Multi bit converters encode analog information into a parallel stream of bits in PCM. The more the bits the more the quantization levels and so the better the accuracy of conversion. However the number of bits in a DAC is a poor method of determining its performance and accuracy. A better measure of performance are the bits themselves. Ideally a 16-bit DAC converts all 16 bits in a linear fashion. However in real practice a 16-bit DAC seldom achieves such a mission. The error in a 16-bit (or any multi-bit) conversion lies in the accuracy of the most significant bit (MSB) of the data word. If it is of the wrong level then half of the amplitude of the reproduced signal is wrong. Having also in mind that audio technology requires very high-speed conversion, more and more problems and errors are introduced.