Log in

View Full Version : eac3to - audio conversion tool


Pages : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 [111] 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308

nautilus7
15th July 2008, 23:04
The libav decoder form eac3to isn't a directshow decoder, therefore it doesn't work for realtime decoding.

Beastie Boy
16th July 2008, 08:16
Is there no news on my feature-request, to be able to convert to audio to match a framerate of 24 fps (for making AVCHD's) ?

24fps is an abbreviation for 23.976fps which eac3to already supports. Blueray is also often quoted as 24fps but is in fact 23.976fps.

Cheers, Beastie.

G_M_C
16th July 2008, 09:29
24fps is an abbreviation for 23.976fps which eac3to already supports. Blueray is also often quoted as 24fps but is in fact 23.976fps.

Cheers, Beastie.

Nope, that wont fly; Its got to be 24.00000 fps, or video wont display (audio will though). Setting 24.0000 fps in tsMuxer is easy, but converting the audio to match the framerate remains the problem (syncing subs is easy too).

B4tm4n
16th July 2008, 12:35
Nope, that wont fly; Its got to be 24.00000 fps, or video wont display (audio will though). Setting 24.0000 fps in tsMuxer is easy, but converting the audio to match the framerate remains the problem (syncing subs is easy too).


I've made plenty of AVCHD's from 23.976fps MKV's and never had to convert to 24fps to playback on my PS3.

G_M_C
16th July 2008, 13:36
I've made plenty of AVCHD's from 23.976fps MKV's and never had to convert to 24fps to playback on my PS3.

Ye. that can be ... But i dont own anything with a Sony-brand on top, so no PS3 here. I have the Panasonic DMP-BD30, wich accepts only AVCHD that adhere to the standard perfectly (see here: http://www.avchd-info.org/format/index.html)

Atak_Snajpera
16th July 2008, 13:44
I have the Panasonic DP30, wich accepts only AVCHD that adhere to the standard perfectly (see here: http://www.avchd-info.org/format/index.html)
So how come PAL AVCHD camcorders like Canon record in progressive 25 fps ? :) (NTSC version also supports 30p)

24 fps is in fact 23.976 !
60 fps is 59.94 , 30 fps is 29.97

Do not change frame rate for god sake.

rickardk
16th July 2008, 13:57
24 fps is in fact 23.976 ! Do not change frame rate for god sake.

I know that when talking about 24 fps you often mean 23.976fps. Most manufacturer also talks about 24fps and means 23.976fps (24/1.001).

But I read in some document (a while ago) discussing correct timings for consumer electronics that 23.976fps are actually called 23fps and 24.0fps are called 24fps.

Arn and Relvolver uses 24.0fps. I know I have seen at least 5 or 6 of my blu-rays that uses 24.0fps.

Atak_Snajpera
16th July 2008, 14:03
23.976fps are actually called 23fps
Stupid idiots do not know how math works :)

G_M_C
16th July 2008, 14:03
So how come PAL AVCHD camcorders like Canon record in progressive 25 fps ? :) (NTSC version also supports 30p)

24 fps is in fact 23.976 !
60 fps is 59.94 , 30 fps is 29.97

Do not change frame rate for god sake.

Well, realise that i thought that too at first; But ive made AVCHD's with 23,976, and they dont work. Only when converted to 24.000 fps do they work, so what i say is based on actual fact.

I can try to see is speeding up to PAL works, but i doubt it; Based on my experiances; Bur 24 fps is closer to original than 25 fps, so i would still prefer 24 fps.

But we'd better get back on topic ;)

dorati
16th July 2008, 18:09
I have Problems with gaps.
I can run eac3to many times, but alwas the same message:

[a03] Audio overlaps for 5ms at playtime 0:19:08.
[a03] Audio overlaps for 5ms at playtime 1:20:15.
[a03] The audio file was demuxed without making use of the gap/overlap information.
[a03] Please rerun the same eac3to command line. That will correct the gaps/overlaps.

