eac3to - audio conversion tool - Page 688

Boulder · 12th December 2015, 18:22

Is it possible to output the file created during the first pass to some different location than where the actual destination file will be written to? It would make things quite a bit faster if I could output it to a different HD.

arrgh · 12th December 2015, 19:28

Quote:

Originally Posted by Q-the-STORM

madshi, do you have any intention of including speedup/slowdown functionality for chapters? As far as I can see there are no chapter programs that let you speedup via CLI, so speeding up multiple files is a big hassle...

try ChapterGen.exe... it is old and a Little bit bitchy, but works in CLI....

Thunderbolt8 · 12th December 2015, 20:28

is eac3to actually multithreaded, does it use multiple cpu cores? I wonder how remux/demux speed would be affected in the not too distant future when PCIe drives hit the 1000MB/s write speed wall.

heerschop · 13th December 2015, 17:16

Quote:

Originally Posted by Groucho2004

Try building it with GCC.

Thank you for your suggestion. I tried to compile with gcc (mingw) but no luck. I kept keep getting "undefined reference" errors.
I am not a developer so I don't know where to begin to fix these errors.
After fiddling about with visual studio and gcc with no success I give up on building the dcadec.lib.

thanks

Boulder · 13th December 2015, 17:36

In Code Generation, you could try enabling the Enhanced Instruction Set option according to your CPU.

heerschop · 13th December 2015, 17:59

Quote:

Originally Posted by Boulder

In Code Generation, you could try enabling the Enhanced Instruction Set option according to your CPU.

In visual studio I have build the libdcadec.dll a zillion times with all sort of settings enabled or disabled. No matter what I did, all my builds were very slow in decoding a dts stream.
I even tried different visual studio versions (2010,2012,2013,2015).

Boulder · 13th December 2015, 20:05

Weird..I've just tested with VS2015 and the DLL it creates, doesn't appear to be slow. I didn't make any specific comparisons but looking at the eac3to progress meter, I couldn't tell if the DLL was updated or not.

I have a question of my own: Moonrise Kingdom has a DTS-HD MA track which shows up as 5.0ch but actually is 5.1ch. Decoding with eac3to doesn't proceed because of that, and dcadec.exe reports "Error writing WAV file: PCM output parameters changed". Is there a way to patch the header to make it work?

nevcairiel · 13th December 2015, 20:22

Quote:

Originally Posted by Boulder

I have a question of my own: Moonrise Kingdom has a DTS-HD MA track which shows up as 5.0ch but actually is 5.1ch. Decoding with eac3to doesn't proceed because of that, and dcadec.exe reports "Error writing WAV file: PCM output parameters changed". Is there a way to patch the header to make it work?

It probably changes from 5.0 to 5.1ch somewhere, and if dcadec and eac3to don't handle such changes, there isn't much that can be done.

You could try instructing it to mix to 5.1 always, maybe that helps, not sure.

heerschop · 13th December 2015, 20:35

Quote:

Originally Posted by Boulder

Weird..I've just tested with VS2015 and the DLL it creates, doesn't appear to be slow.

I am new to building with visual studio. This is how I made the build.
- First download and installed Visual studio 2015
- Downloaded the dcadec source files => dcadec-master.zip (from https://github.com/foo86/dcadec )
- extracted the dcadec-master.zip
- opened the dcadec.sln in visual studio
- visual studio asks if I want update the solution to version 140 => Yes
- Build the libdcadec as a release dll. ( The resulting build is smaller in size then the one that came with eac3to)
- Copy the libdcadec.dll to my eac3to folder

Eac3to uses my build without a problem but the decoding is very slow.
Am I missing something or do I use the wrong source files?

Boulder · 13th December 2015, 20:39

It seems to change almost immediately because the error message kicks in right after the decoding starts. The -down6 parameter doesn't help here either

Music Fan · 14th December 2015, 11:24

ffmpeg could maybe decode it.

hightime · 15th December 2015, 10:56

Just discovered eac3to (brilliant!) in the context of decode/converting TrueHD. The 7.1 audio I have needs to be converted to 5.1 so I used -down6 but I am curious about how it works. The side channels must get mixed into the back channels but do they also get mixed into the front channels? Also, how is clipping avoided? I converted the audio to a .flac file but this would not load into audacity so I made another conversion to .wav but this appeared to only convert about half the length of the audio. Should I use .pcm instead, or something else? The converted .wav file length is long (over 7 gbytes).

