Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion. Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules. |
15th December 2018, 09:44 | #14741 | Link |
Registered User
Join Date: May 2016
Posts: 197
|
This means that either something is wrong with your source or that something is wrong with ffmpeg's decoder. The best place to continue would probably be ffmpeg's bugtracker.
What happens if you use the new truehd_core bitstream filter (designed to drop atmos extensions; usage: ffmpeg -i <inputfile> -vn -sn -c:a copy -bsf:a truehd_core -f <outputfile>)? If the errors always happen at the same place, you could upload just the few seconds of the track for an issue. (Btw: Does this Bluray use seamess branching and is the error at a branchpoint?) |
15th December 2018, 11:05 | #14742 | Link | |
Registered User
Join Date: May 2005
Posts: 1,462
|
Quote:
Thx for the tip! To what can I output the TrueHD stream, though? FFmpeg won't let me output it to .wav (probably because wav is 6 channels only, iirc). I will need a stream I can downmix with eac3to (-down6) again.
__________________
Gorgeous, delicious, deculture! |
|
15th December 2018, 11:50 | #14743 | Link |
Registered User
Join Date: May 2016
Posts: 197
|
The reason that ffmpeg won't let you mux the truehd stream into wav is because wav is a container for uncompressed PCM data, not for truehd data. You can e.g. mux it into Matroska. Just use a file with a "mka" extension and ffmpeg will know that you want to mux into Matroska.
Also keep in mind that if you intend to open a ticket in ffmpeg's bugtracker, you need to provide the truehd track with the atmos extension (or at least a snippet of it that allows to reproduce the issue). |
15th December 2018, 12:02 | #14744 | Link | |||
Moderator
Join Date: Feb 2005
Location: Spain
Posts: 6,915
|
Same to what?
Quote:
Please don't use .pcm, use .w64 or .wav instead. Some messages can be ignored. Quote:
Quote:
If: eac3to valerian.m2ts 2: c:\video\temp.w64 and: ffmpeg -i valerian.m2ts -vn -acodec pcm_s24le c:\video\temp.w64 produce the same messages and output you can ignore them or assume is a corrupt input. After that you can downmix the w64 (recommended with MeGUI or BeHappy).
__________________
BeHappy, AviSynth audio transcoder. Last edited by tebasuna51; 15th December 2018 at 12:09. Reason: typo |
|||
15th December 2018, 15:40 | #14745 | Link | ||
Registered User
Join Date: May 2005
Posts: 1,462
|
Good catch. Too used to .m2ts files. This test actually was on the mkverge .mkv version of the .m2ts (video and audio only), on my media server. You got me thinking, though, that maybe it got corrupted somehow. I just loaded up the disc again, but it makes no difference: about the same amount of errors.
Quote:
Quote:
__________________
Gorgeous, delicious, deculture! |
||
15th December 2018, 22:35 | #14746 | Link |
German doom9/Gleitz SuMo
Join Date: Oct 2001
Location: Germany, rural Altmark
Posts: 6,781
|
Raw PCM has no header. No information which sample rate, which bit depth, which number format, how many channels, which channel layout ... no details, only a byte stream of sample data. You will have to tell a reading application every attribute, or it won't be able to interpret the byte stream.
WAV has a header but is limited to 32 bit size fields (max. 4 GB); W64 expands this limit. FLAC is losslessly compressed. Last edited by LigH; 15th December 2018 at 22:38. |
16th December 2018, 02:36 | #14747 | Link | |
Registered User
Join Date: May 2005
Posts: 1,462
|
Quote:
__________________
Gorgeous, delicious, deculture! |
|
16th December 2018, 13:09 | #14749 | Link | |
Registered User
Join Date: May 2005
Posts: 1,462
|
Quote:
And yeah, even a 5.1 LPCM stream is nearly 7G. So, good tip on using .w64 format.
__________________
Gorgeous, delicious, deculture! |
|
21st December 2018, 18:33 | #14750 | Link |
Registered User
Join Date: Jul 2018
Posts: 1
|
I have a bunch of MKVs with FLAC audio (varying from FLAC 1.0, 2.0 and 7.1) that I want to convert to DTS-HD MA.