Here is the complete log:
eac3to v2.53
command line: eac3to f:\cars_fra 3) -demux
------------------------------------------------------------------------------
M2TS, 1 video track, 12 audio tracks, 8 subtitle tracks, 1:56:35
1: Chapters, 33 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: TrueHD/AC3, English, 5.1 channels, 48khz
4: AC3, English, 5.1 channels, 640kbit/s, 48khz, dialnorm: -27dB
5: DTS-ES, Italian, 5.1 channels, 24 bits, 1536kbit/s, 48khz, dialnorm: -4dB
6: AC3, Italian, 5.1 channels, 640kbit/s, 48khz, dialnorm: -27dB
7: DTS-ES, German, 5.1 channels, 24 bits, 1536kbit/s, 48khz, dialnorm: -4dB
8: AC3, German, 5.1 channels, 640kbit/s, 48khz, dialnorm: -27dB
9: AC3, English, 2.0 channels, 192kbit/s, 48khz, dialnorm: -27dB
10: AC3, English, 2.0 channels, 192kbit/s, 48khz, dialnorm: -27dB
11: AC3, Italian, 2.0 channels, 192kbit/s, 48khz, dialnorm: -27dB
12: AC3, Italian, 2.0 channels, 192kbit/s, 48khz, dialnorm: -27dB
13: AC3, German, 2.0 channels, 192kbit/s, 48khz, dialnorm: -27dB
14: AC3, German, 2.0 channels, 192kbit/s, 48khz, dialnorm: -27dB
15: Subtitle (PGS), English
16: Subtitle (PGS), English
17: Subtitle (PGS), Italian
18: Subtitle (PGS), German
19: Subtitle (PGS), Italian
20: Subtitle (PGS), German
21: Subtitle (PGS), English
22: Subtitle (PGS), English
Creating file "00147 - Chapters.txt"...
Audio gap description file detected, can't be used for TrueHD/MLP, though.
Audio gap description file detected, will be used for processing...
Audio gap description file detected, will be used for processing...
Audio gap description file detected, will be used for processing...
Audio gap description file detected, will be used for processing...
Audio gap description file detected, will be used for processing...
Audio gap description file detected, will be used for processing...
Audio gap description file detected, will be used for processing...
Audio gap description file detected, will be used for processing...
Audio gap description file detected, will be used for processing...
Audio gap description file detected, will be used for processing...
Audio gap description file detected, will be used for processing...
Audio gap description file detected, will be used for processing...
[v02] Extracting video track number 2...
[a03] Extracting audio track number 3...
[a03] Extracting audio track number 3...
[a04] Extracting audio track number 4...
[a05] Extracting audio track number 5...
[a06] Extracting audio track number 6...
[a07] Extracting audio track number 7...
[a10] Extracting audio track number 10...
[a09] Extracting audio track number 9...
[a08] Extracting audio track number 8...
[a12] Extracting audio track number 12...
[a14] Extracting audio track number 14...
[a11] Extracting audio track number 11...
[a13] Extracting audio track number 13...
[s15] Extracting subtitle track number 15...
[s16] Extracting subtitle track number 16...
[s17] Extracting subtitle track number 17...
[s18] Extracting subtitle track number 18...
[s19] Extracting subtitle track number 19...
[s21] Extracting subtitle track number 21...
[s20] Extracting subtitle track number 20...
[s22] Extracting subtitle track number 22...
[a03] Extracting TrueHD stream...
[a03] Extracting AC3 stream...
[a03] Removing dialog normalization...
[a03] Realizing (E-)AC3 gaps...
[a05] Removing dialog normalization...
[a07] Removing dialog normalization...
[a05] Realizing DTS gaps...
[a07] Realizing DTS gaps...
[a04] Removing dialog normalization...
[a06] Removing dialog normalization...
[a08] Removing dialog normalization...
[a08] Realizing (E-)AC3 gaps...
[a06] Realizing (E-)AC3 gaps...
[a04] Realizing (E-)AC3 gaps...
[a11] Removing dialog normalization...
[a10] Removing dialog normalization...
[a12] Removing dialog normalization...
[a14] Removing dialog normalization...
[a14] Realizing (E-)AC3 gaps...
[a13] Removing dialog normalization...
[a11] Realizing (E-)AC3 gaps...
[a10] Realizing (E-)AC3 gaps...
[a12] Realizing (E-)AC3 gaps...
[a09] Removing dialog normalization...
[a13] Realizing (E-)AC3 gaps...
[a09] Realizing (E-)AC3 gaps...
[v02] Creating file "00147 - 2 - h264, 1080p24.h264"...
[a05] Creating file "00147 - 5 - DTS-ES, Italian, 5.1 channels, 24 bits, 1536kbps, 48khz.dts"...
[a07] Creating file "00147 - 7 - DTS-ES, German, 5.1 channels, 24 bits, 1536kbps, 48khz.dts"...
[a03] Creating file "00147 - 3 - TrueHD+AC3, English, 5.1 channels, 48khz.thd"...
[a03] Creating file "00147 - 3 - TrueHD+AC3, English, 5.1 channels, 48khz.ac3"...
[a04] Creating file "00147 - 4 - AC3, English, 5.1 channels, 640kbps, 48khz.ac3"...
[a06] Creating file "00147 - 6 - AC3, Italian, 5.1 channels, 640kbps, 48khz.ac3"...
[a08] Creating file "00147 - 8 - AC3, German, 5.1 channels, 640kbps, 48khz.ac3"...
[s21] Creating file "00147 - 21 - Subtitle (PGS), English.sup"...
[a10] Creating file "00147 - 10 - AC3, English, 2.0 channels, 192kbps, 48khz.ac3"...
[a09] Creating file "00147 - 9 - AC3, English, 2.0 channels, 192kbps, 48khz.ac3"...
[a11] Creating file "00147 - 11 - AC3, Italian, 2.0 channels, 192kbps, 48khz.ac3"...
[a12] Creating file "00147 - 12 - AC3, Italian, 2.0 channels, 192kbps, 48khz.ac3"...
[a13] Creating file "00147 - 13 - AC3, German, 2.0 channels, 192kbps, 48khz.ac3"...
[a14] Creating file "00147 - 14 - AC3, German, 2.0 channels, 192kbps, 48khz.ac3"...
[s22] Creating file "00147 - 22 - Subtitle (PGS), English.sup"...
[s16] Creating file "00147 - 16 - Subtitle (PGS), English.sup"...
[s18] Creating file "00147 - 18 - Subtitle (PGS), German.sup"...
[s15] Creating file "00147 - 15 - Subtitle (PGS), English.sup"...
[s17] Creating file "00147 - 17 - Subtitle (PGS), Italian.sup"...
[a03] Audio overlaps for 5ms at playtime 0:19:08.
[a03] Audio overlaps for 5ms at playtime 1:20:15.
[a03] The audio file was demuxed without making use of the gap/overlap information.
[a03] Please rerun the same eac3to command line. That will correct the gaps/overlaps.
[s19] Creating file "00147 - 19 - Subtitle (PGS), Italian.sup"...
[s20] Creating file "00147 - 20 - Subtitle (PGS), German.sup"...
Video track 2 contains 167736 frames.
eac3to processing took 32 minutes, 8 seconds.
Done.

