Log in

View Full Version : Problems with timing and spliting...


*711
6th March 2008, 04:55
First of all, thanks for this wonderful forum... been trying to accomplish the backup of some of my DVD9's in x264/mkv and even though it has been like "hell" for me due to various VARIOUS hours of constant trial and error, the stuff I found here helped me a lot.

Now for the stuff that matters, I will post my pipeline just so you guys can take a glimpse of my process...

1. A quick pass on DVD Decrypter /// IFO (as described on the tutorial) with streams and chapters.

2. MeGui phase... D2v create, adjust ratio and interlace on AVS creator and then its just encode the audio (using nero-ac LC-Multichanel HQ 256 presset keeping original channels) and the video (AE-Max quality, even though I have to say it have MAJOR overkill choices, I simply want to keep things as "overkilled" as possible if I ever think about re-encoding this in the future ^^)

3. Now for SubRIP. Do a quick extraction using default stuff and save it in .srt w/ font color tags.

4. Out MeGui, enter MKVmuxer GUI. I use this to mux all up, 2 audio tracks 1 subtitle and 1 video. Save the original .ogg chapters in .xml and ONLY change the chapter names, the rest is left intact.

--------------- here is the first problem. After my muxing ends both my sub's and my chapters are "delayed" a bit. While it is very little, it IS quite annoying since the chapters are always a moment or 2 ahead of the original ones and the subtitles are always some milliseconds behind the audio. /// I tried to work them out by adding "minus 250 milliseconds" to the sub's witch didn't work out 100% perfect, some are still delayed. /// The chapters I simply remade the trimming ONE by ONE. Even though it turned out perfect I will never have the patience to make this again on my future projects ^^

5. After I got the "whole" muxed its time to split the whole thing so I can re-link it latter and skip repeated intros and stuff. I go again on MKV GUI and select to split "...after timecodes". Since I manually corrected the chapters I simply copy the timings from there to this place adding the appropriates commas between values and hit go.

----------------------- now comes my next problem. First because copying the values is a pain and I haven't found a way to "split by chapter" as I do with mp4's and Yamb yet. Second because the cuts aren't perfect. Some files just come 4 - 5 seconds wrong witch ruin the whole process.

6. IF I ever get to the stage where I have the files splited and ready to be relinked I will need a tutorial to relink all togheter, if anyone could point me a good one (already found some, but they are not much "noob-proof") I would be a happy fella.

So, sorry for the huge post but I thought it was necessary or you guys wouldn't understand where my issues were, witch are wrong splinting by mkvmerge and wrong timing (probably caused by MeGui?! I really don't know this one). AND because this way you guys can also recommend where I could improve my pipeline to get better results and / or solve my problems :)

Thanks for your time.

[P]ako
6th March 2008, 16:26
Check the frame rate with something like avinaptic. The frame rate is the first thing to look at when you are facing syncing problems.

*711
6th March 2008, 17:04
Downloaded the program and made a report (nice and useful software btw, thanks for the recommendation):


[ About file ]

Name: Witchblade Vol 02.mkv
Date: 5/03/2008 22:56:20
Size: 3,368,630,377 bytes (3212.576 MB)

[ Generic infos ]

Play duration: 01:34:41 (5680.7 s)
Container type: matroska
Creation time: 6/03/2008 01:52:16 UTC
Number of streams: 4
Type of stream nr. 1: video (V_MPEG4/ISO/AVC)
Type of stream nr. 2: audio (A_AAC (AAC Low complexity)) {2_0ch 256Kbps}
Type of stream nr. 3: audio (A_AAC (AAC Low complexity)) {3_2ch 448Kbps}
Type of stream nr. 4: subtitle (S_TEXT/UTF8) {Subtitle Track 01}
Audio streams: 2
Muxing Application: libebml v0.7.7 + libmatroska v0.8.1
Writing Application: mkvmerge v2.1.0 ('Another Place To Fall') built on Aug 19 2007 13:40:07

[ Relevant data ]

Resolution: VERY HIGH (720 x 480)
Width: multiple of 16 (GOOD)
Height: multiple of 32 (GOOD)
Average DRF quality: VERY HIGH (12.068405)
Standard deviation quality: HIGH (1.367039)
Std. dev. weighted mean: HIGH (1.087824)

[ Video track ]

