Scenarist Simulation Works, but Disc Ignores Logic Programming!

[BassPig]

---See note below---

We're in a bit of a pickle.. it's after tech support hours and we have a rush job that was in rework because of some edits that were omitted.

The problem is that we have a program flow logic set up with setting GPRMs to control flow of programs.

There are four features plus a "play all", which looks at the GPRMs. The pre command of the firstplay PGC has a Mov GPRM10, 1. The Play All button has a Mov GPRM5, 1 command.
The post command of each feature has a compare to look and see if GPRM 5=10, if so, then play the next PGC in line.

The works on the simulation in Scenarist.

And it worked on the final DVDs that were delivered to the client.

However, after re-encoding the MPEG and AC3 files with the new edits, and re loading them into Scenarist, it still works perfect in the simulator, but the final DVD ignores the return to Root menu commands and behaves as if the Play All button and GPRM 5=10 situation were true.

This seems completely impossible. Simulator works fine, so we cannot troubleshoot the disc layout as there is nothing wrong with the disc programming. But the finished DVD has it's logic inverted. Play All only plays the first PGC now, and when one selects just ONE feature, the entire disc plays from start to finish, all features!

We've been working on this problem for the last 24 hours and we thought it was due to the disc capacity being too close to limit, so we reencoded all the video to a lower bitrate to bring the project down to 4.14GB, and burned a new disc, but the disc still behaves OPPOSITE of how the simulation of this disc behaves. Play All is play one track and Play one track is Play All on the finished DVD.

Maybe there are gremlins in the system, because all we did was update the MPEG & AC3 files, something we do all the time, but it never caused disc program flow to change before.

The problem we are having is that individual chapters end and don't return to the Root menu like they should. They behave on the disc as if GPRM5 were set, but it's not. We've been over the programming code many times--it's the same as it was when the discs worked. One last edit to the program has us really stuck now. This should have been a slam dunk and this should have been out the door to the replicator this morning, but for this absurd--almost mystical mystery.

What are we overlooking?

---NOTE---
Things are even stranger than first thought!
The player that acted up is an Oppo 981HD.
I took this disc over to a Pioneer DV343 and it played correctly there.
Then I took it to some Chinese-made Karoke player and it played correctly there too.
Then I played it on a Dell Inspiron laptop and ... it played correctly.

Then.. I returned it to the Oppo 981HD and... it now plays correctly!! Go figure!

Theory: we were in such a hurry to test, that the disc was still hot from the burner when inserted in the Oppo. Perhaps the dye in the disc had not settled out yet, hence producing some errors that manifest themselves in strange ways. That's my best guess.

[Eric69]

Kinda of hard to tell based on what you wrote...sorry. But why are you using GPRMs for the play all? No need for this. Just create 5 title. 4 titles contain the individual tracks and the 5th contains all your tracks back to back in seperate PGs - easy,elegant and no using any registers.

You can also track your register settings during simulation to debug your title as well but if it plays correctly on some players it's definitely a player issue.

E

[BassPig]

We've been authoring discs this way for about 4 years now, since the method you mentioned was our first method of achieving this goal, but it resulted in using a larger amount of disc space, because, as I recall, it duplicated the video streams twice. The GPRM method makes most efficient use of space on the disc, because PGCs are not duplicated.

[kumi]

Have you tried tracing through PgcEdit? You can watch the registers in real-time, as you playback. You can also dump the PGC commands in PgcEdit to pinpoint exactly what VM command(s) have changed between Disc A (successful) and Disc B (failure).

[bbjay]

If you are using Scenarist 3 and up you can use the same video clip as many times as you want in the Scenario and it will not take up anymore space than the one track takes up. It will not duplicate the video over and over again. Now if you want the "play all" Title to reorder the chapters/Scenes you will have a nonseamless break between the them if the are out of order. This is how I do all my projects will a play all now. The only gprms I really set for this type of project is for what button to highlight when pressing menu.

