Skip to main content

Topic: When converting to/encoding MP3, which is best: stereo or joint stereo (Read 17791 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • firework
  • [*]
When converting to/encoding MP3, which is best: stereo or joint stereo
I use a piece of software called Switch Sound File Converter, usually for coverting various files to mp3. Within 'Encoder Options' it has choices for 'Channel Encoding Mode' which are: stereo, joint, force and mono. It also has a tick box for 'Include CRC to Detect Errors'.

My question is in two parts:

(i) which is usually best, stero or joint?

and

(ii) what is 'CRC to Detect Errors?'

Thank you in advance.


  • tpijag
  • [*][*][*][*][*]
When converting to/encoding MP3, which is best: stereo or joint stereo
Reply #1
(i) joint
(ii)  Used search box for you
  • Last Edit: 06 September, 2011, 06:15:06 AM by tpijag

  • firework
  • [*]
When converting to/encoding MP3, which is best: stereo or joint stereo
Reply #2
(i) joint
(ii)  Used search box for you


Thank you for both. It appears CRC only detects errors. Can I ask why joint? Thanks again.

  • Northpack
  • [*][*][*][*]
When converting to/encoding MP3, which is best: stereo or joint stereo
Reply #3
Because it is generally agreed upon and verified through listening tests that LAME's joint stereo mode works flawlessly and therefore generally offers better quality than dual channel stereo, because the encoder can spend the available bits more efficently.

In many frontends, channel mode isn't even an option, because joint stereo is the standard mode and ideal for almost 100% of all material.

A side note: I wanted to link the joint stereo wiki article, but I'm a bit puzzled that it contains information that will be rather misleading for novice users.
Especially this sentence...
Quote
Very few MP3 encoders make ideal decisions about what mode to use from frame to frame in joint stereo files, so there has long been a misconception that joint stereo means too many (perhaps 100%) mid-side frames, which is rarely desirable in stereo music and can result in haphazard channel separation and/or suboptimal quality as measured by other criteria.

...could easily give the impression that joint stereo is something better to be avoided. Of course with recent generation encoders it is only true from an strictly theoretical perspective, not from a practical. It definitely should be changed. Ideas?
  • Last Edit: 06 September, 2011, 07:07:49 AM by Northpack

  • mjb2006
  • [*][*][*][*][*]
When converting to/encoding MP3, which is best: stereo or joint stereo
Reply #4
Simple enough: I changed "Very few MP3 encoders" to "Some early MP3 encoders didn't" and appended "This is not an issue with modern encoders." I also moved the text out of the lead and used it to replace some redundant info at the end of the article.
  • Last Edit: 07 September, 2011, 03:22:41 AM by mjb2006

  • Northpack
  • [*][*][*][*]
When converting to/encoding MP3, which is best: stereo or joint stereo
Reply #5
Simple enough: I changed "Very few MP3 encoders" to "Some early MP3 encoders didn't" and appended "This is not an issue with modern encoders." I also moved the text out of the lead and used it to replace some redundant info at the end of the article.

I guess that should avoid any misconception, thanks.

When converting to/encoding MP3, which is best: stereo or joint stereo
Reply #6
I don't get it. Why is mid/side stereo something bad? I know that the side channel is encoded with lower precision than the middle channel, but M/S stereo can be perfectly fine (and even better than simple stereo) for stereo music when the difference between channels is not that huge. Having that sentence in the wiki can be confusing for people who use tools like EncSpot and notice that 50% of their file or more is made of M/S stereo frames.

  • mjb2006
  • [*][*][*][*][*]
When converting to/encoding MP3, which is best: stereo or joint stereo
Reply #7
I was trying to speak in vague terms because I don't want to name the encoders (though the once-popular Radium FhG one comes to mind) or assume too much about their internals. But basically the situation with these encoders was that if you chose joint stereo, and your music was "too stereo", and the bitrate wasn't high enough, it was a pretty safe bet that the MP3 would have less stereo separation and more noise than if you chose regular/simple stereo, and so many people got the impression that the abundance of M/S frames or joint stereo itself was to blame, when really it was just an optimization issue in the encoder.

However, you're right, the ratio of M/S frames isn't the issue, per se, it's the starving of the side channel, so I've adjusted the text further. It now reads:

Quote
Some early MP3 encoders didn't make ideal decisions about what mode to use from frame to frame in joint stereo files, or how much bandwidth to allocate to encoding the side channel. This led to a widespread but mistaken belief that an abundance of M/S frames, or the use of joint stereo in general, always negatively impacts channel separation and other measures of audio quality. This is not an issue with modern encoders. Modern, optimized encoders will switch between mid-side coding or simple stereo coding as necessary, depending on the correlation between the left and right channels, and will allocate channel bandwidth appropriately to ensure the best mode is used for each frame.