Limiting is an extreme approach to compression. Where compression reduces the degree by which sounds can go louder than the threshold, limiting is designed to stop sounds from being any louder than the threshold at all. Limiters usually have simpler controls to compressors, but are functionally similar to compressors with high ratio and fast attack.
Limiters are useful for reducing peak level (the level that machines “hear”) of a sound without affecting the average level (the level that humans hear). This reduces the headroom that the sound needs so that it can be made louder without distorting.
The downside the using limiters is that they reduce the level in the same way that compressors do – by applying gain. That is, they turn down the volume. Ideally, a limiter does this fast enough that the transient peaks are reduced but the steady-state sound is not audibly affected. In many cases though, the gain reduction is audible. It can give the sound a soft, wooly character, or even a random tremolo effect if the threshold is too low. Often using a limiter reduces the power and impact of a sound. This effect is sometimes hidden because the limiter also increases the overall volume, making it more difficult to notice that the character of the sound is changed.
An alternative to limiting is clipping. Yes, this is the same clipping that usually engineers try to avoid when recording and processing audio. Most of the time clipping is undesirable, but in some situations it can be used deliberately and beneficially.
While limiting reduces the level of the transient peaks by turning down the volume (applying negative gain), clipping reduces the level of transient peaks by distorting them. If you view the waveform of a clipped sound, you’ll see that the waveform looks like it’s had the tops and bottoms chopped off. This is distortion! However, if only the transient peaks are clipped, the clipping only occurs for a very short period of time, and is not very noticeable.
When this happens, the excess level in the transient peaks is transformed into upper harmonics. That is, the transients become noisier and dirtier. For some kinds of music, this can be a desirable alternative to reducing gain. The power and impact of the sound is often retained (or even enhanced!), but at the expense of fidelity.