First of all, MPEG Layer III is not just one simple method of encoding, but a wide range of methods and sub-methods that can vary from frame to frame.

Especially, Layer III uses an iterative process for finding the most efficient coding of the audio samples. Most MP3 compressors are using an "HQ switch".

•Without the "-hq" switch, the encoder stops iterating after a certain number of iteration loops and uses the best encoding found so far. •With the "-hq" switch set, it iterates until no more improvements are found (= highest quality).

For example, if you watch the frame counter for L3ENC, you will notice that some frames encode very quickly, while it will "think" about others for a long time.

The difference is often very subtle. You may not tell a difference right away, but later (maybe after listening to the CD) you notice that the mp3 isn't quite as good.

There are generally two differences.

The first is what I call "warble" (it is a very poor description). You will hear this both with and without "hq", but "hq" improves it. It happens when the sound is very "busy", meaning a lot of different sounds are playing, covering the full acoustic range. You will hear the high frequencies quickly coming and going (hence the name "warble").

Also the singers voice will be very muddy. Almost like a radio station that is not quite tuned right. This happens because the encoder simply runs out of bits. It is a sign that the bitrate is not high enough. If you encode something at a low bitrate, you will hear very gross versions of this. At 128kbps, it only happens now and then. The hq will do a better job of getting the most out of the bits available, but there are only so many to go around.

The second is what I call lack of sharpness or crispness. It is usually much harder to point out, and is more of a general sense of the quality of the recording. It is related to transient response. It just seems that without the hq switch, just aren't quite as "live".

Incidentally, it should be compared to the CD (or wave), not to the non-hq version. Usually when there is a difference, both will be off from the original, but the hq will handle it better.

First, the simple answer.

If you want the best quality sound, use the HQ mode. Pure and simple. While there is often no real difference, "hq" will always sound as good or better than anything else.

Second, is it worth the extra time ?

If you just run it in the background or leave the computer running while you do something else, the time is not that important. It really is better, but not "night and day" better. Sometimes it is just a matter of bad and worse.  

If a link on this page doesn't work, let me know It doesn't fuckin'work!

All this web page is designed by Dave Guile © 1998