BLKMGK
17th July 2008, 00:48
Pastebin might be a good idea for long log posts...

Warrex
17th July 2008, 11:26
@ Madshi

Any chance you could include an option to demux the audio AND subtitle streams (thanks for including pgs support btw.) only?

I am asking this because:


m2ts files work fine for me in MeGui without remuxing them or indexing them with DGAVCDec.
Remuxing them is slow and wastes a lot of disk space.
Using "1:filename ..." is annoying and wastes time when having to enumerate all streams of interest (especially as there does not seem to be an option to let eac3to autoname the files)


****

Enumerating is also somewhat of a hassle for me concerning video streams that consist of multiple .m2ts files. It would be cool if eac3to would accept .mpls files as a source like tsMuxer does and get the correct files names and file order from it.

Rectal Prolapse
17th July 2008, 17:55
Keep in mind that there are a few titles out there where the video is explicitly set to 24.000 fps, NOT 23.976 fps (ie. Infernal Affairs Blu-ray, and another Hong Kong movie as well).

If you rip them and remux at 23.976 fps, your audio will be out of sync! Strange, but true!

Thunderbolt8
17th July 2008, 22:49
eac3to already should recognize them correctly so that you dont have to do any changes manually when remuxing (at least not so far)

zeropc
18th July 2008, 09:47
-core extract the DTS core of a DTS-HD track

does this also apply for the core extraction on a true-hd stream?

i'm asking cause it looked like it get's re-encoded.

nautilus7
18th July 2008, 13:27
No, it doesn't. To extract the AC3 core of a TrueHD track, you just set .ac3 as output extension. Note that only Blu-ray TrueHD tracks are TrueHD/AC3 combo. HD DVD TrueHD tracks are plain TrueHD.
If you still have problems, it would be useful to post the eac3to log.

fkl
18th July 2008, 22:13
hi everybody and sorry for my English...
we're not agree on tracker hdfrench about d.true-hd...is there a reason to convert d.true-hd with ac3 640kb/sec from blu-ray to dts 1536kb/sec? or downconvert this track to ac3 640kb/sec is the best way? many thanks for your response

Ranguvar
19th July 2008, 05:21
hi everybody and sorry for my English...
we're not agree on tracker hdfrench about d.true-hd...is there a reason to convert d.true-hd with ac3 640kb/sec from blu-ray to dts 1536kb/sec? or downconvert this track to ac3 640kb/sec is the best way? many thanks for your response
DTS 1,536Kbps from the TrueHD is going to be higher quality... extracting the 640kb/s AC3 stream is going to be smaller... and the first will take longer, as the second is just an extract, the first is a transcode.

fkl
19th July 2008, 11:32
DTS 1,536Kbps from the TrueHD is going to be higher quality... extracting the 640kb/s AC3 stream is going to be smaller... and the first will take longer, as the second is just an extract, the first is a transcode.

ok...thanks
in this case eac3to works with the core from the TrueHD or it makes 6 waves from de ac3 file?

bmnot
19th July 2008, 22:02
madshi, do you think eac3to in the future could have the ability to remove pulldown from 1080i60 AVC Blu-rays and mux to mkv at 1080p23.976?

A few 1080i60 FILM blu-rays have been popping up.

nautilus7
19th July 2008, 22:29
1080i60 isn't interlaced? How would muxing to 23,976 be possible?

Ranguvar
19th July 2008, 22:38
ok...thanks
in this case eac3to works with the core from the TrueHD or it makes 6 waves from de ac3 file?
I don't know...

