AC3 TrueHD to DTS

[GZZ]

I have a question about AC3 TrueHD conversion to DTS. I can do a conversion of a AC3 TrueHD (640 kbit core + HD) to DTS 1536 kbit. But the question is, will this give better quality sound compared to just the 640 kbit Core AC3.

The DTS 1536 kbit will be far larger in size compared to AC3 640 kbit, but will it sound better when the source is AC3 TrueHD ?

GZZ

[madshi]

Might depends a bit on the DTS encoder. If you have access the latest official DTS encoder and encode with 1.5Mbps, that should easily beat 640kbps AC3 is audio quality. However, a DTS track encoded by the Surcode DTS encoder might not sound much better than a studio provided 640kbps AC3 track. But in the end why don't you try it out and check yourself which sounds better to your ears?

[GZZ]

But when eac3to decode the AC3 TrueHD to 5.1 channel wav does it then also read the HD part of the track or only the core. If it only read the core when converting to wav, then keeping the AC3 Core will be just as good as the DTS at 1536 kbit. ?

[rica]

If i were you i would extract the thd first:
eac3to input output (*.thd)

Then i would convert to dts:

eac3to input(*.thd) output (*.dts -1536)

[GZZ]

didnt know that was possible. So then it only use the HD part of the AC3 to encode the DTS. I will try that out. Thanks.

[rica]

It is gonna work, don't worry.

[madshi]

When asking to transcode, eac3to decodes TrueHD and not AC3, of course.

[hollydoc]

I have a question about AC3 TrueHD conversion to DTS. I can do a conversion of a AC3 TrueHD (640 kbit core + HD) to DTS 1536 kbit. But the question is, will this give better quality sound compared to just the 640 kbit Core AC3.

The DTS 1536 kbit will be far larger in size compared to AC3 640 kbit, but will it sound better when the source is AC3 TrueHD ?

GZZ

Is there a way to convert TrueHD to DTS-HD ?

Hollydoc

[rica]

Yes, there is one: 1500 USD :)

[baudi]

Might depends a bit on the DTS encoder. If you have access the latest official DTS encoder and encode with 1.5Mbps, that should easily beat 640kbps AC3 is audio quality. However, a DTS track encoded by the Surcode DTS encoder might not sound much better than a studio provided 640kbps AC3 track. But in the end why don't you try it out and check yourself which sounds better to your ears?

Which is, in your opinion, the best DTS encoder?

I have realised that Surcode DTS encoder is not precise, fail with timestamp. Before encoding ==> 1:36:19.840, after encoding ==> 1:36:28.453. Very imprecise encoder!

Thank you.

[madshi]

Which is, in your opinion, the best DTS encoder?
The official one from DTS, of course.

I have realised that Surcode DTS encoder is not precise, fail with timestamp. Before encoding ==> 1:36:19.840, after encoding ==> 1:36:28.453. Very imprecise encoder!
Wrong. Just run the DTS track through eac3to to cleanup the zero padding. Afterwards runtime will be correct again. Surcode is not imprecise at all.

[GZZ]

Wrong. Just run the DTS track through eac3to to cleanup the zero padding. Afterwards runtime will be correct again. Surcode is not imprecise at all.
Is this needed at all or does it only correct the timestamp, nothing more. ?


About the AC3 TrueHD to DTS.

Are there any difference between:

(Extracting AC3 TrueHD from m2ts)
eac3to Input.m2ts Output.dts -1536

and

(Extracting AC3 TrueHD from m2ts to THD)
eac3to Input.m2ts output.thd

(Convertering THD to DTS)
eac3to Output.thd Output.dts -1536

Will the result be the same or does the first only use the Core AC3 (not the HD) ?

[rica]

Wrong. Just run the DTS track through eac3to to cleanup the zero padding. Afterwards runtime will be correct again. Surcode is not imprecise at all.

Yes,
After getting dts rerun eac3to:

eac3to input.dts output.dts

(I think it is related with version number; if you have Surcode 1.0.21 you have to make a second pass.

[GZZ]

(I think it is related with version number; if you have Surcode 1.0.21 you have to make a second pass.

I have 1.0.23. Does this also need the second pass ?

