At which bitrate does Dolby Digital (AC3) achieve transparency? - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 11 Old 11-23-2012, 03:39 AM - Thread Starter
Member
 
clancy688's Avatar
 
Join Date: Jul 2012
Location: Germany
Posts: 52
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Hello,

a simple question: At which bitrates do AC3 encoded audio files achieve transparency compared to uncompressed samples?

Why I'm asking this: There's German BD publisher who apparently likes to put audio tracks in DD 2.0 (48 kHz / 320 kbit/s) on the disc, even though there are uncompressed tracks available (International releases of the same films feature DTS-HDMA or in most cases LPCM 2.0 (48 kHz / 16 bit)).

Personally I've the impression that DD encoded audio comes not even close to LPCM, DTS-HDMA or TrueHD (the sound seems to be... very shallow, while lossless formats seem to provide far more depth), but I could be imagining things... so, are there any impartial studies regarding this...?

Thanks for all answers!
clancy688 is offline  
Sponsored Links
Advertisement
 
post #2 of 11 Old 11-23-2012, 08:25 AM
AVS Addicted Member
 
ccotenj's Avatar
 
Join Date: Mar 2005
Location: the toxic waste dumps of new jersey
Posts: 21,915
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 89
^^^

you are imagining things... smile.gif

the difference is subtle at best, and then only noticeable on a very well setup system/room... in the great majority of home environments (read: almost all), those subtle differences will be inaudible, as they'll be swallowed up by other issues...

- chris

 

my build thread - updated 8-20-12 - new seating installed and projector isolation solution

 

http://www.avsforum.com/t/1332917/ccotenj-finally-gets-a-projector

ccotenj is offline  
post #3 of 11 Old 11-23-2012, 12:41 PM
AVS Special Member
 
MarkHotchkiss's Avatar
 
Join Date: Mar 2007
Location: Long Beach, California
Posts: 1,476
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 82
What might be clearly audible is if the AC3 track was placed on the disc at a lower volume than the other tracks. That might explain the symptoms you hear.

Also, your AVR might not decode the AC3 to the same volume level.
MarkHotchkiss is offline  
post #4 of 11 Old 11-23-2012, 08:43 PM
Senior Member
 
Joe Skubinski's Avatar
 
Join Date: May 2009
Location: Buffalo, NY
Posts: 474
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 19
Should also include the fact that different D/A schemes exist so one may prefer a particular format over another simply due to variations in how the receiver (or whatever the playback device) converts to analog.

Be the sage.
Joe Skubinski is offline  
post #5 of 11 Old 11-23-2012, 09:37 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,849
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 564 Post(s)
Liked: 342
Quote:
Originally Posted by clancy688 View Post

Hello,
a simple question: At which bitrates do AC3 encoded audio files achieve transparency compared to uncompressed samples?
Why I'm asking this: There's German BD publisher who apparently likes to put audio tracks in DD 2.0 (48 kHz / 320 kbit/s) on the disc, even though there are uncompressed tracks available (International releases of the same films feature DTS-HDMA or in most cases LPCM 2.0 (48 kHz / 16 bit)).
Personally I've the impression that DD encoded audio comes not even close to LPCM, DTS-HDMA or TrueHD (the sound seems to be... very shallow, while lossless formats seem to provide far more depth), but I could be imagining things... so, are there any impartial studies regarding this...?
Thanks for all answers!
I have not seen any published data on AC-3 in *stereo* at 320 kbps. Here is a double blind study of it compared to other codecs at 192 kbps and lower (at 44.1Khz which is somewhat easier than 48 Khz):

i-Xr9dcsq.png

You can see that at it does poorly at lower data rates but starts to improve a lot once you reach 192 kbps. You can also see that it starts to asymptote. Same is true of other codecs once you reach 320 kbps. The trick then is to guess at the slope of that graph and interpolating smile.gif.

Double blind tests of AAC at 320 kbps achieves near transparency. On some clips it gets 100%, on others it is a hair short. Assuming AC-3 underpforms it by some, I think it is fair to say that it doesn't achieve transparency in all cases.

