multiangle size end up too big

[HyperYagami]

I have a concert DVD which the whole thing is seamless multiangle'd (angle 1 = with lyrics for karaoke, angle 2 = without). After doing it via DIF4U DC4U and RA and the final Scenarist product ends up 400MB too big, which is about the same size as all the audio tracks combined together. Apparently, even though both video streams are using the same audio tracks Scenarist create a copy for each video streams.

Is it what seamless multiangle is supposed to be, where no space is saved? Or there's something in Scenarist that I need to change?

Thanks.

[influenza]

Seamless angles use the same audio for both angles. How did you set up the tracks in scenarist?

[HyperYagami]

well, I tried both ways, via using the RA-generated script and did it manually ->

1) create a track via dragging video-stream #1
2) create a new angle
3) dragging video-stream #2 onto the new angle
4) create the audio tracks

the size of the audio tracks is about 400MB, and the size different between a project with audio tracks and w/o is 800MB, which is about twice the size as the audio tracks, and that's why my conclusion at this point is that with multiangle you're not saving any space. Even when I checked the original, the VOBs is 7.46GB conbined and the demux (m2v, ac3, dts) is only 6.58GB and the audio tracks (ac3, dts) is 760MB combined.

[influenza]

But you've only added one audio track right? I never noticed the audio being used twice and so the project ending up too big.

How did you demux, by vobid?

[HyperYagami]

Yes, only 1 track. When I check the angle'd videostream it has no audio tracks attached to it.

I think it's just that most of the time people don't notice it. Most of the time you have a movie which only part of it is multiangle'd, while in this case the whole thing is MA'd (hence no need to rip via VOBID):

PGC_1 (program chain): [Title(TTN): 1] [01:09:47.15 / 30 fps] (Programs: 8) (Cells: 18) (uses VOB-IDs: 1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2)
[Ch 01] [Pg 01] [Cell 01] [V/C Id: 1/ 1] (Angle 1): time: 00:22:57.18 / 30 fps [Pos: 00:22:57.18] [Frames: 41328]
[Cell 02] [V/C Id: 2/ 1] (Angle 2): time: 00:22:57.18 / 30 fps [Pos: 00:22:57.18] [Frames: 41328]
[Ch 02] [Pg 02] [Cell 03] [V/C Id: 1/ 2] (Angle 1): time: 00:11:47.27 / 30 fps [Pos: 00:34:45.15] [Frames: 62565]
[Cell 04] [V/C Id: 2/ 2] (Angle 2): time: 00:11:47.27 / 30 fps [Pos: 00:34:45.15] [Frames: 62565]
[Cell 05] [V/C Id: 1/ 3] (Angle 1): time: 00:09:09.03 / 30 fps [Pos: 00:43:54.18] [Frames: 79038]
[Cell 06] [V/C Id: 2/ 3] (Angle 2): time: 00:09:09.03 / 30 fps [Pos: 00:43:54.18] [Frames: 79038]
[Ch 03] [Pg 03] [Cell 07] [V/C Id: 1/ 4] (Angle 1): time: 00:04:11.00 / 30 fps [Pos: 00:48:05.18] [Frames: 86568]
[Cell 08] [V/C Id: 2/ 4] (Angle 2): time: 00:04:11.00 / 30 fps [Pos: 00:48:05.18] [Frames: 86568]
[Ch 04] [Pg 04] [Cell 09] [V/C Id: 1/ 5] (Angle 1): time: 00:04:36.27 / 30 fps [Pos: 00:52:42.15] [Frames: 94875]
[Cell 10] [V/C Id: 2/ 5] (Angle 2): time: 00:04:36.27 / 30 fps [Pos: 00:52:42.15] [Frames: 94875]
[Ch 05] [Pg 05] [Cell 11] [V/C Id: 1/ 6] (Angle 1): time: 00:02:43.18 / 30 fps [Pos: 00:55:26.03] [Frames: 99783]
[Cell 12] [V/C Id: 2/ 6] (Angle 2): time: 00:02:43.18 / 30 fps [Pos: 00:55:26.03] [Frames: 99783]
[Ch 06] [Pg 06] [Cell 13] [V/C Id: 1/ 7] (Angle 1): time: 00:07:50.27 / 30 fps [Pos: 01:03:17.00] [Frames: 113910]
[Cell 14] [V/C Id: 2/ 7] (Angle 2): time: 00:07:50.27 / 30 fps [Pos: 01:03:17.00] [Frames: 113910]
[Ch 07] [Pg 07] [Cell 15] [V/C Id: 1/ 8] (Angle 1): time: 00:04:18.24 / 30 fps [Pos: 01:07:35.24] [Frames: 121674]
[Cell 16] [V/C Id: 2/ 8] (Angle 2): time: 00:04:18.24 / 30 fps [Pos: 01:07:35.24] [Frames: 121674]
[Ch 08] [Pg 08] [Cell 17] [V/C Id: 1/ 9] (Angle 1): time: 00:02:11.21 / 30 fps [Pos: 01:09:47.15] [Frames: 125625]
[Cell 18] [V/C Id: 2/ 9] (Angle 2): time: 00:02:11.21 / 30 fps [Pos: 01:09:47.15] [Frames: 125625]