Codec ID: V_MPEG4/ISO/AVC
Resolution: 720 x 480
Frame aspect ratio: 3:2 = 1.5
Pixel aspect ratio: 175:144 = 1.215277
Display aspect ratio: 175:96 = 1.822916
Framerate: 23.976024 fps
Stream size: 2,992,896,994 bytes
Play duration: 01:34:40 (5679.548693 s)
Bitrate: 4215.682837 kbps
Qf: 0.508764

[ Audio track nr. 1 ]

Codec ID: A_AAC (AAC Low complexity)
Channels (container): 2
Sample rate: 48000 Hz
Stream size: 183,406,337 bytes
Bitrate (container): 258.286953 kbps

[ Audio track nr. 2 ]

Codec ID: A_AAC (AAC Low complexity)
Channels (container): 6
Sample rate: 48000 Hz
Stream size: 188,794,519 bytes
Bitrate (container): 265.875006 kbps

[ About H.264 encoding ]

User data: x264
User data: core 58 svn-736M
User data: H.264/MPEG-4 AVC codec
User data: Copyleft 2003-2008
User data: http://www.videolan.org/x264.html
User data: cabac=1
User data: ref=16
User data: deblock=1:1:1
User data: analyse=0x3:0x113
User data: me=umh
User data: subme=7
User data: me-prepass=0
User data: brdo=1
User data: mixed_ref=1
User data: me_range=16
User data: chroma_me=1
User data: trellis=2
User data: 8x8dct=1
User data: cqm=0
User data: deadzone=21,11
User data: chroma_qp_offset=0
User data: threads=3
User data: nr=0
User data: decimate=1
User data: mbaff=0
User data: bframes=16
User data: b_pyramid=1
User data: b_adapt=1
User data: b_bias=0
User data: direct=3
User data: wpredb=1
User data: bime=1
User data: keyint=250
User data: keyint_min=25
User data: scenecut=40(pre)
User data: rc=2pass
User data: bitrate=4215
User data: ratetol=1.0
User data: rceq='blurCplx^(1-qComp)'
User data: qcomp=1.00
User data: qpmin=10
User data: qpmax=51
User data: qpstep=4
User data: cplxblur=20.0
User data: qblur=0.5
User data: ip_ratio=1.40
User data: pb_ratio=1.30
User data: aq=1:0.5:13.0
SPS id: 0
Profile: High@L5.1
Num ref frames: 16
Chroma format idc: YUV 4:2:0
PPS id: 0 (SPS: 0)
Entropy coding type: CABAC
Weighted prediction: No
Weighted bipred idc: B slices - implicit weighted prediction
8x8dct: Yes
Number of frames: 136173
Drop/delay frames: 0
Corrupted frames: 0

P-slices: 42787 ( 31.421 %) ########
B-slices: 91879 ( 67.472 %) #################
I-slices: 1507 ( 1.107 %)
SP-slices: 0 ( 0.000 %)
SI-slices: 0 ( 0.000 %)

[ DRF analysis ]

Average DRF: 12.068405
Standard deviation: 1.367039
Max DRF: 17

DRF<10: 0 ( 0.000 %)
DRF=10: 22341 ( 16.406 %) ####
DRF=11: 18861 ( 13.851 %) ###
DRF=12: 50996 ( 37.449 %) #########
DRF=13: 22190 ( 16.295 %) ####
DRF=14: 15691 ( 11.523 %) ###
DRF=15: 5122 ( 3.761 %) #
DRF=16: 940 ( 0.690 %)
DRF=17: 32 ( 0.023 %)
DRF>17: 0 ( 0.000 %)

P-slices average DRF: 10.888540
P-slices std. deviation: 1.059117
P-slices max DRF: 16

B-slices average DRF: 12.642442
B-slices std. deviation: 1.104255
B-slices max DRF: 17

I-slices average DRF: 10.569343
I-slices std. deviation: 0.901044
I-slices max DRF: 14

[ Profile compliancy ]

Profile to check: MTK NTSC
Resolution: Ok
Framerate: 23.976024 <> 30
Buffer underflow: 00:00:19 (frame 447)
Buffer underflow: 00:00:20 (frame 482)
Buffer underflow: 00:00:27 (frame 654)
Buffer underflow: 00:00:27 (frame 658)
Buffer underflow: 00:00:28 (frame 664)
Buffer underflow: 00:00:28 (frame 669)
Buffer underflow: 00:00:28 (frame 675)
Buffer underflow: 00:00:28 (frame 680)
Buffer underflow: 00:00:29 (frame 686)
Buffer underflow: 00:00:29 (frame 691)
Buffer underflow: 00:00:29 (frame 696)
Buffer underflow: 00:00:29 (frame 701)
Buffer underflow: 00:00:29 (frame 706)
Buffer underflow: 00:00:30 (frame 712)
Buffer underflow: 00:00:30 (frame 717)
Buffer underflow: 00:00:30 (frame 722)
Buffer underflow: 00:00:30 (frame 727)
Buffer underflow: 00:00:31 (frame 737)
Buffer underflow: 00:00:32 (frame 777)
Buffer underflow: 00:00:33 (frame 786)
Error: Too many violations