[Eric69]

I'm with bbjay. Do earlier versions of scenarist increase disc size this way? It shouldn't.

[bigotti5]

No version of scenarist does duplicate (unless you place the track into another VTS or language area).

[bbjay]

oops kind of forgot to provide some sort of solution to your original problem and not a substitute. Having worked with Sonic products (creator, Producer, Scenarist) for many many years I have found to never trust what you see in the simulator. Especially in Scenarist!! Even their tech support will tell you that. don't get me wrong I still use the simulator all the time but I dont look at the video playing I watch the SPRMs and GPRMs. Memorize them and you dont need to see any video at all to know if your DVD will work or not. It is the best way to know what is working and what is a bug with the program (which as we all know Scenarist is full of them).

So this bring me to tell you that if it didn't worked on one player one time, and then worked on it eventually this tells me that your Oppo player may be the problem. Some players are very strict to comply with the DVD spec and others do not. For Example, If your player can play SVCDs or mp3s then it does not follow the strict DVD specification. It could be possible the Oppo player does not store the registers properly.

If you still think it is on the Scenarist side, which it very well could be, I would delete any files not being used in the project and export a script file. Then import the script file as a new project. This usually cleans out any critters lurking from within the project file. Especially if you are replacing a lot of assets. Also when building the project always Build to an empty directory. Basically delete the build files it had before, this include mux directory, 0, Image files, lst files etc.

I hope this helps.

[BassPig]

Thanks for all the comments and suggestions, everyone.

I think the problem MAY have been the fact that I rushed the disc out of the burner and into the Oppo too quickly, and the dye on the disc may not have had enough time to settle (the other time I had trouble playing a disc, fresh out of the burner, it played as if someone had smeared peanut butter on the disc surface, but ten minutes later, it played the same part of the disc on the same player, flawlessly).

The discs are playing properly now in the Oppo and 3 other players tested. It does raise an interesting new question though: what is happening chemically with the dye on a DVD-R media, in the seconds following the burn process? It seems that the BERR is very high until the dye has had time to settle down.

In general, we have had few problems (that were not caused by our own attempts at trying to be clever with programming DVD titles) with Scenarist. This quirk discussed above is probably not a Scenarist issue but a DVD-R issue. I never thought that a DVD-R was like a cake, fresh out of the oven--you have to let it cool first before you do anything with it, but it may be the case. We were just backlogged with work and rushed at the beginning of the week. That's when strange things often happen.

[PhillipWyllie]

I'd open up you original DVD in PGCEdit and compare it with you newly authored one(in PGCEdit) to see if there are any differences.

Have you deleted your "mux" and "0" directory prior to re-authoring?
Have you renamed the new video(as opposed to naming it the same)? You can then, in the data editor using the property browser to specify a new location of the video asset(pointing to the new video). Scenarist won't use the old vif file.
Are the chapter points exactly the same?

Scenarist seems very picky when you change assets on the fly, if every thing isn't the same then things go wrong. You may need to reauthor though.

[BassPig]

Yes, I deleted the mux and 0 folder contents each time. I thought perhaps there was residual stuff in there causing a problem.

Ultimately, we've concluded that the problem was due to rushing the disc out of the burner and to a test on a set top player too quickly. The dye on the TDK media we are using may not have settled down, and so the disc behaved strangely on the initial play, but after moving it to a number of other players, it performed correctly, and then back to the player on which it had malfunctioned, it worked correctly once again. Same player that had the problem now worked, so we concluded that it must be something with the disc media not settling.

The usual list of gotchas we encounter with replacement of video in a scenario is if the length changes, sometimes the technician forgets to change the lengths manually in the track editor. On complex projects this happens more often.

Our next project is going to be interesting, because it is a symphony concert with a world-renowned conductor and guest pianist and we will be working with a surround mix plus a stereo mix and the DVD will want a special menu that allows the user to select the audio track of their preference. I've never built an audio menu before, so that will present some new challenges with implementation. But it's an exciting project that will stretch my skills to the limit with DVD authoring.