To test, try extracting the TrueHD (.thd extension) and then encoding it, and try just specifying .dts extension... make a checksum of each, and compare.

If they're the same, eac3to is smart and encodes from the TrueHD. If not, it is encoding from the AC3.

bmnot
19th July 2008, 23:07
1080i60 isn't interlaced? How would muxing to 23,976 be possible?

Maybe I have the wording screwed up. Would it be apply pulldown? Doesn't 1080i60 stuff derived from film contain the instructions on how to recompose it back to 1080p23.976? if the pulldown sequence is constant?

nautilus7
19th July 2008, 23:56
I got your point now... but I don't think that's possible. 29,97p with pulldown flags --> 23,976p is possible, but with interlaced video I don't think this can been made.

Ranguvar
20th July 2008, 01:23
I got your point now... but I don't think that's possible. 29,97p with pulldown flags --> 23,976p is possible, but with interlaced video I don't think this can been made.
Agreed...

Unless it's 29.97p telecined that's pretending to be interlaced.

madshi
20th July 2008, 16:05
I got a HD DVD where I cannot demux the subtitles using evodemux into subrip (also crashes supread). If I rebuild the evo with just the subtitles. What is the command to demux it with eac3to?
Easiest way is to use "eac3to c:\HdDvdMovieFolder 1) -demux". That will demux all video, audio and subtitle tracks. If you want to demux the subtitle tracks, only, please browse a bit through this thread. It has been explained about a million times how it works.

Is there no news on my feature-request, to be able to convert to audio to match a framerate of 24 fps (for making AVCHD's) ?
It is on my to do list. But there are many things on my to do list and some of them I consider more important than your wish. As of yet you're the only person who has asked for 24.000 speedup/slowdown support. Find some more people who want/need that and I might bump up it's priority.

I have Problems with gaps.
I can run eac3to many times, but alwas the same message
It's half a bug in eac3to and half not a bug.

The problem is that you want to *DEMUX* a TrueHD track in its original form. You don't want it transcoded to WAV or PAM or FLAC, you want to keep it as TrueHD. And while that is not a problem at all for eac3to, TrueHD tracks can principally not be edited as well as other kinds of bitstreams. eac3to cannot delay TrueHD tracks and eac3to can also not fix gaps/overlaps in TrueHD bitstream tracks. That's due to the nature of how the TrueHD bitstream works. If I deleted a TrueHD audio frame in the middle of the stream, the decoders would complain about that and maybe even not decode the stream properly, anymore. So if you want gaps/overlaps to be removed you will have no other choice than to transcode the TrueHD track to something else.

It's "half a bug" because the message "Please rerun the same eac3to command line" doesn't make sense in this situation. But apart from this there isn't really anything eac3to can do right now.

However:

[a03] Audio overlaps for 5ms at playtime 0:19:08.
[a03] Audio overlaps for 5ms at playtime 1:20:15.
[a03] The audio file was demuxed without making use of the gap/overlap information.
[a03] Please rerun the same eac3to command line. That will correct the gaps/overlaps.
We're talking about an overall overlap of only 10ms here. You can really ignore this. No normal human will be able to notice a lipsync mismatch of 10ms.

Any chance you could include an option to demux the audio AND subtitle streams (thanks for including pgs support btw.) only?
I generally don't like adding hundreds of options because every new option makes the eac3to help more complicated. Yes, an option to demux only XYZ would save some time, but saving time is the only thing you'd gain which doesn't make adding such an option very attractive for me.

Check out the "eac3to and more GUI" (see first post). Maybe that GUI will do the trick for you.

in this case eac3to works with the core from the TrueHD or it makes 6 waves from de ac3 file?
eac3to always uses the TrueHD track - except if you explicitely ask to get an AC3 track. In that case of course the embedded AC3 bitstream will be extracted.

Maybe I have the wording screwed up. Would it be apply pulldown? Doesn't 1080i60 stuff derived from film contain the instructions on how to recompose it back to 1080p23.976? if the pulldown sequence is constant?
How does eac3to report these tracks? Are they reported as "1080i60" or as "1080p24 with pulldown flags"?

madshi
20th July 2008, 23:27
eac3to v2.54 released

http://madshi.net/eac3to.zip

* VC-1 pulldown removal rewritten (comparable to vc1conv 0.4, but faster)
* VC-1 pulldown removal is activated by default
* VC-1 pulldown removal can be manually deactivated by "-keepPulldown" option
* VC-1 pulldown removal is also available and activated when muxing to MKV now
* fixed Blu-Ray subtitle demuxing for seamless branching movies
* better task separation when doing multiple operations with an audio track
The seamless branching subtitles demuxing needs to be tested. Also the VC-1 pulldown removal needs some testing. It should be about twice as fast as vc1conv (if the harddisk is fast enough). The VC-1 pulldown removal done by eac3to is even a bit more complete compared to vc1conv. vc1conv leaves some flags in the bitstream but just sets them to zero. eac3to completely removes these.

