Almost 6 years ago (this makes me feel old) I wrote
this post about the difference between analog, bucket brigade-based delays and digital ones. Although in that post there are some embarrassing/outdated statements:
Consider that most delays are 8 bit, which is about as low as you can get in quality.
I have always considered the argument that analog delays sound better because of their use of bucket brigade chips to be a moot point. Finally, today, someone else repeated this argument back to me (albeit in a more formal way)! A better way of describing the difference between a bucket brigade-based circuit and a digital delay is that the bucket brigade is discrete time, but continuously-valued, while the digital circuit is discrete time and discrete valued.
Above is a Ross Stereo Delay, using the less-common Reticon delay chips... Tuesday Guts!Now, to tie that in to my argument, the only real difference between a bucket brigade and a digital chip performing a delay function is that the bucket brigade chip is probably sampled much more slowly (maybe 8khz at the slowest vs 44.1+ khz), the BBD introduces some distortion, noise and loss, and there is some kind of minor round-off error in the digital circuit. Now, as a result of those differences (really only the first two), an analog echo circuit gets a LOT of "wrapper" circuitry - in particular, some very heavy low-pass filtering, compression, and expansion. In the end, the only difference that isn't due to the wrapper circuitry are the BBD nonlinearities, but my point of my original post was that the nonlinearities of low fidelity digital conversion were comparable, if not just as musical. At any rate, it was nice to hear someone else discussing this difference!
-Colin
