. The chapter points correctly skipped to the start of each song (which wouldn't happen if the reason the file reported as 36:29 was that the track was sped up) and the last track played completely to the end of the song and a few seconds more for a total of 37:28.

My experience is this not an issue that affects the playback of the finished disc.

 

[schultz1010]

Hi AoQ

I've encountered this phenomenon a number of times, and while I am by no means the expert in DVD creation in my view this is just a "quirk" in how DLP2 reads the DTS file and does not actually affect playback. I've been so confident this is a quirk rather than a glitch that I am in the habit of ignoring it.

I am curious which DTS encoder you are using.

I then created a DVD-A/V ISO in DiscWelder and burned a DVD-RW disc. I played the disc in my Panasonic DVD recorder (the only machine I have left that plays the DVD-V layer when a DVD-A layer is present) with the audio set to DTS. The chapter points correctly skipped to the start of each song (which wouldn't happen if the reason the file reported as 36:29 was that the track was sped up) and the last track played completely to the end of the song and a few seconds more for a total of 37:28.

A very nice feature of the OPPO brand of BluRay/DVD players (I have the Oppo 103) is that DVD-V or DVD-A zone playback is a user selectable option. You can use the setup menu to select which zone gets played when a DVD-Audio disc is detected - that's how I can check out both the Video_TS and Audio_TS areas to make sure they work in a set top box rather than only a computer.

 

[Neb-Maat-Re]

I use SurCode DVD-DTS.

I am aware that Oppo players have that capacity.

 

[schultz1010]

I use SurCode DVD-DTS.

when specifying the output file type do you use the '*.dts' (Compact) or the '*.dtp' (Padded) type?



There has been discussion on various threads that DVD Lab Pro works best with the Padded type.

 

[Neb-Maat-Re]

The file was *.dts.

I create both the Dolby and DTS files using Audiomuxer which calls up Surcode. That makes it a one-click process straight from the multi-channel FLAC and I don't have to load the individual WAVs into Surcode. I think that process means *dts is the default. I don't specify it.

The next time I am playing around with DLP2 I could load the WAVs into Surcode and create a *.cpt file to test if that overcomes the false time report, but as I have said I have never found the inaccurate time report on loading the *dts file to be an issue with the completed VIDEO_TS file or disc playback.

 

[Neb-Maat-Re]

I created a *cpt file from the same source.

It also incorrectly reported as 36:29 when loaded into DLP2.

 

[schultz1010]

I created a *cpt file from the same source.

It also incorrectly reported as 36:29 when loaded into DLP2.

CPT is the compact form - not the padded form so this new file is the same compact form that DLP has reported incorrect time for in the past. According to

http://wiki.multimedia.cx/index.php?title=DTS

.cpt is the compact form. This is also what http://www.reviversoft.com/file-extensions/cpt says. Somehow it appears that the same compact type of file as before has been created. The expectation was that the file extension for a padded file would be .dtp (as shown in the picture in #16). Is this not the case and the manual is wrong?

The Surcode manual shows the .dtp extension rather than .cpt for the padded variant. Is the manual incorrect and the encoder does not create .dtp (packed) files?

I thought the experiment was to create a .dtp file rather than a .cpt file (which we know shows the incorrect time).

 

[Neb-Maat-Re]

The drop down list for my installation of Surcode differs from the image you have posted.

Mine lists DTS Padded as *.dts and DTS Compact as *.cpt.

There is no option to create an *.dtp file.

It would appear from that I have been using a padded version all along.