The Preferences settings panel is reached through the Options menu.

The Advanced section is intended to optimise AE for use with faster or slower computers. The first two controls set how often the LED displays are updated. On a slow computer, set a slow speed here to use less of the processor's time.

The Sound Card Delay Correction is used for matching the timing of a recorded sample with the samples already on the Timeline. It is common to play back the existing tracks while recording a new track, and without this adjustment they may be slightly out of sync.

To measure the delay from the sound card, connect the line input directly to the line output. Load a sample with easily recognisable peaks, such as a percussion loop. Turn the monitor level well down, and make a short recording.

Now find the exact time of some peak in the first sample. To do this, drag the T marker so that it lines up exactly with the point to be measured - while you are dragging it, there will be a guide line over the sample. Then double-click on the green marker flag to see the time in milliseconds. Now measure the time of the same peak in the newly recorded copy sample. It will probably be 10-60 millisecs later. That is the figure to enter in the Prefs.

Audio Evolution uses a series of buffers to store data in memory. The samples are kept on the hard drive, and as playback proceeds through the song, parts of each sample are loaded into RAM for mixing and effects calculations. Loading from disk takes time. The big problem in a program of this kind is to do everything that is needed fast enough to keep up with the music. A final output value including mixing and all effects has to be calculated in 1/44100 second.

The buffering layers are:

1.) Big disk buffer (from 8K to 64K) : reading small pieces of data each time is very inefficient (due to the overhead of seek times).

2.) Middle layer for effects calculation. This is adjustable from 1/50 to 1/2 second. 1/50th of a second (see 3) can cause a high CPU load, but can be used if you have CPU cycles to burn and not many tracks or effects. The advantage of using a smaller buffer is that the real time controls respond faster. With a large buffer there may be a noticeable delay before anything happens.

3.) AHI layer which uses a buffer of 1/50th of a sec. for smooth volume and panning changes. This is not adjustable.

The Mixer section of Preferences controls some CPU-intensive display features: the LEDs next to the level control sliders in the channels and the Master, and the movement of the controls showing the automation. You can turn these off if your processor is reaching its limits.

The Timeline section has the option to not create a PCG file automatically for big audio files. On a slow system, this can take a while. Generally, this option should only be enabled if you have a very slow Amiga.