Here is a suggestion and homework assignment smile.gif. Take itunes or other favorite AAC encoder and rip some music in it at 320 kbps and see if you can tell the difference from the source. Try this across multiple clips. Listen especially to high-frequency transients. If you can tell the difference here, then you certainly can in AC-3 also. If you can hunt down the MPEG reference clips, ,the job would be even simpler (I don't recall the public link I found for them just now).

To show you the dependency on clip selection, here is AC-3 performance at 384 kbps but in 5.1:

i-zTNwbzR.png

You can see how performance is variable. Again, keep in mind that this is for 5.1 and not stereo which is harder case although not nearly 3X so.

All of this said, the choice they are making for BD is really poor. There is no reason to get into this debate. They should be using lossless and be done with it.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #6 of 11 Old 11-24-2012, 12:34 AM
AVS Special Member
 
Bigus's Avatar
 
Join Date: Feb 2003
Location: The South
Posts: 4,258
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 50
Quote:
Originally Posted by Joe Skubinski View Post

Should also include the fact that different D/A schemes exist so one may prefer a particular format over another simply due to variations in how the receiver (or whatever the playback device) converts to analog.
That's like stopping to worry about the shape of a snowflake if you're caught in an avalanche. smile.gif

Bigus is offline  
post #7 of 11 Old 11-25-2012, 06:33 AM - Thread Starter
Member
 
clancy688's Avatar
 
Join Date: Jul 2012
Location: Germany
Posts: 52
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
First of all, thanks for all answers, especially to amirm! Your post was most helpful. smile.gif
Quote:
Originally Posted by amirm View Post

I have not seen any published data on AC-3 in *stereo* at 320 kbps. Here is a double blind study of it compared to other codecs at 192 kbps and lower (at 44.1Khz which is somewhat easier than 48 Khz):
[You can see that at it does poorly at lower data rates but starts to improve a lot once you reach 192 kbps. You can also see that it starts to asymptote. Same is true of other codecs once you reach 320 kbps. The trick then is to guess at the slope of that graph and interpolating smile.gif.

Is this simply *one* channel? Therefore 2 * 160 kbit/s AC3 = 320 kbit/s would be right on the threshold of having a perceptible difference to the source?
Quote:
Double blind tests of AAC at 320 kbps achieves near transparency. On some clips it gets 100%, on others it is a hair short. Assuming AC-3 underpforms it by some, I think it is fair to say that it doesn't achieve transparency in all cases.

Those 320 kbps clips are stereo, I assume?
Quote:
Here is a suggestion and homework assignment smile.gif. Take itunes or other favorite AAC encoder and rip some music in it at 320 kbps and see if you can tell the difference from the source. Try this across multiple clips. Listen especially to high-frequency transients. If you can tell the difference here, then you certainly can in AC-3 also. If you can hunt down the MPEG reference clips, ,the job would be even simpler (I don't recall the public link I found for them just now).

Thanks, I think I'll try this. But it may take some time as I've not much spare time available right now. But I know someone researching audio coding at Fraunhofer IIS (they created MP3 and were involved in AAC), so I hope I can get those MPEG reference clips you're mentioning.
clancy688 is offline  
post #8 of 11 Old 11-25-2012, 06:53 AM
AVS Special Member
 
Theresa's Avatar
 
Join Date: Nov 2010
Location: Minnesota
Posts: 2,430
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 50
I've often perceived that DTS is less harsh than AC3. I attribute this, perhaps wrongly, to DTS usually having twice the bitrate of AC3.
Theresa is offline  
post #9 of 11 Old 11-25-2012, 09:42 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,849
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 564 Post(s)
Liked: 342
Quote:
Originally Posted by clancy688 View Post

First of all, thanks for all answers, especially to amirm! Your post was most helpful. smile.gif
You are welcome. It used to be what I did for a living (managing development of audio technology including lossy and lossless compression).
Quote:
Is this simply *one* channel? Therefore 2 * 160 kbit/s AC3 = 320 kbit/s would be right on the threshold of having a perceptible difference to the source?
No. AC-3 like most other lossy codecs uses two methods to save bits in stereo/multi-channel encodings;

1. Shared buffering. There is a single buffer pool (data rate to be used) that each channel dips into. Think of it as having a family meal at a restaurant with a party of 2 or more. If one channel is more difficult to encode than the other, it will use more bits than the other. This pays huge dividends in multi-channel coding where most of the channels are idle or playing at low levels. Think of the center channel carrying most of the audio signal in a movie track. AC-3 will then allocate most of its bandwidth to that channel, resulting in far higher bit rate and much reduced distortion than as if it statically allocated bandwidth to each channel.

