These days the audio world is full of possibilities for anyone deep into digital music. Things became extremely sophisticated over the years. That's obviously a good thing to pick and choose from particularly in this genre. From an engineering perspective, the road to achieving the goal of converting signal from digital to analogue can be most varied. Practice shows how different approaches achieve various results.


Before we go further, there's an important thing to remember. The overall sound quality of any audio product depends heavily on the engineer's skill of connecting the various pieces of the puzzle. Here the human touch is crucial. What parts are used is important of course but a great cook can create a grand meal with literally nothing in the fridge. To some this is common knowledge but it had to be pointed out once again. That's because our hobby is filled with individuals who get deeply invested in a design based just on its topology. On many levels that even makes sense. For instance, if one is after a class A/B power amplifier to drive a pair of difficult speakers, it's very probable that a 300B SET becomes a no-go. Conversely someone in search of a low-power pure class A amp will discard heavyweight A/B brutes by design.


The same goes for speakers. Open baffle fanatics won't take small two-way bass-reflex affairs under consideration and vice versa. These scenarios are understandable. It's quite different with DACs though. Many people focus primarily on the conversion method, be it a single Sigma-Delta chip, several of these stacked, vintage R2R via PCM1704 or TDA1543, chipless, FPGA upsampling with a simple low-pass filter and so forth. But there's so much more to it like the execution of the power supply, digital receivers, current-to-voltage conversion, output stage etc. Each of these can be implemented in many different ways and is partly responsible for the final outcome. Many highly regarded engineers insist that the power supply is of greater importance than the choice of conversion silicon for example.


Moving right along, it's easy to see why enthusiasts with big coin avoid Sigma-Delta chips. These are so often used in very affordable products to not seem kosher enough. A narrow price frame in fact is the main limiting factor which seriously reduces an engineer's options. He can't go after a fancy power supply, FPGA code writing or tubes, all of which may lead to an outcome well below the spectacular but to the minds of many, ΣΔ chips are the only culprits. Odd, ain't it?