[rica]

I have 1.0.23. Does this also need the second pass ?

You can try using this way:

Just try to open created dts with TSMuxer; if it doesn't recognize the dts; it means eac3to has to be run again.

[GZZ]

You can try using this way:

Just try to open created dts with TSMuxer; if it doesn't recognize the dts; it means eac3to has to be run again.

TSmuxer recognize it just fine. So this problem is properly fixed in this version (1.0.23).

[GZZ]

d:\eac3to>eac3to "E:\ripfiles\new\output.thd" "E:\ripfiles\new\output.dts" -1536

TrueHD, 5.1 channels, 48khz
Decoding with libav/ffmpeg...
Writing WAVs...
Creating file "E:\ripfiles\new\output.LFE.wav"...
Creating file "E:\ripfiles\new\output.C.wav"...
Creating file "E:\ripfiles\new\output.SR.wav"...
Creating file "E:\ripfiles\new\output.R.wav"...
Creating file "E:\ripfiles\new\output.L.wav"...
Creating file "E:\ripfiles\new\output.SL.wav"...
This audio track has a constant bit depth of 24 bits.
Encoding DTS <1536kbps> with Surcode...
Found Surcode DTS Encoder version 1.0.23.0.
Surcode says/asks: "At least one valid source file must be specified to encode."
.
Pressing the Surcode "Encode" button didn't seem to work...
Closing Surcode...

Is this because the constant bit depth is 24 bits and that it isnt supported ? Should I use the -16 to convert 24 bit to 16 bit wav ?

[nautilus7]

No this isn't the problem. Try changing the path to your output file to something sorter, like E:\output.dts

Surcode is known to behave strangely with long paths.

[lexor]

No this isn't the problem. Try changing the path to your output file to something sorter, like E:\output.dts

Surcode is known to behave strangely with long paths.
I doubt that's the issue either. I used much longer paths, with spaces and special characters to boot.

Are we talking about the latest eac3to here? An older version didn't work with surcode for a while.

[rica]

I think -16 might work.

[baudi]

(I think it is related with version number; if you have Surcode 1.0.21 you have to make a second pass.

This issue is present in Surcode DTS version 1.0.29 also.

Rerun eac3to is required (or select DTS Compact (*.cpt) as output and renaming as DTS).

Thank you madshi and rica.

[GZZ]

d:\eac3to>eac3to "E:\ripfiles\new\output.thd" "E:\ripfiles\new\output.dts" -1536 -16
TrueHD, 5.1 channels, 48khz
Decoding with libav/ffmpeg...
Writing WAVs...
Creating file "E:\ripfiles\new\output.L.wav"...
Creating file "E:\ripfiles\new\output.LFE.wav"...
Creating file "E:\ripfiles\new\output.C.wav"...
Creating file "E:\ripfiles\new\output.R.wav"...
Creating file "E:\ripfiles\new\output.SR.wav"...
Creating file "E:\ripfiles\new\output.SL.wav"...
This audio track has a constant bit depth of 24 bits.
Encoding DTS <1536kbps> with Surcode...
Found Surcode DTS Encoder version 1.0.23.0.
Surcode encoding successfully started. Please wait...
Closing Surcode...
eac3to processing took 33 minutes, 12 seconds.
Surcode encoding took 31 minutes, 50 seconds.
Done.

The -16 did the job, maybe this option should be forced in eac3to when encoding DTS using surcode, so it dosnt crash on surcode because it expect a 16 bit stream ?

[rica]

The -16 did the job, maybe this option should be forced in eac3to when encoding DTS using surcode, so it dosnt crash on surcode because it expect a 16 bit stream ?

No, i guess i made a mistake; the option must be -down16

Surcode seems to work but your wavs are still 24bit.

Take a look at:

C:\>eac3to\eac3to C:\Output\audio.dts C:\Output\audio_out.wavs -down16
DTS, 5.1 channels, 0:37:26, 24 bits, 768kbps, 48khz
Decoding with ArcSoft DTS Decoder...
Reducing depth from 24 to 16 bits...
Writing WAVs...
Loading white noise (needed for dithering)...
Creating file "C:\Output\audio_out.R.wav"...
Creating file "C:\Output\audio_out.L.wav"...
Creating file "C:\Output\audio_out.C.wav"...
Creating file "C:\Output\audio_out.SL.wav"...
Creating file "C:\Output\audio_out.LFE.wav"...
Creating file "C:\Output\audio_out.SR.wav"...
eac3to processing took 2 minutes, 25 seconds.
Done.