And what I have for files:

08/30/2003 11:22 AM 161,792 VTS_01_0.IFO
08/31/2003 03:29 AM 142 VTS_01_PGC_01_ANGLE_1 - Chapter Information - CCE.txt
08/30/2003 11:22 AM 468 VTS_01_PGC_01_ANGLE_1 - Stream Information.txt
08/31/2003 03:29 AM 142 VTS_01_PGC_01_ANGLE_2 - Chapter Information - CCE.txt
08/30/2003 11:31 AM 474 VTS_01_PGC_01_ANGLE_2 - Stream Information.txt
08/30/2003 11:32 AM 100,598,784 VTS__01_P01A1-80-192K-[-66]ms(fixed)-ch2Chinese.AC3
08/30/2003 11:32 AM 100,598,784 VTS__01_P01A1-81-192K-[-66]ms(fixed)-ch2Chinese.AC3
08/30/2003 11:32 AM 201,197,568 VTS__01_P01A1-82-384K-[-66]ms(fixed)-ch6Chinese.AC3
08/30/2003 11:34 AM 395,319,772 VTS__01_P01A1-8B-768K-[-66ms]ms(fixed)-ch5Chinese.DTS
08/30/2003 11:35 AM 345,822 VTS__01_P01A1.I-BFF.4~3_1.d2v
08/30/2003 11:22 AM 3,104,487,240 VTS__01_P01A1.I-BFF.4~3_1.M2V
08/30/2003 11:35 AM 256 VTS__01_P01A1.I-BFF.4~3_1.nopull.AVS
08/30/2003 11:36 AM 345,936 VTS__01_P01A2.I-BFF.4~3_1.d2v
08/30/2003 11:31 AM 3,170,023,240 VTS__01_P01A2.I-BFF.4~3_1.M2V
08/30/2003 11:36 AM 256 VTS__01_P01A2.I-BFF.4~3_1.nopull.AVS

[influenza]

If you have only one video stream and the angle has no audio attached to it as you suggest that would mean your audio stream is shorter the the video duration?

I would do it by vobid. i guess you won't have the oversize problem anymore then.

[HyperYagami]

Originally posted by influenza
If you have only one video stream and the angle has no audio attached to it as you suggest that would mean your audio stream is shorter the the video duration?

I would do it by vobid. i guess you won't have the oversize problem anymore then.

no, what I meant is just the angle'd scene has no seperate audio tracks attached, i.e., it uses the same tracks (supposedly anyway).

Via vobid would get your the exact same thing, since the whole vobid #1 is the original and vobid #2 is the angle'd scene. I'm pretty sure Scenarist is making 2 copies of the audio tracks one for each angle, although I'm not sure if this is what it's supposed to be.

[influenza]

I should have to test it to be sure, I really never noticed it. Ending up 400 MB too big is really annoying though. Reencoding again I guess?

[HyperYagami]

Originally posted by influenza
Reencoding again I guess?

Yeah it's the only way, I'm just surprised that no one had deal with that before...

Thx for your help though.

[Eyes`Only]

You shouldn't ever get duplicate audio tracks-- In the last few versions of DoItFast4U, I coded the demux to only keep one audio track when doing angles.

[influenza]

The question is:

when you reauthor a movie with multiple angles and add only one audio stream to it. Will the audio be repeated for the second angle. I wouldthink no, but HyperYagami says it happens. So that would have nothing to do with doitfast4u, but with the muxing of audio and video in the authoring package?

I think it's a strange problem. i've done a lot of multi-angle stuff and never noticed this. Although I have to admit I always did this by vobid and HyperYagami just extracts the entire PGC and imports this into scenarist.