[ Chapters ]

Episode 5 - Prelude: 00:00:00,000 - 00:00:27,021
Episode 5 - Opening: "Psych: 00:00:27,021 - 00:01:57,075
Episode 5 - Title: "Search": 00:01:57,075 - 00:06:40,855
Episode 5 - Intermediate 01: 00:06:40,855 - 00:11:47,076
Episode 5 - Intermission: 00:11:47,076 - 00:16:12,090
Episode 5 - Intermediate 02: 00:16:12,090 - 00:21:40,085
Episode 5 - Closing: "Mamik: 00:21:40,085 - 00:23:10,052
Episode 5 - Preview: "Chang: 00:23:10,052 - 00:23:40,038
Episode 6 - Prelude: 00:23:40,038 - 00:23:55,052
Episode 6 - Opening: "Psych: 00:23:55,052 - 00:25:25,099
Episode 6 - Title: "Change": 00:25:25,099 - 00:28:26,992
Episode 6 - Intermediate 01: 00:28:26,992 - 00:34:17,148
Episode 6 - Intermission: 00:34:17,148 - 00:39:01,667
Episode 6 - Intermediate 02: 00:39:01,667 - 00:45:20,085
Episode 6 - Closing: "Mamik: 00:45:20,085 - 00:46:50,050
Episode 6 - Preview: "Past": 00:46:50,050 - 00:47:20,083
Episode 7 - Prelude: 00:47:20,083 - 00:47:35,101
Episode 7 - Opening: "Psych: 00:47:35,101 - 00:49:05,100
Episode 7 - Title: "Past": 00:49:05,100 - 00:53:05,433
Episode 7 - Intermediate 01: 00:53:05,433 - 00:59:05,116
Episode 7 - Intermission: 00:59:05,116 - 01:03:13,433
Episode 7 - Intermediate 02: 01:03:13,433 - 01:09:00,089
Episode 7 - Closing: "Mamik: 01:09:00,089 - 01:10:30,051
Episode 7 - Preview: "Recip: 01:10:30,051 - 01:11:00,080
Episode 8 - Prelude: 01:11:00,080 - 01:11:15,101
Episode 8 - Opening: "Psych: 01:11:15,101 - 01:12:45,145
Episode 8 - Title: "Recipro: 01:12:45,145 - 01:17:07,865
Episode 8 - Intermediate 01: 01:17:07,865 - 01:21:08,104
Episode 8 - Intermission: 01:21:08,104 - 01:26:42,066
Episode 8 - Intermediate 02: 01:26:42,066 - 01:32:40,136
Episode 8 - Closing: "Mamik: 01:32:40,136 - 01:34:10,098
Episode 8 - Preview: "Sadne: 01:34:10,098 - [undefined]

This report was created by AVInaptic (18-11-2007) on 6 mar 2008, h 13:14:43


Don't know if this helps a lot but... at least now you guys have all the pedigree on my encode ^^

As you can see: 23.976024 fps @ 4125 avg. bitrate.

As far as I knew this is the usual stuff after IVTC is applied isn't it? Also, could this be the reason? And if so, how to workaround this while still deinterlacing? The fact is that on my progressive monitor it looks a WHOLE lot better deinterlaced and I wouldn't want to spoil that.

Also, I made 2 marks (bold) on 2 things I didn't understand what they meant.... if you could please explain I would be thankful. First what are all those "buffer underflow" that resulted in some sort of "error". Second what means "MTK NTSC" (the NTSC part I understood tho ^^).

Thanks.

[P]ako
7th March 2008, 16:13
Are the subs delayed by approximately the same amount of time or they are further delayed as the video goes?

*711
7th March 2008, 21:07
It seems to get worse by the last 20 minutes or so of the movie.

[P]ako
8th March 2008, 03:39
This post
http://forum.doom9.org/showthread.php?t=135616&highlight=witchblade

is related to this one. I think the problems you are experiencing are due to the hybrid nature of the material. There are two solutions to it, one is to create a VFR video and the other one is to invert decimate each 23.976 and 29.97 section to 59.94 fps.