eac3to - audio conversion tool

[tebasuna51]

I think it expects whole numbers for this?
You are right seems only work with integer values.

It does work with -no2ndpass, but then it has some, albeit minor, clipping.
Don't worry with the clipping, they are always encoder/decoder imperfections, the original source can't have peaks over 0dB.

Recode from lossy formats never can be exact.

[ramicio]

Eac3to is due for a refresh...........................................

[Overdrive80]

Had an interesting problem today when trying to go from AC3 -> FLAC with a channel remap (in preparation for oggenc2 encoding). Here's the log:

[CODE]AC3, 2.0 channels, 1:35:39, 224kbps, 48kHz
The Nero decoder doesn't seem to work, will use libav instead...

Do you have tried to use nero ac3 decoder??? Apparently, it isnt installed.

[dream88]

Eac3to is due for a refresh...........................................

Madashi has stated that once MadVR reaches v1.0 work on eac3to will continue.

[`Orum]

Do you have tried to use nero ac3 decoder??? Apparently, it isnt installed.

No. IIRC you needed a fairly old version of Nero for this, and I have a newer version installed which I needed for some new features (I forget what exactly).

[rapscallion]

Remember that theres a bigger Problem with the NeroAudio Decoder in eac3to! The DRC isnt removed properly so the Decoder is not recommended

I wouldnt use NERO AC3/EAC3 Decoder
http://forum.doom9.org/showthread.php?p=1404212#post1404212
Looks like DRC isnt disabled properly

Better use libav..works without Problems. And i dont think you hear big a Quality Difference generally between Nero and libavcodec on AC3/EAC3




AC hasn't been around for some time now.

Does anyone know if the above applies to both DD and True-HD ?

Anybody ??

[Overdrive80]

If eac3to doesnt remove DRC properly and I dont want use libavcodec. What could it be other option??? Maybe Audition??

[ramicio]

How doesn't it remove DRC properly? It's just a gain metadata. Why don't you want to use libavcodec? Why would one want to go from AC3 to FLAC?

[rapscallion]

@ramico, see my post above re the same, #11708.

[ramicio]

What?

[rapscallion]

What do you mean "what"? Did you read the post and the links.

[ramicio]

First off, my first post wasn't even replying to you, it was to Overdrive80. Secondly, I read it, and I don't see what it has to do with anything I said. The problem is not with eac3to, it's with a codec. I was asking Overdrive80 what was wrong with libavcodec, because it will function perfectly.

[rapscallion]

Whoa, slow down ther ramico.

I know you weren't replying to me, but Overdive's question was in response to my post, right above his.
You commented re DRC and metadata so my post has everything to do with what you said. I was hoping you could clarify things for him and me.

My post(s), I asked this back on the 15th as well (w/no responses): http://forum.doom9.org/showpost.php?p=1582897&postcount=11699 referenced the Nero codec, which ACrowley stated removal of drc was a problem.

We are here to help each other, no ?

[Overdrive80]

Ramicio, I havent any reason for not use libavcodec. My question is mainly based on the curiosity to know if there are alternatives.

my question is a consequence of assertion of rapscallion. Is there technical data for it??? I actually use nero for disable drc on my projects, and it worry me.

Thanks

[ramicio]

Why not just use libavcodec, which comes with eac3to, instead of a third party codec that doesn't even work?

[rapscallion]

Why not just use libavcodec, which comes with eac3to, instead of a third party codec that doesn't even work?
From the first post in this thread:
The libav/ffmpeg (E-)AC3 and DTS decoders work really well, but the output is not 100% identical to the Dolby/DTS reference decoders.

So, it has always seemed safer to me ( and Overdreive) to use Nero. Until, of course, the drc question. Do you know if the "problem" applies to both DD and TrueHD ?

[ramicio]

It shouldn't apply to TrueHD. It's really MLP in disguise, and is lossless. Specs weren't hidden for it. EVERY different lossy decoder will give a different decoded result. This is the nature of all of this psychoacoustic lossy pseudoscientific pseudomathematic BS. Want consistency? Go lossless.

[rapscallion]

It shouldn't apply to TrueHD. Go lossless.
Thanks for the info...that is what I was hoping !

Always go lossless. However, and just my personal prefrence, I use eac3to to create the wavs from a BD TrueHD track and then use DTS MA Suite to create a DTS-HD track to mux back with the original video.

Thus, my reason for asking.

[ramicio]

Why the love for DTS-HD?

[rapscallion]

Well, and this has certainly been a matter of discussion on the boards, I'm not a fan of DRC.

[ramicio]

I really don't get what you're talking about again. Just ripping the track will remove the DRC metadata. The audio inside of TrueHD will remain lossless. TrueHD > DTS-HD Master IMO because it allows you to run just the lossless portion without a core. But I don't like either codec in the end. I use FLAC.