[HyperYagami]

Originally posted by Eyes`Only
You shouldn't ever get duplicate audio tracks-- In the last few versions of DoItFast4U, I coded the demux to only keep one audio track when doing angles.

yes, I'm not getting duplicate audio tracks after DIF4U, it's only after the Scenarist step, that I notice the final Scenarist product is too big - just about the same size as all the audio tracks. I got the same result even if I do it manually (i.e., w/o using the RA script) and even the **original source** shows the same behavior. For example:

07/08/2003 03:33 AM 13,944,832 VTS_01_0.VOB
07/08/2003 03:33 AM 1,073,565,696 VTS_01_1.VOB
07/08/2003 03:36 AM 1,073,565,696 VTS_01_2.VOB
07/08/2003 03:39 AM 1,073,565,696 VTS_01_3.VOB
07/08/2003 03:42 AM 1,073,565,696 VTS_01_4.VOB
07/08/2003 03:46 AM 1,073,565,696 VTS_01_5.VOB
07/08/2003 03:49 AM 1,073,565,696 VTS_01_6.VOB
07/08/2003 03:52 AM 695,885,824 VTS_01_7.VOB
8 File(s) 7,151,224,832 bytes

(VTS_01_0.VOB shouldn't be there but it's only 13MB so that's OK.)

After demux:


09/01/2003 08:35 PM 2,796,682,056 VTS__01_P01A1.I-BFF.4~3_1.M2V
09/01/2003 08:43 PM 2,796,683,976 VTS__01_P01A2.I-BFF.4~3_1.M2V
2 File(s) 5,593,366,032 bytes

09/01/2003 08:43 PM 88,951,296 VTS__01_P01A1-80-192K-[-66]ms(fixed)-ch2Chinese.AC3
09/01/2003 08:44 PM 88,952,064 VTS__01_P01A1-81-192K-[-66]ms(fixed)-ch2Chinese.AC3
09/01/2003 08:44 PM 177,904,128 VTS__01_P01A1-82-384K-[-66]ms(fixed)-ch6Chinese.AC3
09/01/2003 08:45 PM 349,551,802 VTS__01_P01A1-8B-768K-[-66ms]ms(fixed)-ch5Chinese.DTS
4 File(s) 705,359,290 bytes

so 5,593,366,032+705,359,290 = 6298725322, which is way less than 7,151,224,832.

but 5,593,366,032+705,359,290+705,359,290 = 7004084612, which is about right, and I really do not think the extra 700+MB is just overhead.

[influenza]

Well let's say there's an overhead of about 5-10%. 6298725322*0,05=314936266. Wich makes more sense then saying that the audio is duplicated and there's virtually no overhead?

Note: I'm guessing here, no idea what the actual overhead would be.

[HyperYagami]

Originally posted by influenza
I'm guessing here

Well I'm guessing too. But here is the thing, like I said I have tried to make 2 Scenarist projects that contain 2 video streams (one is the angle'd version of the other) but one with audio and the other one w/o. After authoring both projects with Scenarist the size difference between the final VOBs is about twice the size of all the audio tracks combined. It's an experiment that you can try...

[gbr]

thanks to influenza for pointing out this thread to me. i am having the exact same problem and yes scenarist is muxing the audio with each video track. you can check this by looking at the muxed data directory (unclick auto delete). what i have is 3 angles, each full length and two audio tracks also full length. let's say for ease of understanding that the video is 700mb each and the audio is 300mb each. when scenarist does the mux it will take video track one and mux it with both audio tracks then do the same with angle 1 and angle 2 ending up with 3 vob.t files ~1.3gb now those are combined to make the final vob set of 3.9gb (ignoring the overhead that authoring progs add) when what i wanted was 3 angles and 2 audio tracks totaling 2.7gb (only adding the audio track once). i guess this helps me understand the dvd structure even more and makes sense when you think about what is happening...the audio is multiplexed with the video...not played seperately...but it certainly sucks because it feels like wasting 1.2gb.

in the end...because this is 60's film of the apollo mission and low quality audio, i used besweet to convert from 448k to 128k audio, thus cutting my audio track by just over 1/3 and gaining the extra room.

eyes and aqua may want to take this into consideration when doing their dvd size calculations (unless of course i am out to lunch)

[HyperYagami]

yeah I've been trying to get some dvd tech papers about this and see how this is supposed to be handled, but no luck. :-(

so I don't know if that's a scenarist thing or not.