EPiPH0NE
21st July 2008, 00:01
eac3to v2.54 released

http://madshi.net/eac3to.zip

* VC-1 pulldown removal rewritten (comparable to vc1conv 0.4, but faster)
* VC-1 pulldown removal is activated by default
* VC-1 pulldown removal can be manually deactivated by "-keepPulldown" option
* VC-1 pulldown removal is also available and activated when muxing to MKV now
* fixed Blu-Ray subtitle demuxing for seamless branching movies
* better task separation when doing multiple operations with an audio track
The seamless branching subtitles demuxing needs to be tested. Also the VC-1 pulldown removal needs some testing. It should be about twice as fast as vc1conv (if the harddisk is fast enough). The VC-1 pulldown removal done by eac3to is even a bit more complete compared to vc1conv. vc1conv leaves some flags in the bitstream but just sets them to zero. eac3to completely removes these.

So does this mean that VC-1 pulldown removal will actually work now? Up unitl now eac3to never actually removed the flags for me it just changed the FPS -> 23.976 and I would have to us vc1conv anyway. Well, I hope it works right now :) Thnx....Keep up the great work!

madshi
21st July 2008, 08:10
So does this mean that VC-1 pulldown removal will actually work now?
Yes.

Up unitl now eac3to never actually removed the flags for me it just changed the FPS -> 23.976 and I would have to us vc1conv anyway.
eac3to v2.53 and older were based on vc1conv 1.0 which didn't remove the pulldown properly. At one time I had even removed the pulldown removal functionality because of that problem but I put it back in on request. Now with v2.54 eac3to should be equal to (or even better than) the latest vc1conv.

Usage: Basically whenever eac3to sees a VC-1 stream with pulldown flags, it automatically removes the pulldown. So it's done when demuxing video from HD DVD or Blu-Ray. It's also done when remuxing video to MKV. It's also done when you do "eac3to source.vc1 dest.vc1".

Beastie Boy
21st July 2008, 08:16
Excellent update madshi. Once again, many, many thanks.

Cheers, Beastie.

G_M_C
21st July 2008, 08:29
[...]
It is on my to do list. But there are many things on my to do list and some of them I consider more important than your wish. As of yet you're the only person who has asked for 24.000 speedup/slowdown support. Find some more people who want/need that and I might bump up it's priority.
[...]

NP, glad to see its on the to-do list and i'll wait for it. In the mean time I'm using AviSynth / NicSound.dll / Soundout. My only problem with that option is, that i haven't found out how to output/manipulate to 6 wavs yet.

rickardk
21st July 2008, 10:26
Problem remuxing a dvd:

eac3to v2.53
command line: EAC3TO vts_01_1.vob+vts_01_2.vob+vts_01_3.vob+vts_01_4.vob+vts_01_5.vob 2: side.mkv 3: side.ac3
------------------------------------------------------------------------------
VOB, 1 video track, 1 audio track, 14:19:45
1: Joined VOB file
2: MPEG2, 480p24 /1.001 (16:9) with pulldown flags
3: AC3, 5.1 channels, 448kbit/s, 48khz, dialnorm: -27dB, -109ms
[v02] Extracting video track number 2...
[a03] Extracting audio track number 3...
[a03] Removing dialog normalization...
[a03] Applying (E-)AC3 delay...
[v02] Muxing video to Matroska...
[a03] Creating file "side.ac3"...
[v02] Video has a gap of 5 frames at playtime 0:00:01.
[v02] Video has a gap of 6 frames at playtime 0:00:01.
[v02] Video has a gap of 6 frames at playtime 0:00:02.
[v02] Video has a gap of 6 frames at playtime 0:00:02.
[v02] Video has a gap of 6 frames at playtime 0:00:02.
[v02] Video has a gap of 6 frames at playtime 0:00:03.
[v02] Video has a gap of 6 frames at playtime 0:00:03.
[v02] Video has a gap of 6 frames at playtime 0:00:04.
[v02] Video has a gap of 6 frames at playtime 0:00:04.
[v02] Video has a gap of 6 frames at playtime 0:00:04.

etc etc etc to the end...
Length is wrong.
Also delay of the ac3 track seems wrong. PgcDemux reports -25ms.




Pulldown cannot be removed

eac3to v2.53
command line: EAC3TO side.mpg side.mkv -strippulldown
------------------------------------------------------------------------------
MPEG2, 480p24 /1.001 (16:9) with pulldown flags
Removing MPEG2 pulldown...
Muxing video to Matroska...
The MPEG2 stream is a mixture of video and movie content.
The pulldown cannot be removed from this type of MPEG2 stream.
Will have to abort processing and redo everything.
Muxing video to Matroska...