I thought i could attempt to do this with eac3to alongside the arcsoft, surcode or DTS-HD Master Suite software. I'm running Windows 10 FYI. Has anyone had good experience with latest windows trying to get this done? I run eac3to (using the UsEac3to GUI) and select the functionality to convert Track #2: FLAC to DTS from within eac3to. I always end up getting the error: "Pressing the Surcode "Encode" button didn't seem to work..." I tried making the file names 1 character long and on a very short file path (i.e. F:/Movies/a.mkv) but still get the same error, on either 1.0.21 or 1.0.29 Surcode version. I thought...well how about I just convert the FLAC to wavs (I tried it on a mono FLAC giving me just one center .wav file) then run it through Surcode...well Surcode converts that center mono file into a five channel DTS track (which I can successfully mux with the video but I feel like this is wrong...). So lastly, I wanted to try with the actual DTS-HD Master Suite software, which seems like the most viable. I tried converting the single center channel .wav file but the encode just seems frozen in my queue as pending... So any idea what would be the best way to tackle my goal of converting FLAC 1.0/2.0/7.1 from an MKV into DTS-HD MA 1.0/2.0/7.1 (respectively) in an MKV? I saw this which seemed to indicate people don't reccommend DTS-HD Master Suite... https://forum.doom9.org/showthread.php?t=158611 Also Arcsoft 6 doesn't even work on Window 10 anymore. I used this work around from another post that, when I run eac3to test, it detects Arcsoft dtsdecoder dll to be working (idk if it actually is though...). https://forum.doom9.org/showthread.php?t=154070 |
22nd December 2018, 04:16 | #14751 | Link |
Moderator
Join Date: Feb 2005
Location: Spain
Posts: 6,915
|
1) ArcSoft is a decoder, not a encoder. And is not necesary anymore because the free libdcadec.dll can do the decode of DTS-MA.
2) Surcode is a DTS encoder than can run with eac3to but can't produce DTS-MA, only standard DTS 768 or 1536 Kb/s. Forget it to recode FLAC. 3) If you want DTS-MA you need DTS-HD Master Suite and you only need eac3to to decode the FLAC's to monowav's: eac3to input.flac output.wavs Or in UsEac3to selecting 'wavs' in Output format After that you can use DTS-HD Master Suite with the monowavs input: Stream type DTS-HD Master Audio Channel Layout: from 1.0 to 7.1
__________________
BeHappy, AviSynth audio transcoder. |
18th January 2019, 17:09 | #14753 | Link |
Registered User
Join Date: May 2005
Posts: 1,462
|
Ran into this problem: HEVC multi-part demuxing? Any chance HEVC demuxing will become a reality with eac3to?
Thanks.
__________________
Gorgeous, delicious, deculture! |
18th January 2019, 17:46 | #14754 | Link |
RipBot264 author
Join Date: May 2006
Location: Poland
Posts: 7,815
|
Can you just join them with simple
copy /b 800.m2ts+801.m2ts+802.m2ts+803.m2ts+804.m2ts combined.m2ts
__________________
Windows 7 Image Updater - SkyLake\KabyLake\CoffeLake\Ryzen Threadripper |
7th February 2019, 06:33 | #14756 | Link | |
the Interrogator
Join Date: Oct 2001
Location: DownUnder
Posts: 664
|
Quote:
The audio streams I wanted were not the problem E-AC3 ones, so I extracted just that stream => X.thd+ac3. Then I ran eac3to [path] 1) -demux -core and let it extract all the streams. The X.thd+ac3 produced was binary identical to the one above. As you suggested, I would have imagined using -core would lose something, but it doesn't seem to have here, do you have any explanation? |
|
7th February 2019, 11:07 | #14757 | Link | |
Moderator
Join Date: Feb 2005
Location: Spain
Posts: 6,915
|
Quote:
A thd+ac3 is not a single audio track but 2 independent tracks interleaved. To extract each track you can use: eac3to X.thd+ac3 X.thd X.ac3 The -core parameter is ignored here.
__________________
BeHappy, AviSynth audio transcoder. |
|
12th February 2019, 01:40 | #14760 | Link |
Registered User
Join Date: Oct 2009
Posts: 42
|
Correction: the audio is 3.1 - Front & Side L R. I decoded part of it and found the L voice channel (must be channel 0) silent. I did come across a 2.0 mix of this audio, it had voice in one channel. Downmixing as it is probably won't do. How do I go about mixing or remapping this? Other than decoding to WAVs and replacing the silent channel. Mono? I didn't keep the log and the source AC3 but if you think it can help I'll get it.
Last edited by Masutin; 12th February 2019 at 02:45. Reason: Correction |
Tags |
eac3to |
Thread Tools | Search this Thread |
Display Modes | |
|
|