Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[tombert]

Anybody an idea how to find out if a file is coded joint stereo or real stereo? In mp3tag I get it by clicking the file properties ...

thx
thomas

[MusicHawk]

Sorry, I don't know the term "joint stereo". Do you mean a file with 2 channels/tracks, but both are identical, AKA monaural?

Or do you mean 2 channels/tracks but the audio is so-called rechanneled where monaural audio is put into 2 tracks, then each is filtered to be somewhat different from the other? There are lots of ways this has been done over the years, usually with an unpleasant result. This is sometimes called fake stereo.

Either way, if there are 2 tracks/channels of audio, I don't believe there is anything in the file's properties that would indicate whether the tracks are identical or different. What info do you get in mp3tag?

Determining monaural vs. fake stereo is often easy by listening with headphones. Otherwise it requires something like an audio editor that can display both tracks for visual comparision, or analyze them. I've wished MC's small left/right channel spectrum display put the channels side-by-side for easy visual comparison, as a normal audio system would do.

Or are you referring to whether the file contains 2 tracks, or just 1? That information is visible in MC tag "Channels".

[rick.ca]

Sorry, I don't know the term "joint stereo".

For a short answer, see Wikipedia. It includes a reference to Joint Stereo: The Myths ... and The Realities which is an interesting read.

I know very little about it, but the encoders I've used in the past included a "mode" setting (i.e., stereo, joint, mono). I don't see the option in any of the MC encoders, but maybe it's available as a command line parameter for some. It seems I've encoded most of mine in joint stereo. I can't hear the difference, but I'm pleased that second article seems to say that was the right choice.

Anybody an idea how to find out if a file is coded joint stereo or real stereo?

The file property can be displayed in MP3Tag by defining a column with the value %_mode%. If you add the ENCODERSETTINGS field to the tagging panel, you can then fairly easily select all files with mode "Joint Stereo" and write this to ENCODERSETTINGS. This will be saved to MC's Encoding Settings field upon Update Library (from tags). Check that you're not overwriting anything already in that field you want to keep.

You would think there's a way to set ENCODERSETTINGS = %_mode% (or even a combination of such variables), but I don't see it. Of course, it would be even better if the encoder could be configured (a command line parameter?) to write this information to the tag.

[MusicHawk]

Thanks for the link. I'm a (former) broadcast engineer familiar with the stereo encoding techniques of FM radio (and AM and vinyl records), but I hadn't heard them described as joint stereo, though it makes sense to have a term that encompasses several techniques including digital audio.

I'm not aware that I have any joint-stereo MP3 files to check, so I'm wondering, is the file-size-reduction really worth the audio outcome?

[rick.ca]

The second link explains it better than I ever could. What I found interesting is that as a mathematical concept, it very simple and completely lossless. It's main purpose is to facilitate compression, the idea being the "difference" data of the "average" + "difference" representation can be compressed much more than "left" or "right." Now for the part I don't really understand, but it seems to make sense: All else being equal, this should allow more "room" for a compression scheme to retain more quality. The result should as good or better, not worse. The exception to the rule would be the rare case (for most music) where the left and right channels are completely different. Then the "difference" has as much information as the "average," and attempts to compress it too much could degrade it.

So the main thing I learned is it doesn't really have anything to do with a file size/quality trade-off. It provides the means for a more efficient compression. Depending on what encoder is used and how it's configured, the result might be better quality in a smaller file size. Also, I believe I read an encoder doesn't have to use one method—so we can expect a good one to recognize those situations where this won't work well and switch to L + R encoding.

Maybe one of our audio experts will comment on this. I've almost convinced myself joint stereo mode should always be used. Now the obvious question: What are the MC encoders configured to do? Should I be adding a command line parameter to enable joint stereo?

[tombert]

So MC itself is not able to read this %_mode% field?

I read lot of "opinions" that joint stereo is worse - and I'am playing with my pro-logic decoder - which gives me noise when playing joint stereo files ... so thats the reason why I asked ...

thx

[rick.ca]

So MC itself is not able to read this %_mode% field?

%_mode% is the MP3Tag variable for that file property, not a field. There doesn't appear to be a field/tag dedicated to Mode. Maybe that's because neither MC or MP3Tag developers think it's important or relevant.

I'am playing with my pro-logic decoder - which gives me noise when playing joint stereo files

Based on what I read today, I would think that must have something to do with how those files were encoded generally, rather than necessarily having anything to do with joint stereo. I suppose the obvious thing to do is get the mode information into MC, and then see if there really is a correlation between that and the noise you're hearing. But maybe you need to consider other variables as well.

[Alex B]

Joint Stereo - How to find out?

Select an MP3 file and open the Tag panel in Action Window. Click the topmost line to see detailed format information that comes directly from the selected file (alternatively you can click the Format field if it has been added to the visible tags in AW > Tag). The channel mode is one of the displayed technical details.

In general, the so called joint stereo mode in modern MP3 encoders is designed to improve quality. In lossless encoders the mode is completely lossless, but because MP3 is a lossy format the decoded output naturally differs if you have encoded the same file using two different settings. Most of the time the "intelligent" joint stereo mode produces less lossy results than the "stupid" forced stereo mode.

When MC's LAME encoder is used "joint stereo" is automatically enabled. You can set LAME to use other channel modes with a command line switch.

A couple of my related replies at hydrogenaudio.org:

http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=70847&view=findpost&p=624679
http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=54421&view=findpost&p=490936

[tombert]

Thx - there it is. So its missing only in a library field 

[rick.ca]

So its missing only in a library field.

If there is any point to doing so, you can import the information using the method I described.