[linux-audio-user] Re: Digital Fidelity

[Paul Davis]

On Wed, 2006-03-01 at 16:06 -0800, Maluvia wrote:
> Not to beat a dead dog, but . . . 
> 
> Before we blow a once-a-year tax refund on an Apogee converter, could
> someone - a compassionate scientist perhaps - explain to me why an Apogee
> converter is so vastly superior to the converter in, say, an RME hdsp9632 -
> as to justify it's second-mortgage price tag?
> (I know the best way to find out would be to just listen, but we can't
> afford to order one just to test it out.)
> 
> At least Zaolla has 6 pages of detailed engineering specs, data, tests,
> charts, etc. to back up their claims, whereas with Apogee - they just seem
> to expect you to take their word for it.

apogee used to have some very detailed tech stuff.

the reason their stuff is so good is not directly related to the
converters themselves, but to the sample clock that drives them. the
less jitter you have in the clock, the less likely it is that you will
see differences between the original and reconstructed waveform after
going through an A->D then D->A pass.

how come? imagine a perfect system with zero jitter in the clock. every
N nanoseconds, the converters generate a sample value for the A->D side;
every N nanoseconds, they accept a sample and feed it into the D->A
reconstruction process. so if you feed the digital signal back into the
D->A process, you will get an identical waveform. this isn't because
apogee is cool - this is basic digital audio theory.

now give that clock a little jitter, and it turns out that you end up
with the reconstruction phase having a slightly different sense of time
than the sampling phase. as a result, the waveform that is reconstructed
can differ (albeit in minute ways) from the original.

so if you can reduce the clock jitter, you can reduce this effect, and
apogee has basically the best clocks you can buy.

their gear is not worth the money, but it is better than just about
anything else. however, my compromise is to buy an apogee clock, when
possible. so for example, the mackie digital 8 bus that i own has
mackie's own burr-brown based converter circuitry, but it also has the
optional apogee clock card to provide word clock to the converters.

--p