[nautilus7]

Wait a minute.......

Do you see any message telling that "reducing bitdepth to 16 bits"? No, because you didn't tell eac3to to do it.
-16 is not meant to be used here. It is only for setting pcm track parameters.
rica had -down16 in mind but he didn't think that he should check it first.
I told you that 24 bit source wan't the problem anyway, but you didn't listen.


Anyway, you run exactly the same command as before and it worked. That means that Surcode was acting up. That's why i proposed changing output paths in the first place.

Take care.


EDIT: rica was a bit faster in his last reply.

[rica]

EDIT: rica was a bit faster in his last reply.

Yes, i suppose we wrote at the same minutes :)



But btw, i found -down16 is working in wavs output (look my previous post) while dts to dts -down16 doesn't work.
Is this a bug?

eac3to v2.66
command line: eac3to\eac3to C:\Output\audio.dts C:\Output\audio_new.dts -down16
------------------------------------------------------------------------------
DTS, 5.1 channels, 0:37:26, 24 bits, 768kbps, 48khz
Creating file "C:\Output\audio_new.dts"...
eac3to processing took 4 seconds.
Done.

And the info:

C:\>eac3to\eac3to C:\Output\audio_new.dts
DTS, 5.1 channels, 0:37:26, 24 bits, 768kbps, 48khz

[GZZ]

I see, so surcode does like 24 bit wave files. But what did the -16 option then do ? Nothing ?


Another thing. I ran the eac3to Input.dts output.dts - which should fix the zero padding in the dts stream. Well it didnt fix anything the output dts stream is identical with the input dts stream, so I think this is fixed in version 1.0.23 of surcode.

[rica]

Nautilus7, my friend;

I see GZZ used the same output folder, didn't he:

E:\ripfiles\new\output.dts

Check posts 17 and 22 :)

That's why i proposed changing output paths in the first place.


EDIT: GZZ, can you please try again without -16? Do not change the output path.

[GZZ]

I trying to do a eac3to input.m2ts output.dts -1536 kbit to see if the output DTS file is identical to the one created from your command: "eac3to input.m2ts output.thd" and then "eac3to output.thd output.dts -1536. I think both way should result in the same DTS file, but the first command do both steps in one.

But I will try the eac3to input.thd output.dts -1536 (without the -16 option) and see if the output DTS also i identical to the one with the -16 command. If surcode crash without the -16 command then it dosnt like 24 bit wave files and then the -16 option did change the wav files (but eac3to still reported it as 24 bit).

[nautilus7]

But btw, i found -down16 is working in wavs output (look my previous post) while dts to dts -down16 doesn't work.
Is this a bug?DTS doesn't know what bitdepth is. What you see is nothing more than the bitdepth of the source pcm files that were used to make the dts. So you ask eac3to to reduce the bitdepth of what? It doesn't make sense.

But what did the -16 option then do ? Nothing?Yes nothing!

Another thing. I ran the eac3to Input.dts output.dts - which should fix the zero padding in the dts stream. Well it didnt fix anything the output dts stream is identical with the input dts stream, so I think this is fixed in version 1.0.23 of surcode.There's nothing to be fixed guys... There is an option in every dts encoder to output compact, padded or or dts in wav form. We want the first, but Surcode outputs the second by default. eac3to manages to force the encoder to output compact form.

That's very simple.

Nautilus7, my friend;

I see GZZ used the same output folder, didn't he:

Check posts 17 and 22 :)Anyway, you run exactly the same command as before and it worked. That means that Surcode was acting up. That's why i proposed changing output paths in the first place.
I bet if he runs the same (initial) command multiple times, some will work, some not. Changing path will make Surcode work everytime, probably.

[GZZ]

I bet if he runs the same (initial) command multiple times, some will work, some not. Changing path will make Surcode work everytime, probably.