Apologies if these topics have been covered (I searched but couldn't find anything ... and I'm quite new to this ... ).

with thanks.

LigH · 15th December 2015, 11:19

WAV is restricted to 32 bit chunk sizes (<4 GB; depending on the interpretation of the first bit as sign, in carelessly programmed tools, possibly even <2 GB). You can try ".wavs" which generates several mono WAV files per channel, location appended to filename. Or try ".w64" for the WAV64 format.

kit90 · 16th December 2015, 15:44

When eac3to downconverts a 24-bit audio file to 16-bit, it applies Triangular PDF dithering automatically. Does it also apply any noise-shaping?

madshi · 16th December 2015, 15:51

No noise shaping, only TPDF dithering.

kit90 · 16th December 2015, 21:18

Quote:

Originally Posted by madshi

No noise shaping, only TPDF dithering.

Thanks

ACrowley · 21st December 2015, 15:02

Hi

I have a Problem with some AC3 5.1 Tracks from a HDTV TS Capture.
Its surely 5.1 Channel, but eac3to detects it as 2.0 and outputs only 2 Channel Left and Right .wavs

Is there a Way to correct the Header Info in the AC3 or a override in eac3to to Output 6 .wavs ?

thanks

Music Fan · 21st December 2015, 15:08

This is maybe 2.0 followed by 5.1 in the same track, it happens on some tv channels (movie in 5.1, ads in 2.0).
TSdoctor can cut audio/video on audio format change but it's not free.

ACrowley · 21st December 2015, 15:54

Hi

I cut the TS Streams with Videoredo Frame Accurate

And i tried TS Doctor with the AC3 Function, but eac3to still detects it as 2.0 AC3

EDIT
When i enable the TSDoctor Option Insert Ac3 5.1 Frames if needed then eac3to detects 5.1 AC3 and decodes to 6 CH .wavs

But the demuxed AC3 from this fixed TS Stream shows still 2.0 again

What works: i simply cut a few Single Frames at the Stream Beginning with Videoredo, then its 5.1 Channels

Q-the-STORM · 24th December 2015, 00:51

you can use ProjectX for that...
Go to PreSettings -> Audio -> check "replace all non-3/2 AC-3 by 3/2lfe silence"

demux the ac3, open it with projectX and click on "Quick Start"...
ProjectX will then replace all 2.0 audio in the track with 5.1 silence, making it possible for eac3to to detect it as 5.1 audio...

a feature like that was actually requested on the VideoReDo forums years ago, but it was never implemented...

12th December 2015, 18:22	#13741 \| Link
Boulder Pig on the wing Join Date: Mar 2002 Location: Finland Posts: 5,733	Is it possible to output the file created during the first pass to some different location than where the actual destination file will be written to? It would make things quite a bit faster if I could output it to a different HD. __________________ And if the band you're in starts playing different tunes I'll see you on the dark side of the Moon...

12th December 2015, 20:28	#13743 \| Link
Thunderbolt8 Registered User Join Date: Sep 2006 Posts: 2,197	is eac3to actually multithreaded, does it use multiple cpu cores? I wonder how remux/demux speed would be affected in the not too distant future when PCIe drives hit the 1000MB/s write speed wall. __________________ Laptop Lenovo Legion 5 17IMH05: i5-10300H, 16 GB Ram, NVIDIA GTX 1650 Ti (+ Intel UHD 630), Windows 10 x64, madVR (x64), MPC-HC (x64), LAV Filter (x64), XySubfilter (x64) (K-lite codec pack)

13th December 2015, 17:36	#13745 \| Link
Boulder Pig on the wing Join Date: Mar 2002 Location: Finland Posts: 5,733	In Code Generation, you could try enabling the Enhanced Instruction Set option according to your CPU. __________________ And if the band you're in starts playing different tunes I'll see you on the dark side of the Moon... Last edited by Boulder; 13th December 2015 at 17:39.

13th December 2015, 20:05	#13747 \| Link
Boulder Pig on the wing Join Date: Mar 2002 Location: Finland Posts: 5,733	Weird..I've just tested with VS2015 and the DLL it creates, doesn't appear to be slow. I didn't make any specific comparisons but looking at the eac3to progress meter, I couldn't tell if the DLL was updated or not. I have a question of my own: Moonrise Kingdom has a DTS-HD MA track which shows up as 5.0ch but actually is 5.1ch. Decoding with eac3to doesn't proceed because of that, and dcadec.exe reports "Error writing WAV file: PCM output parameters changed". Is there a way to patch the header to make it work? __________________ And if the band you're in starts playing different tunes I'll see you on the dark side of the Moon...

