Ame one Loves You

I been thinking about mp3 conversion lately. I don’t know exactly how it works. I know the general theory but I don’t know anything else. One of the questions I have always wondered, “Is it necessary to use bit reduction and convert a 24 bit song to 16bits before mp3 conversion?”

After thinking about it, I realized that it probably doesn’t matter. It would most likely depend on the encoder’s ability to deal with 24 bit files. Just going through the simple logic of it, any gains by using dither would be negligible due to the fact that the file is being converted into a lossy format. Which mean any distortion caused by truncation at the 16th bit, would probably not be noticible for two reasons. Reason 1 is that it would mostly likely be masked by the degradation caused through the encoding process. Reason 2, either way the effect of the dither will be un-noticible except in extreme circumstances of bit reduction or in certain listening situations.

My answer to the question of whether a 24bit wav should be converted to 16bits with dithered applied prior to conversion to an mp3 would be, no. Recently I had to make an mp3 out of one of my projects in Logic Studio. The project was a combination of audio and virtual instruments. I do everything in 24bit when I write music. I didn’t even bother to bounce it out and convert to 16, instead iIjust did a straight bounce to mp3 conversion that is in the bounce options. I did not notice any terrible “nasty distortion” from the truncation to 16 bit. All I heard was the usual lossy sound of a 128kbps mp3.

Dither really seems to be this mystery thing in audio that most people don’t quite have a grasp on. You can use math and plot graphs to explain and show what it does. Bob Katz does a wonderful example of explaining dither with math and graphs. Can’t really argue with the math behind it because the numbers are there to support what it does. Is this really enough evidence to support its use? Just because the math makes sense is it necessary? Apparently Steve Massey of Massey Plugins doesn’t think so. I recently came across an interesting post on his blog where he makes a great argument that dither “has no impact on the end product.”

In my other post about dither I recommended that you go to http://www.24-96.net/dither/ and take the blind dither test. Steve Massey looked at the fine print of the test and points out with math that what we are hearing in the test is equivalent of 7 bit audio and “that dither is highly effective for 7 or 8 bit audio.” The reason for this according to Massey is that:

They had to [use such bit reduced audio] or the test never would have worked. Quantization noise, which dither is meant to mask, is not humanly audible for typical, normalized, 16-bit audio, especially not in the everyday listening environments where we experience music. 16-bit audio has over 90dB of dynamic range — this is the difference between the maximum signal amplitude and the quantization noise. In other words, the quantization noise is 90dB below what you’re actually listening to….Expecting to hear the effect of dither is like expecting to hear the fluttering of a fly through the din of a jackhammer.

Massey is pointing out the the test is somewhat flawed because it is not using real world listening conditions, but relies on an artificial simulation in order to bring the noise of the dither forward enough so that it can be heard.

To illustrate his point that dither is not as big a deal as we think and whether we really need it, Massy points out something he discovered using his limiter plugin in conjunction with the TDM POW-R plugin in Pro Tools.

A while back I discovered that the TDM POW-r dithering plugin that’s provided with Pro Tools had a bug, making it incompatible with my L2007 limiter plugin. The POW-r plug-in would stop producing dither when activated on the same DSP as the L2007…. I discovered the bug on my own, more or less by accident, while using an FFT analysis plugin. But, guess how many users called to report it themselves? Zero. Just like myself, no one ever heard this on their own. Only with tweaky analysis tools is dither measurable.

In the real world environment no one was able to recognize that the dither plugin had failed to work in conjunction with the L2007 limiter plugin. This illustrates how the dither does not effect the final outcome of the product enough that it is noticeable in a real world listening environment and because it is supported by math, it has become a set it and forget plugin. We just accept that it does what it is supposed to do without questioning it.

In the end we (the people using it) know it is there and what it is theoretically doing. The people listening to dithered music don’t know that it is there or don’t care. After all, “the only people on the planet perceiving and discussing dither are those people who have been informed of its existence.” Only the audio engineers know what flavor of dither they are using. I think Massey really sums it up best when he says, “When have you ever listened to a CD and thought, ‘Wow! This is a really incredible album. But, damn! The dithering algorithm that the mastering engineer used is total crap! I can’t believe they used POW-r type 3 instead of type 2!’…Regardless of its questionable efficacy, no one can reasonably argue that dither has an impact on the experiential enjoyment of music.”

Is dither really necessary? People will argue both sides to that. On one side you have the math and the graphs and the sound tests supporting it. On the other side you have the real environment and experience where most people will not notice it. I believe that we just take it for granted and accept it for several reasons. Those are that it is mathematically proven. It has been shown to us and we have seen examples of how it works compared to other dithers and non-dithered material even though the tests do not reflect on the real world environment. In the end I feel that dither has simply become something that has been ingrained into us and has developed into a set it and forget type of plugin. In the end the only person who really knows what dither on what song is the engineer that worked on it.

You can view Steve Massey’s original post at:
http://www.smassey.com/blog.html#29_apr_2008

Which is the best dither? Truthfully I have no idea. There are several different types available. Before we tackle which is the best dither, let’s examine what dithering does.

In short dither is used during bit reduction i.e. 24bit to 16bit. What it essentially does is add noise shaping to smooth out the reduction. It is something that you use only once and it is placed at the very end when you are converting the bit rate. Because it causes degradation, do not use it more than once. That is dithering in a nutshell. For a more in depth description of what dither is and isn’t check out this Article by Bob Katz Explaining Dither.

There are several different types of dither algorithms that I always hear being mentioned when people talk about it. The most commons are POW-R, UV22 (Apogee), MBit + (Izotope), and the Waves dither found in the L2 limiters. I have used all but the Mbit +. I personally don’t have a preference. I try to use the one I feel is the best available in the program I’m using. If i’m using Cubase I use the UV22hr, if I am using Logic or ProTools I use the POW-R. You have probably heard a recording that has used eachone of these dither algorithms, except you just don’t know which one.

If you really want to know which is the best, don’t depend on me to tell you. What you should do is two things. First research the different dithers and how they work. Second try a blind dither test. I have done both of those. In the blind test I found that I prefered the Waves L2 dither.