2. Channel coupling (also called Joint Stereo in other codecs). High frequencies, especially those that have transients, are very difficult for lossy codecs to encode. Depending on bit rate and number of channels, AC-3 encodes frequencies above certain level (theoretically 3.5 KHz but practically 10 Khz) in mono and then sends a signal to the decoder to play these at the levels of the original signal. The idea being that your ear is more sensitive to the envelop of the sound (level) more than its spectrum make up. So if you make sure that you get the relative levels right, the fact that you have a mono signal won't be noticed. Or at least, it won't be noticed as much as compression artifacts. Needless to say, you can't pull this trick for mono signals.
Quote:
Those 320 kbps clips are stereo, I assume?
The test was 384 kbps and not 320. All tracks were encoded as 5.1. However, some of them used to be stereo and were converted to multi-channel using signal processing.
Quote:
Thanks, I think I'll try this. But it may take some time as I've not much spare time available right now. But I know someone researching audio coding at Fraunhofer IIS (they created MP3 and were involved in AAC), so I hope I can get those MPEG reference clips you're mentioning.
The names of them were in the 384 kbps test graph I showed. But yes, your FHG contact most definitely has it. It is like a hammer to a carpenter smile.gif. He may also have more direct info on this topic. I used o have an AC-3 encoder years ago and did a bunch of tests with it but no longer do these days.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
post #10 of 11 Old 11-25-2012, 10:22 AM - Thread Starter
Member
 
clancy688's Avatar
 
Join Date: Jul 2012
Location: Germany
Posts: 52
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by amirm View Post

You are welcome. It used to be what I did for a living (managing development of audio technology including lossy and lossless compression).
No. AC-3 like most other lossy codecs uses two methods to save bits in stereo/multi-channel encodings;
1. Shared buffering. There is a single buffer pool (data rate to be used) that each channel dips into. Think of it as having a family meal at a restaurant with a party of 2 or more. If one channel is more difficult to encode than the other, it will use more bits than the other. This pays huge dividends in multi-channel coding where most of the channels are idle or playing at low levels. Think of the center channel carrying most of the audio signal in a movie track. AC-3 will then allocate most of its bandwidth to that channel, resulting in far higher bit rate and much reduced distortion than as if it statically allocated bandwidth to each channel.

Is this perhaps the so-called "Bit-Reservoir" technique...?
Quote:
The names of them were in the 384 kbps test graph I showed. But yes, your FHG contact most definitely has it. It is like a hammer to a carpenter smile.gif. He may also have more direct info on this topic.

Well, he used to give an Audio Coding seminar at my college and took us students on a trip to Fraunhofer IIS. I can remember him doing coding demonstrations with MPEG-samples. Speaking of which, there were two guys holding this seminar, the other one working for Dolby Nuremberg... ^^
Guess I could've asked them directly, but I figured that I don't want to bother them with asking for information which may be trivial and easily to obtain through the internet. Moreover, both are (of course) not unbiased, so I wanted to ask for a broader opinion first.
clancy688 is offline  
post #11 of 11 Old 11-25-2012, 11:06 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 17,849
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 564 Post(s)
Liked: 342
Quote:
Originally Posted by clancy688 View Post

Is this perhaps the so-called "Bit-Reservoir" technique...?
Yes. That would be a name FHG would especially use since I think they have a patent related to it smile.gif.

Amir
Founder, Madrona Digital
"Insist on Quality Engineering"

amirm is online now  
Reply Audio theory, Setup and Chat

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off