The MPEG2 stream is not a mixture I'm pretty sure.

nautilus7
21st July 2008, 11:57
eac3to v2.54 releasedMadshi thanks for the new version. Seamless branching subtitles are working now.

fkl
21st July 2008, 12:37
thanks Madshi for your response and thanks from all hd-fans from France for your great work!

madshi
21st July 2008, 12:54
Problem remuxing a dvd:

eac3to v2.53
command line: EAC3TO vts_01_1.vob+vts_01_2.vob+vts_01_3.vob+vts_01_4.vob+vts_01_5.vob 2: side.mkv 3: side.ac3
------------------------------------------------------------------------------
VOB, 1 video track, 1 audio track, 14:19:45
1: Joined VOB file
2: MPEG2, 480p24 /1.001 (16:9) with pulldown flags
3: AC3, 5.1 channels, 448kbit/s, 48khz, dialnorm: -27dB, -109ms
[v02] Extracting video track number 2...
[a03] Extracting audio track number 3...
[a03] Removing dialog normalization...
[a03] Applying (E-)AC3 delay...
[v02] Muxing video to Matroska...
[a03] Creating file "side.ac3"...
[v02] Video has a gap of 5 frames at playtime 0:00:01.
[v02] Video has a gap of 6 frames at playtime 0:00:01.
[v02] Video has a gap of 6 frames at playtime 0:00:02.
[v02] Video has a gap of 6 frames at playtime 0:00:02.
[v02] Video has a gap of 6 frames at playtime 0:00:02.
[v02] Video has a gap of 6 frames at playtime 0:00:03.
[v02] Video has a gap of 6 frames at playtime 0:00:03.
[v02] Video has a gap of 6 frames at playtime 0:00:04.
[v02] Video has a gap of 6 frames at playtime 0:00:04.
[v02] Video has a gap of 6 frames at playtime 0:00:04.

etc etc etc to the end...
Length is wrong.
Also delay of the ac3 track seems wrong. PgcDemux reports -25ms.
Hmmmmm... Does it help if you use the "-60i" option? Can I have a sample of the DVD (first 50MB should be enough, if the eac3to complaint still occurs with those 50MB).

eac3to v2.53
command line: EAC3TO side.mpg side.mkv -strippulldown
------------------------------------------------------------------------------
MPEG2, 480p24 /1.001 (16:9) with pulldown flags
Removing MPEG2 pulldown...
Muxing video to Matroska...
The MPEG2 stream is a mixture of video and movie content.
The pulldown cannot be removed from this type of MPEG2 stream.
Will have to abort processing and redo everything.
Muxing video to Matroska...

The MPEG2 stream is not a mixture I'm pretty sure.
How do you know? This "is a mixture" comment does not mean to say that the movie in itself is a mixture of native movie and video content. The comment means that the MPEG2 encoder has produced a mixture or movie and video type encoding (some frames are encoded progressively, others interlaced). That happens very often with MPEG2 DVDs.

Encoder888
22nd July 2008, 04:30
This has probably been asked before, but has anyone heard anything about when we can expect a DTS-HD decoder capable of fully decoding 7.1 DTS-HD ? So far I've only encountered one title with DTS-HD 7.1 (The Bank Job) and I was surprised to find out a decoder out there simply doesn't exist that fully converts DTS-HD 7.1 without downsizing to 5.1, for eac3to to use... :(

Coolpplse
22nd July 2008, 05:16
This has probably been asked before, but has anyone heard anything about when we can expect a DTS-HD decoder capable of fully decoding 7.1 DTS-HD ? So far I've only encountered one title with DTS-HD 7.1 (The Bank Job) and I was surprised to find out a decoder out there simply doesn't exist that fully converts DTS-HD 7.1 without downsizing to 5.1, for eac3to to use... :(

Look back at the first post

The ArcSoft DTS decoder seems to be perfect for DTS and DTS-HD decoding. It seems to support every format and channel configuration that exists including 6.1 and 7.1.

This is now the best decoder for such conversion replacing Sonic.

Also tried it myself and it works with Flash Point's DTS HD Master Audio 7.1 track, thanks Madshi!

Encoder888
22nd July 2008, 06:00
I see, wow I can't believe I missed that. Two quick questions then:

Is there a way to only get the decoder or do you have to get the whole Total Media Theater?

And second, if I've used Sonic to this point on tracks that it's been able to perfectly decode that doesnt mean that ArcSoft would have done a better job, does it? In other words, I won't get better results if I reencode all of my files with Arcsoft, correct? ArcSoft helps on tracks where Sonic fails, but on ones where they would both work, ArcSoft doesnt outperform Sonic, does it ?

Coolpplse
22nd July 2008, 07:31
I see, wow I can't believe I missed that. Two quick questions then:

Is there a way to only get the decoder or do you have to get the whole Total Media Theater?

And second, if I've used Sonic to this point on tracks that it's been able to perfectly decode that doesnt mean that ArcSoft would have done a better job, does it? In other words, I won't get better results if I reencode all of my files with Arcsoft, correct? ArcSoft helps on tracks where Sonic fails, but on ones where they would both work, ArcSoft doesnt outperform Sonic, does it ?

