songwriter wrote:I need to understand once and for all about converters.
So I am more confused now than before. This engineer mixes and matches converters?? Why? What's the difference in sound and why would it have an effect on program material?
This is actually easier than it looks.
A converter has a job - polling your analog signal with a given frequency to transform it into a stream of samples with the desired word length.
If it were possible to build a perfect real-world converter, it would do that job and that one would be all that you even needed.
But it isn't. Real-world converters are less than perfect.
So different converters (based on different classes of tech or different designs in the same class) produce slight different errors. The sample streams they produce are (often very, very slightly) different.
That results into slightly different analog signal when the samples are used to reconstruct the (supposedly) original waveform. Aka "different sound".
Errors can come from timing (frequency must be ultra precise), how the gain-to-digital level is interpreted (ideally should be a straight line, but physical components may not behave so) and so on. Also boundary conditions like operating temperature (often linked to the voltage) can affect the signal detection and different designs will attempt to compensate for these physical effects in different ways (usually the better they are, the more costly, with the inevitable law of diminishing returns creeping in).
Keep in mind these effects are really minimal
- for example, the time scale with which music is understandable to us (say milliseconds for a note) is way bigger than the time scale with which a converter operates (even a humble 44.1Hz is 44100 times per second, an order of magnitude greater).
Nevertheless: these errors all result is slightly different (but musically equivalent) sample streams and reconstructed-analog-wave.
Some people claim to be able to hear the differences. How much that claim is true or self delusion, it's hard to say - nobody's ever organized a proper blind test insofar I know.
One funny thing to think about, is that the same micro-errors are sure to occur (for the same physical-vs-theoretical reasons) in D/A conversion.
So, are these fellows hearing the A/D conversion errors, or the D/A conversion errors? A mix? :-) Who knows. My $.10 is that we people are great at fooling ourselves, especially when we've used a hefty dosh of cash on something. But your opinion may differ.
Beyond that, exactly like you can't evaluate a microphone unless your room is pretty good (a fantastic mic will capture fantastically a horrible comb-filtered signal, if the room is very reflective, resulting in a horrible recording), evaluating a converter without having a really great chain before is nonsensical.
, if you have a great room, a really good fidelity microphone, a great (low distortion, low noise, as-linear-as-possible) preamp... you can hear what the converters errors do. Maybe. Dunno. I don't.
I also understood that the converter acts as the audio interface, like the fireface UFX. Is that right? Or can you have a different converters like the ones above integrated into your normal recording setup (in my case with the UFX) to change at will?
No - "interface" stands for "computer interface": that is, an interface is what takes the samples from the converter to the PC (via USB, thunderbolt etc). A stand alone converter usually doesn't "talk" these computer protocols, so usually need to take the (already digital) sample stream and send it into an interface. which then converts them to USB etc.
That's due (my guess!) to historical reasons, with protocols to transmit audio signals (like AES/EBU or S/PDIF) having evolved in parallel with device-to-computer communication protocols (like USB), and from slightly different viewpoints.
Bit like Nikon and Canon being both cameras, but needing adapters to use each other's lenses. Or Apple and PCs being both computers but.. you get the gist.
On the other side, all interfaces that expose analog inputs (line or mic) contain A/D converters. There are interfaces which are digital to digital (to translate between the various digital audio protocols, say AES to MADI etc) and then get the signal into a computer. These won't contain converters, but are far fewer and normally oriented to the pro market rather than the consumer.
Due to the theory-vs-real-world-effect above, at a given price point a design that deals only with conversion (i.e. an A/D box) vs. one that has to deal with preamp + conversion + translation to computer protocol (i.e. an interface) will likely do a slightly worse job at A/D. But in reality really good A/D conversion chips, supporting reasonably good analog front-ends, have become commonplace so it's not really something you should concerns yourself unless you can
(as opposite to want
) squeeze the last 0.0001% of performance from your chain.
Like it's not a big deal - for the purpose of driving fast - to worry about the micro-meter smoothness of a cars paint coat if you don't have a great pilot, engine, suspensions, injection, reliability etc..
Shameless plug: if you like, you can always have a look at my blog http://theaudioblog.org
where I write about these things.
Hope it helps!