[rapscallion]

Ok, I left out another important reason.

When playing a BD, with a TrueHD track, in the bedroom non-surround system, the volume is always too low, even when turned up to the tv's max.

[ramicio]

But this is before you do anything to it. Before you were obviously referring to copying your discs onto your computer. The DRC would be removed.

Playing back a blu-ray the TV would be decoding an AC-3 track, anyway, not a TrueHD track, or even a DTS-HD track.

I am still not following you.

[rapscallion]

BD to pc, pc to disc, to be played on a SA player.

Correct, however the core volume between DD and DTS is dramatically different.
I've found that eac3to doesn't extract TrueHD correctly (Mediainfo doesn't recognixe it and VLC won't play it), but it does fine with wavs as the output.

Tsmuxer demuxes TrueHD with the DRC intact.

Maybe the trouble following is that we have 2 completely differen goals. Thank you for the info, I do learn something new everyday even though I'm retired from 35 years in the audio business.

[ramicio]

BD to pc, pc to disc, to be played on a SA player.

I just have to ask. Why? I am assuming "pc to disc" means you are ripping it, doing something to it, and burning it to another disc. Eac3to does just fine demuxing TrueHD tracks and stripping the core. Trusting VLC for anything is part of your problem. Playing this back on a TV I wouldn't even bother using any audio track except the mandatory 2.0 DD.

[rapscallion]

Correct, ripping it. Doing something is eliminating all the fluff extras, warnings, duplicate studio trailers, menus, etc.

I mentioned the bedroom TV, however the sytem in our den is a complete surround sytem so just the 2.0 track is a no go.

VLC does play the Tsmuxer demuxed HD track fine but not the same by eac3to. Then there's Mediainfo.
You've got me curious though...gonna run a few tests.

[ramicio]

I just don't get the point of ripping something just to burn it again to remove the fluff. People remove the fluff when ripping to hard drives. I would just get into the program of using networked media players, store them on a NAS, and just use something like FLAC for the audio.

[rapscallion]

Well, to each his own. I didn't want to take this thread too far OT.

I prefer using an SA player. Additionally, I only do this for discs that have TrueHD, extremely long load times ( a problem w/my Pio BDP-51) or discs that require a new firmware update (my Sammy 1200 doesn't get fw updates anymore).

Just tested a True HD track w/eac3to and it will extract the TrueHD or the core , but not demux both the way Tsmuxer does.

I was mistaken re VLC, I should have stated Media Player Classic won't play the eac3to TrueHD.

[ramicio]

Have you tried something like "eac3to <input> track: output.thd+ac3"? I don't get what effect the codec would have on load times.

[rapscallion]

I don't use the command line, I use the eac3to and More GUI.

Codec doesn't affect load times, of course. I'm talking about the rip removing all the Java that does affect it.

[ramicio]

And that GUI won't show you the command? I don't get the point of what you're doing. It just seems like a giant waste of time. You're better off just getting a TV with Netflix or something, if you like wasting money.

[rapscallion]

And that GUI won't show you the command?
Yes, however you can't edit/add to it.

[ramicio]

Seems like a piece of crap to me. Eac3to is pretty straight forward to use. Typing commands is trivial compared to the time it takes to rip.

[Sparktank]

I've been curious about the undocumented swtiches and options that eac3to can use.

I've stumbled upon some here and there through the thread and kept an extra text file handy for such occasions. Also found some pointed out in the chagelog.

Is there a command or switch to display a complete list of operations?

-fullhelp

Or something like that?
I tried to google using different search functions and found pretty much the ones I ran into.

-normalize
-analyzebitdepth
-check16bit
-no2ndpass
-neroaacenc="differentpathifnotdefaulteac3tofolder"
-mono
-progressnumbers

[robpdotcom]

I don't use the command line, I use the eac3to and More GUI.

You could try UsEac3To. It has an option to extract as thd+ac3.

[Abradoks]

Is there a command or switch to display a complete list of operations?
There was an attempt to document them on wikibooks. Feel free to add anything missing there (http://en.wikibooks.org/wiki/Eac3to/How_to_Use#Command_Line_Syntax).

[Sparktank]

There was an attempt to document them on wikibooks. Feel free to add anything missing there (http://en.wikibooks.org/wiki/Eac3to/How_to_Use#Command_Line_Syntax).

:goodpost: Wow, I didn't even see that page before!
Thanks!

[tebasuna51]

There was an attempt to document them on wikibooks. Feel free to add anything missing there (http://en.wikibooks.org/wiki/Eac3to/How_to_Use#Command_Line_Syntax).

I add some parameters and info to this page.

[rapscallion]

You could try UsEac3To. It has an option to extract as thd+ac3.
Thank you, I wasn't aware of that GUI....works great.

