L-> 16, R->8

Joint stereo encodes as the sum and difference of the two channels. In this case, the sum is 24 and the difference is 8.

S: 24 D: 8

You can find the original Left and Right values with these formulae: L = S – D and R = S – L.

L=24-8
L=16

R=24-L
R=24-(16)
R=8

Since we got back the original values exactly, and since we can do this regardless of what the values are (I dare you to try and find a case where this doesn’t work), joint stereo is lossless.

However, the computer must allocate enough space to store both values exactly or all that losslessness would be in vain. Assuming we are dealing with 16 bits per channel per sample integer audio (the most common type), to do this in all cases (including the worst case) we would need 17 bits for the sum channel and 16 bits for the difference channel. This means that instead of needing 32 bits per sample (2 channels*16) we would need 33. This means we would actually need MORE space to store joint stereo than plain stereo. On top of that, computers store bits in groups called bytes, and since bytes are almost always eight bits long, we would need 40 bits to store a sample of joint-stereo audio, since under most circumstances, computers don’t store bits by themselves (only in bytes).

However, this is where the magic of data compression kicks in. I’m not going to go into details, but a well implemented data compression algoritm can compress the joint-stereo data much better than individual channels, assuming there is much in common with the left and right channels. This is how FLAC works, even though it calls joint stereo “inter-channel decorrelation” and says joint stereo is lossy when they are really the same (They are probably trying to disassociate joint stereo with the additional sound degradation it can cause when using it with MP3 and other lossy compression algorithms).

The problem is that lossless compression can’t shrink audio enough for many purposes, like sharing music online (or sharing videos online with sound in them). This is where technologies like MP3 come in, though MP3 is a rather outdated technology (Vorbis is better, spec-wise at least). These are lossy compression algorithms, and they can also usually compress joint stereo better (by cutting out the less important data, especially in the difference channel), especially at lower bitrates. The problem here is that the lossiness can be more noticable when using joint stereo, especially at very low and high bitrates (as the article and other comments describe).

So if you can find the disk space/bandwidth to compress losslessy, you can (and should provided the left and right channels are similar; this is usually the case) go joint-stereo with no extra degradation.

If you’re already using a lower bit rate and then opt for full stereo, you will be instantly almost halving the available bit rate for the music, which will have a catastrophic effect on the overall quality at already sub-optimal bit rates.

I suspect a similar case exists here, namely that of discounting certain characteristics of program content in favour of others. Now, I do not pretend to contest the claim of a better frequency response being obtained by employing M/S stereo in favour of independantly encoded left and right channels provided the original source content does not contain grossly out of phase spatial signals, say improperly adjusted azimuth alignment of master tapes or deliberately out of phase signals reproduced at the recording stage. It can be argued (and indeed has been successfully argued in the past) that the lions share of recorded performances are not subjected to the grosser spatial anomalies and, while undoubtedly true, it does however beg the question “but what of the more minor ones?”

Think for a moment of your standard FM radio broadcast. A decent enough signal provides for (in most cases) an acceptable reproduction of the content provided at the studio. However, the further you travel from the transmitter site, the more noise is introduced due to the nature of its M/S component. FM radio encodes its sum channel between 50Hz and 16KHz typically, the difference channel being delivered above the 19KHz pilot tone. As the carrier signal drops in intensity, the bandwidth of the signal drops too, producing what is commonly referred to as multiplex hiss.

Content in mp3s are likewise going to suffer some form of degradation. We are in essence limiting the amount of bandwidth that data (in this case digital) might be used to store this content in much the same way that distance from an FM transmitter limits the (analogue) data our FM receivers might comfortably reconstruct. In the case of mp3′s however, it is not the noise floor which suffers adversely as much as it is frequency response and (in the case of M/S Joint stereo) spatial seperation.

Its always going to be a trade-off between how much loss and of what type of loss the user is prepared to accept when ripping their collections to mp3. M/S joint stereo (as I have learned through experience with a variety of codecs) tends to sacrifice spatial information markedly more so than independantly coded stereo does. This is to be expected as (and do correct me if I am wrong here) much of the data stream is devoted to the sum channel at the expense of the difference channel. This can often result in a muddying of the stereo image, particularly at lower bitrates (I never use intensity stereo for that very reason) and this becomes all the more apparent as the original stereo image widens. And as I undestand it, if the stereo image is widened sufficiently (beyond 120 degrees or so) this can also affect the overall performance of the mid channel as well.

I guess the point needs to be made that there is no solution which is always going to produce a more accurate result than the other. I would estimate that most users with no prejudice towards simple or M/S stereo encoding would be content to set and forget their encoders to M/S joint stereo for the simple reason that much of today’s content (digitally recorded or otherwise) tends to be listened to in less than ideal situations negating the concerns for a true spatial image, the trade-off resulting in more data space for more mp3s.

However, if space is not as much an issue as extracting the “best bang for your byte” then you are probably best served auditioning the results of both methods and carefully as I tend to do for the more critical content. As to what the individual might consider critical, I shall leave that for you to decide, though if you are in the process of remastering ye olde cassette collection I’d recommend giving J/S as wide a berth as is humanly possible.

As with most things in life – there is no such thing as a free lunch. If you want good quality music, you need to have high bit rate. The best codec at the moment is Flac as it’s lossless. However, the bit rate is around 800 kbps on typical music.