You are properly right on this one. Will it help to maybe change the path to short path like C:\Program~1\.... the way old windows did it. But on the other hand. My path (E:\Ripfiles\new\output.dts) is really short, no path name over 8 chars, no spaces, no wierd chars, its hard to imagine that should be the error.

[nautilus7]

It's Surcode!!! You can't be sure!!! :D Really, i don't know.
I always use c:\output.dts to play safe.

[GZZ]

(Extracting AC3 TrueHD from m2ts)
eac3to Input.m2ts Output.dts -1536

and

(Extracting AC3 TrueHD from m2ts to THD)
eac3to Input.m2ts output.thd

(Convertering THD to DTS)
eac3to Output.thd Output.dts -1536

Will the result be the same or does the first only use the Core AC3 (not the HD) ?

To answer my own questions. The two ways to extract a AC3 TrueHD flie results in identical files, so doing the first script will give the same extracting to THD and then encode to DTS. :)

[rica]

OK, thanks for the feedback but how about the second trial?

GZZ, can you please try again without -16? Do not change the output path.

[GZZ]

havent done that yet. But from what I can see in the surcode log file it outputs. Then my first (Thd -> DTS) and second run (Second run was from the M2ts (Ac3 Truehd) -> DTS (without the -16 option).

The log first run (with -16 option)

DTS DVD:

Core Bit Rate = 1536000 bps
Frame Size = 512 samples
Pad zeros at end of frame= No
Channel Layout = 3/2/.1
Sampling Frequency = 48000 Hz
Bits Per Input Sample = 24
QMF Filter = Nonperfect reconstruction
Rear channel attenuation = 0 dB
HDCD = No
ES = No
Compressed audio file = E:\ripfiles\new\output.dts


Second run (without -16 option):

DTS DVD:

Core Bit Rate = 1536000 bps
Frame Size = 512 samples
Pad zeros at end of frame= No
Channel Layout = 3/2/.1
Sampling Frequency = 48000 Hz
Bits Per Input Sample = 24
QMF Filter = Nonperfect reconstruction
Rear channel attenuation = 0 dB
HDCD = No
ES = No
Compressed audio file = E:\Ripfiles\new\output-direct.dts


As you can see they had the same Bits Per Input Sample. So the -16 didnt do ANYTHING at all. Also did a binary compare of the 2 output DTS files and they are identical.

So the crash related bug had nothing to do with the -16 option. It properly just crashed, maybe its not surcode fault, but the method used to call surcode (from eac3to), maybe it should do a few retry before it cancel the conversion. Then you will be sure if its just surcode just acting wierd sometimes.

[rica]

No i didn't say there is an option like -16; i already told i made a mistake coz the right option must have been -down16.

But the question was, without -16, a second or third trial would change the Surcode auto-opening situation or not?
I didn't mean 24 will return to 16.

I think there is a misunderstanding.

[GZZ]

No i didn't say there is an option like -16; i already told i made a mistake coz the right option must have been -down16.

But the question was, without -16, a second or third trial would change the Surcode auto-opening situation or not?
I didn't mean 24 will return to 16.

I think there is a misunderstanding.

Sorry about that. Havent tested the surcode openening problem.

[LegendSeeker]

Hi friends, How TrueHD/AC3 audio to DTS audio conversion? source m2ts file.

[nautilus7]

:search:

hint: eac3to

[LegendSeeker]

:search:

hint: eac3to

my englih bad:) but;

yr_eac3to_gui ok.
my convertion m2ts file truehd/ac3 audio track -> .dts audio file. but converting new audio file .dts but audio info: "dts-hd" why?

[nautilus7]

I guess it was "media info" that told you your new track is dts-hd, right? It's simply wrong. Your new track is dts. Use eac3to (not the gui) to get the real info about your track.

[LegendSeeker]

I guess it was "media info" that told you your new track is dts-hd, right? It's simply wrong. Your new track is dts. Use eac3to (not the gui) to get the real info about your track.

ok thnx nautilus.

p.s.: source 2h 16min but converting new files 2h 14min, correct or incorrect?

[nautilus7]

Can't be sure.

Is your dts track in sync with the video? That's all it matters.