I got TotalMedia Extreme, and the decoders seem locked to the whole package to the point where if you got the 15 day trial and it expires, the codec's refuse to function... and no I don't think you can only get the decoders themselves...

In my honest opinion I always felt sketchy to Sonic's output which was why I never attempted this DTS HD conversion since it was confirmed that it strangely goes up to 6.1 but not 7.1 which gives me some goose bumps that it may only be decoding DTS/DTS ES core tracks instead of the actual Lossless track.... there was no way for me to verify this either with my setup sound quality wise (using 4.1 at the moment) so this is just my pure opinion to the situation with Sonic's output.

EPiPH0NE
22nd July 2008, 07:55
I
And second, if I've used Sonic to this point on tracks that it's been able to perfectly decode that doesnt mean that ArcSoft would have done a better job, does it?



IMHO ArcSoft decoders sound better then Sonic. ArcSoft seems to give me a better sounding mix all the way from the sub range to the surrounds. I don't get really high, overly loud surrounds like I get with Sonic/libav. Since I only use 5.1 I get a bonus with Arcsoft as I can get a nice 7.1 -> 5.1 downmix that makes it that much better then Sonic. Just my 2 tokes. YMMV.

Encoder888
22nd July 2008, 08:14
Wow... I'm currently downloading ArcSoft and I will encode a DTS-HD 5.1 track that I previously encoded with Sonic to FLAC and compare the results... If the file sizes are exactly the same, then there shouldnt be any difference, right? Well, one thing about decoding DTS-HD with Sonic is that it ignores dialog normalization...

madshi
22nd July 2008, 09:13
Sonic and ArcSoft decode DTS bitwise identical, as long as the XXCh DTS extension isn't used. If the XXCh DTS extension is used Sonic falls back to 5.1 decoding. The XXCh DTS extension can be used for either 6.1 and 7.1 channels. For 6.1 usually the XCh DTS extension is used, though, which is fully supported by the Sonic decoder. ArcSoft supports everything.

I don't see any need to redo Sonic DTS-HD -> FLAC tracks, as long as Sonic had output the full number of channels.

The Sonic DTS decoder is the only one which was ever really useful from Sonic, though. All the other Sonic audio decoders are rather bad.

Warrex
22nd July 2008, 09:43
Warrex:
Any chance you could include an option to demux the audio AND subtitle streams (thanks for including pgs support btw.) only?

Madshi:
I generally don't like adding hundreds of options because every new option makes the eac3to help more complicated. Yes, an option to demux only XYZ would save some time, but saving time is the only thing you'd gain which doesn't make adding such an option very attractive for me. Check out the "eac3to and more GUI" (see first post). Maybe that GUI will do the trick for you.

I personally like adding options because it gives you flexibility. You could categorise them if you want to. Lame.exe works fine for people even with that many options. If one of eac3tos main design goals is simplicity you should rename it (as already the name is confusing) and really start working on a GUI yourself .

Using "eac3to and more GUI" is actually slower for me (no checkboxes for tracks) and its whole design is not very streamlined in my book.

Furthermore the option I requested is one that actually even every newbie would understand in contrast to many other ones of eac3to (which are of course still necessary and useful). But after all it's your call and I will have to keep using tsMuxer for demuxing as saving time and/or disk space is big for me.

Encoder888
22nd July 2008, 09:49
Thanks for the info madshi,

so, you're basically saying for any DTS-HD that's not higher than 5.1, Sonic and ArcSoft would produce identical results. well, what about dialog normalization though, isnt there going to be a difference there? and if so, is that gonna result in a difference in file sizes, or is the only way to tell be by listening...? sorry if those are dumb questions.

um, on an unrelated note, for AC3 decoding - i have Nero 8, which also has the Blu-Ray/HD-DVD plug-in, but i take it, it's not used by eac3to since I dont have Nero 7? So, if I have the plug-in with Nero 8 but not Nero 7, the Nero (E-)AC3 decoder is not going to be used for AC3 decoding?

I cant tell if it is being used, or if it is libav/ffmpeg, since eac3to isn't telling me which one its using... does that mean it's using libav/ffmpeg? if it was using Nero, would it say "using Nero AC3" or anything like that? I realize that libav/ffmpeg AC3 decoding doesnt produce a 100% identical copy of the original AC3 stream, which terrifies me a little bit, and I'd much rather use Nero. Any way I can do that w/o getting Nero 7?

Sorry, that was too many questions...

Coolpplse
22nd July 2008, 09:54
Sonic and ArcSoft decode DTS bitwise identical, as long as the XXCh DTS extension isn't used. If the XXCh DTS extension is used Sonic falls back to 5.1 decoding. The XXCh DTS extension can be used for either 6.1 and 7.1 channels. For 6.1 usually the XCh DTS extension is used, though, which is fully supported by the Sonic decoder. ArcSoft supports everything.