[Lincoln Burrows]

Is there a way for me to use this kind of command without changing the bitrate for 128/44?

eac3to originalfile.mp3 stdout.wav -slowdown | lame - new-file-created.mp3

And how do I change the fps from a file (also without changing the bitrate/Hz) from 23.976 to 24? Or 24 > 23.976?

P.S. It seems the command is:

eac3to original-file new-file -23.976 -changeTo 24.000-23.976 = define fps from the source + new fps.

However, if I am dealing with AC3, the bitrate is also changed (this time, to 448 Kbps). Why?

[sneaker_ger]

Changing the speed means eac3to has to re-encode the audio and 448 is the default bitrate.

[tebasuna51]

With Lame you can specify the bitrate (CBR 128 Kb/s is the default) with:

CBR: lame -b 160 - new-file-created.mp3
VBR: lame -V 2 - new-file-created.mp3
With V (0 best .. 5 medium .. 9 low quality)

With internal AC3 encoder (448 default for stereo, 640 for 5.1)
you can use -192, -224, -256, -320, -384, -448, -512, -576 or -640 to specify your desired bitrate

[Skwelcha]

I get a "Patching bitdepth to 24 bits..." message when I try to extract the DTS core of a 16bit DTS-MA track and the DTS core is 24bit after that. Shouldnt it bet 16bit?

[Snowknight26]

Ignore that. Lossy formats don't have a bit-depth.

[ramicio]

Because you are extracting the lossy core. DTS is usually labeled as 24 bit despite people claiming that there is no such thing as bit depth in lossy formats.

[Skwelcha]

alright weird though that mediainfo reports 16bit sometimes and other times 24bit DTS tracks, but I will ignore it from now on then.
Thanks for the help.

[NanoBot]

The header of the DTS core contains a field "Source PCM resolution", which should indicate the bit depth of the original pcm files, before they where compressed with the DTS codec. Nevertheless, this value can be altered manually during / after the encoding, so it can be a fake value. And even if the source PCM really was 24bit PCM, this bit depth is likely to be almost lost, since the DTS core, as already said, is a lossy compression.

Therefore, if I would like to decode a DTS core to a lossless format like PCM or flac using eac3to, I would always use the "-down16" switch, since the additional size of a 24bit target file IMHO is wasted space on my HDD.

C.U. Nanobot

[MGSteve]

Has anyone had issues extracting the Right eye channel from Despicable Me 3D? Every time I do it, I get an out of memory error.

Only the right eye, everything else extracts fine. It even does it if I try and extract it straight from the disc.

[ramicio]

So, for 5.1 eac3to writes a 0x0000060F dwChannelMask, but MadFLAC ignores that and does 0x0000003F. I know madFLAC doesn't change all channel masks, because I have some that are odd and they still show through (0x00000107). Seeing that both things are by the same author, why 0x0000060F? 5.1 uses rear speakers, not side. The "S" when labeling "SR" or "SL" doesn't mean "side," it means "surround." Obviously the audio device needs the correct channel mask for mixing or whatever, so I think when eac3to deals with 5.1 it should do 0x0000003F.

Also, if I strip the header off of a WAV file and specify the input parameters, does it eac3to read the standard WAV channel ordering or does it assume some other funky (lpcm) ordering?

[tebasuna51]

So, for 5.1 eac3to writes a 0x0000060F dwChannelMask, but MadFLAC ignores that and does 0x0000003F. I know madFLAC doesn't change all channel masks, because I have some that are odd and they still show through (0x00000107). Seeing that both things are by the same author, why 0x0000060F? 5.1 uses rear speakers, not side. The "S" when labeling "SR" or "SL" doesn't mean "side," it means "surround." Obviously the audio device needs the correct channel mask for mixing or whatever, so I think when eac3to deals with 5.1 it should do 0x0000003F.
This was discussed already in this thread.

Until XP SP2, if I remember well, only the channel mask 0x0000003F was defined by M$ like standard 5.1 (Surround channels -> Back channels BL-BR), but after XP SP2 the channel mask 0x0000060F was the preferred to 5.1 (Surround channels -> Side channels SL-SR), maintaining the old mask like valid 5.1

Then new soft, M$ compliant, must accept both mask like equivalent for 5.1, and old soft maybe can reject 0x0000060F.
I recommended to madshi use always 0x0000003F, but ...

Also, if I strip the header off of a WAV file and specify the input parameters, does it eac3to read the standard WAV channel ordering or does it assume some other funky (lpcm) ordering?
If you use, for a 6 chan, 16 bits, 48 KHz:

-override -little -6 -16 -48000

the channel order is assumed like WAV

-override -big -6 -16 -48000

the channel order is assumed like PCM