An introduction to computer audio
ASIO bypasses the normal audio path from the user application through layers of intermediary Windows operating system software, so that the application connects directly to the sound card hardware. Each layer that is bypassed means a reduction in latency, the delay between an application sending sound to the sound being reproduced by the sound card. In this way ASIO offers a relatively simple way of accessing multiple audio inputs and outputs independently. Its main strength lies in its method of bypassing the inherently high latency of operating system audio mixing kernels (KMixer), allowing direct, high speed communication with audio hardware. Unlike KMixer, an unmixed ASIO output is "bit identical", that is, the bits sent to the sound card are identical to those of the original WAV file, thus having higher audio fidelity.
Source: http://en.wikipedia.org/wiki/Audio_Stream_Input/Output
The audio stream transferred by the K-Mixer (mixing software implemented in the operating system's kernel) may not be bit-exact; it is often resampled because the K-Mixer can only process audio streams that have the same format (48kHz/16-bit). And the resampling process worsens the audio quality. The windows plugins like Volume, Balance etc.. are only working with 48kHz music signals. Even if the output is bit exact, the resampling algorithms may induce jitter resp. signal alterations.
Source: http://www.aqvox.de/Asio-USB-Audio-installation-e.htm
For the audiophile bypassing XP's KMixer is probably the biggest asset of
using a ASIO driver. Pick them carefully as there are differences in sound
quality between the brands.
In case of Vista, MAC and Linux the use of ASIO drivers
is not recommended by manufactures like Aqvox, Benchmark and Wavelength.
See also: XP vs Vista