Log in

View Full Version : Question about Encoding TV Episode DVD


Angelus
18th February 2005, 02:14
Hi everyone I am trying to back up...well I'll just say it im not embarassed...The OC Season 1 and it seems that there is only one main VTS, VTS 1. The thing is that DIF4U unchecks what seems to be each individual episode, and leaves checked one larger stream which includes all the 4 episodes. I encoded it the first time with the one big stream and all seemed fine with the encode all though I was suspicious since that one stream was marked interlaced. Anyways I tried just checking each individual episode seperately and it only found the 1st episode as "interlaced". When I encoded the project with each episode individually it only came out to 3.78 GB...when I encoded the long stream with all the episodes together it came out fine.

Anyways is there anyway to calculate the bitrate for having multiple "main movies" as in my case if I have 4 episodes in one folder? Or does deinterlacing a non-interlaced source mess it up at all? Or would it just leave it alone? Sorry for my horrible explanation =)

Here's a pic of DIF4U window:

http://img194.exs.cx/img194/7751/dif4u6dk.jpg (http://www.imageshack.us)

jel
18th February 2005, 03:07
The OC Season 1?!?!?!!
ha ha ha ha ha ha ha ha ha <breathes> ha ha ha ha ha! :D
only joking mate, couldnt resist the open invitation though ...

anyways, i have no idea why you would be getting the difference in encode sizes, depending on which pgc's you select, as that vts is flagged as main ...could you post part of your ifo for vts_01, that shows the structure?
oh and also your ccedata.txt files for the two different methods you have tried might tell us something as well.

cheers
j

Angelus
18th February 2005, 03:21
Here's the CCEData of the one big stream with all the episodes:

D:\DVD\OC1\VTS01\VTS__01_P01.I-TFF.4~3_1.nopull.AVS",0,9608,3101,0,4

Here's the CCEData of the all the episodes seperate:

C:\DVD_Backup\OC1Test\VTS01\VTS__01_P02.I-TFF.4~3_1.nopull.AVS",0,9608,3101,0,4
C:\DVD_Backup\OC1Test\VTS01\VTS__01_P03.P.4~3_1.AVS",0,9608,3101,0,4
C:\DVD_Backup\OC1Test\VTS01\VTS__01_P04.P.4~3_1.AVS",0,9608,3101,0,4
C:\DVD_Backup\OC1Test\VTS01\VTS__01_P05.P.4~3_1.AVS",0,9608,3101,0,4

All though I increased the average bitrate for both encodes up to 3175 because there was an extra I could do without in the VideoTS folder (DIF4U provided me with the calc using the 1 large stream).

I just posted the IFO of the original DVD...also I've had this problem with another Episodic DVD before, undersizing, and it was because I used BatchEncodeM2V for the encoding.

Angelus
18th February 2005, 03:24
Here's the IFO sorry

D3s7
18th February 2005, 05:28
Ok..... your initial question seems to potentially be 2 questions mixed together?

If I'm correct, your PGC1 uses all the same vobid's as PGC2-5 (which would make since by how DIF4U reacted) - so, with that in mind, i'm not exactly sure what your asking...

The bitrate for grabbing PGC1 -vs- PGC2-5 should almost be identicial except if PGC2-5 share multiple vobids...

Example
PGC1 = vobid 1,2,3,4,5,6,7
PGC2 = vobid 1,6,7
PGC3 = vobid 2,6,7
PGC4 = vobid 3,6,7
etc
in that case, you'd get a lower bitrate by demuxing PGC 2-5 then just PGC1 (because you can see 6,7 are repeated)

As far as the second question w/ deinterlacing..... The detection routines in DIF4u are VERY good (especially for NTSC) so if it flagged them as interlaced, your probably ok...... Now if your saing PGC1 was flagged as interlaced but 2-5 weren't, then I would bring that up to Eyes`Only because the material is identical and the detection should be as well....

If you are attempting to decomb a progressive source though, I'd highly recommend against taht... Decomb has a very specific purpose, to get rid of the comb in an interlaced movie. It does that by blending the fields. Blending a progressive source will do nothing but lower quality (anyone jump in here if this isn't exactly correct)

Angelus
18th February 2005, 05:51
sorry for not being more clear. I encoded the disc last night using the long stream which includes all 4 episodes in that one stream (PCG 01 in this case). Anyways I started watching it and it seems the quality was somewhat to be desired. So I thought I would uncheck that large stream in DIF4U and check each individual episode by itself (PCG 2-5). When I did this, DIF4U found PCG2 as interlaced, top field first. And the other PCG;s, 3-5 were all marked progressive. So this left me wondering if the quality would be better if I encoded it this way rather than with PCG 1 which has all of the episodes.

Actually looking at it now, all of the episodes are interlaced/hybrid sources I guess. This is the script from VTS__01_P03.P.4~3_1.AVS

import("C:\Tony\DVD Backup Programs\DoItFast4U\new.avs\addaudio.avs")
LoadPlugin("C:\Tony\DVD Backup Programs\DoItFast4U\new.avs\mpeg2dec3.dll")
Mpeg2Source("VTS__01_P03.P.4~3_1.d2v",idct=0)
LoadPlugin("C:\Tony\DVD Backup Programs\DoItFast4U\new.avs\decomb.dll")
Telecide(guide=1)
Decimate(mode=3,threshold=3)
AddAudio()
ConvertToYUY2(interlaced=true)

Yet in the latest DGIndex, it says the frame type is progressive, and the video type flashes between FILM 96% and NTSC 4%. There is some combing when you preview it in there. Maybe this clears things up more? Still encoding it the 2nd way using each episode seperate seems like it would be a better fit?