I'm actually quite curious about why these Lossless DTS HD MA/DD TrueHD DirectShow decoders can go up to 24 bit 96khz in output while current Blu-Ray/HD-DVD Playback software limits all of this to 16 bit 48khz... Was it really cause of PAP? Or do these apps simply limit the analog output to 16 Bit 48 Khz for compatibility with various sound cards? Even with my Xi-Fi prelude, I never got ffdshow to output beyond 16 Bit 48khz analog from FLAC files with higher bit depths and sample rates until I used madFlac...

madshi
22nd July 2008, 13:15
well, what about dialog normalization though, isnt there going to be a difference there?
I already answered that:

"Sonic and ArcSoft decode DTS bitwise identical, as long as the XXCh DTS extension isn't used."

um, on an unrelated note, for AC3 decoding - i have Nero 8, which also has the Blu-Ray/HD-DVD plug-in, but i take it, it's not used by eac3to since I dont have Nero 7?
See first post of this thread.

I cant tell if it is being used, or if it is libav/ffmpeg, since eac3to isn't telling me which one its using...
eac3to should tell you which decoder it uses.

I'm actually quite curious about why these Lossless DTS HD MA/DD TrueHD DirectShow decoders can go up to 24 bit 96khz in output while current Blu-Ray/HD-DVD Playback software limits all of this to 16 bit 48khz... Was it really cause of PAP? Or do these apps simply limit the analog output to 16 Bit 48 Khz for compatibility with various sound cards? Even with my Xi-Fi prelude, I never got ffdshow to output beyond 16 Bit 48khz analog from FLAC files with higher bit depths and sample rates until I used madFlac...
Cyberlink's decoder always downconverts everything to 16bit internally. They probably do that just to play safe. Technically it's no problem decoding to 24bit 96khz at all.

ffdshow uses libav for decoding which currently is limited to 16bit. madFlac uses the original FLAC source code which doesn't know any such limitations.

Encoder888
22nd July 2008, 13:30
@madshi

You're right, sorry I tend to repeat myself... I was just wondering about dialogue normalization, but then I remembered that DTS doesn't use dialogue normalization at all, it's Dolby's evil creation :P, so clearly the Sonic DTS decoder is gonna ignore it, because there is no such thing as dialnorm with DTS to begin with, be that DTS-HD or any kind of DTS stream.

I hate to be a pain, but with AC3 tracks eac3to really doesn't tell me which decoder it uses. Here's what I get:

eac3to v2.54
command line: D:\eac3to\eac3to.exe 00001.m2ts 3: Audio.ac3
------------------------------------------------------------------------------
M2TS, 1 video track, 4 audio tracks, 6 subtitle tracks, 1:21:05
1: h264/AVC, 1080p24 /1.001 (16:9)
2: DTS Master Audio, English, 5.1 channels, 24 bits, 48khz, -8ms
3: AC3, French, 5.1 channels, 448kbit/s, 48khz, dialnorm: -27dB, -19ms
4: AC3, Spanish, 5.1 channels, 448kbit/s, 48khz, dialnorm: -27dB, -19ms
5: AC3, English, 2.0 channels, 224kbit/s, 48khz, dialnorm: -27dB, -15ms
6: Subtitle (PGS), English
7: Subtitle (PGS), Spanish
8: Subtitle (PGS), Korean
9: Subtitle (PGS), Chinese
10: Subtitle (PGS), French
11: Subtitle (PGS), Spanish
[a03] Extracting audio track number 3...
[a03] Removing dialog normalization...
[a03] Applying (E-)AC3 delay...
[a03] Creating file "Audio.ac3"...
Video track 1 contains 116639 frames.
eac3to processing took 7 minutes, 57 seconds.
Done.


And I assume since I have the Nero plug-ins, but NOT Nero 7, it just uses libav/ffmpeg... :( Correct ?

Thanks again.

About the whole downconverting to 16-bit - I use CoreFLAC, doesn't that fully decode 24-bit, regardless of whether its by itself or in a container?

nautilus7
22nd July 2008, 13:40
Since source and output track is AC3 why do you think that a decoder is needed? Does eac3to anywhere say that is decoding the track? No, it says "extracting"...

Encoder888
22nd July 2008, 13:47
Oops, sorry as you can see I'm a newbie... So, I did try a different output format, and it does indeed use libav like I thought. Thanks anyway... For putting up with my questions hah.

madshi
22nd July 2008, 13:53
but then I remembered that DTS doesn't use dialogue normalization at all, it's Dolby's evil creation :P, so clearly the Sonic DTS decoder is gonna ignore it, because there is no such thing as dialnorm with DTS to begin with, be that DTS-HD or any kind of DTS stream.
Actually DTS does know dialnorm, it's just rarely used, and I believe both Sonic and ArcSoft ignore it.