13th December 2015, 20:39	#13750 \| Link
Boulder Pig on the wing Join Date: Mar 2002 Location: Finland Posts: 5,733	It seems to change almost immediately because the error message kicks in right after the decoding starts. The -down6 parameter doesn't help here either __________________ And if the band you're in starts playing different tunes I'll see you on the dark side of the Moon...

14th December 2015, 11:24	#13751 \| Link
Music Fan Registered User Join Date: May 2009 Location: Belgium Posts: 1,744	ffmpeg could maybe decode it.

15th December 2015, 10:56	#13752 \| Link
hightime Registered User Join Date: Mar 2014 Posts: 1	how does down6 work for 7.1 files Just discovered eac3to (brilliant!) in the context of decode/converting TrueHD. The 7.1 audio I have needs to be converted to 5.1 so I used -down6 but I am curious about how it works. The side channels must get mixed into the back channels but do they also get mixed into the front channels? Also, how is clipping avoided? I converted the audio to a .flac file but this would not load into audacity so I made another conversion to .wav but this appeared to only convert about half the length of the audio. Should I use .pcm instead, or something else? The converted .wav file length is long (over 7 gbytes). Apologies if these topics have been covered (I searched but couldn't find anything ... and I'm quite new to this ... ). with thanks.

15th December 2015, 11:19	#13753 \| Link
LigH German doom9/Gleitz SuMo Join Date: Oct 2001 Location: Germany, rural Altmark Posts: 6,782	WAV is restricted to 32 bit chunk sizes (<4 GB; depending on the interpretation of the first bit as sign, in carelessly programmed tools, possibly even <2 GB). You can try ".wavs" which generates several mono WAV files per channel, location appended to filename. Or try ".w64" for the WAV64 format. __________________ New German Gleitz board MediaFire: x264 \| x265 \| VPx \| AOM \| Xvid Last edited by LigH; 15th December 2015 at 11:22.

16th December 2015, 15:44	#13754 \| Link
kit90 Registered User Join Date: Nov 2015 Posts: 11	When eac3to downconverts a 24-bit audio file to 16-bit, it applies Triangular PDF dithering automatically. Does it also apply any noise-shaping?

16th December 2015, 15:51	#13755 \| Link
madshi Registered Developer Join Date: Sep 2006 Posts: 9,140	No noise shaping, only TPDF dithering.

21st December 2015, 15:02	#13757 \| Link
ACrowley Registered User Join Date: Apr 2006 Posts: 1,008	Hi I have a Problem with some AC3 5.1 Tracks from a HDTV TS Capture. Its surely 5.1 Channel, but eac3to detects it as 2.0 and outputs only 2 Channel Left and Right .wavs Is there a Way to correct the Header Info in the AC3 or a override in eac3to to Output 6 .wavs ? thanks

21st December 2015, 15:08	#13758 \| Link
Music Fan Registered User Join Date: May 2009 Location: Belgium Posts: 1,744	This is maybe 2.0 followed by 5.1 in the same track, it happens on some tv channels (movie in 5.1, ads in 2.0). TSdoctor can cut audio/video on audio format change but it's not free.

21st December 2015, 15:54	#13759 \| Link
ACrowley Registered User Join Date: Apr 2006 Posts: 1,008	Hi I cut the TS Streams with Videoredo Frame Accurate And i tried TS Doctor with the AC3 Function, but eac3to still detects it as 2.0 AC3 EDIT When i enable the TSDoctor Option Insert Ac3 5.1 Frames if needed then eac3to detects 5.1 AC3 and decodes to 6 CH .wavs But the demuxed AC3 from this fixed TS Stream shows still 2.0 again What works: i simply cut a few Single Frames at the Stream Beginning with Videoredo, then its 5.1 Channels Last edited by ACrowley; 21st December 2015 at 17:14.

24th December 2015, 00:51	#13760 \| Link
Q-the-STORM Registered User Join Date: Sep 2012 Posts: 174	you can use ProjectX for that... Go to PreSettings -> Audio -> check "replace all non-3/2 AC-3 by 3/2lfe silence" demux the ac3, open it with projectX and click on "Quick Start"... ProjectX will then replace all 2.0 audio in the track with 5.1 silence, making it possible for eac3to to detect it as 5.1 audio... a feature like that was actually requested on the VideoReDo forums years ago, but it was never implemented...