Signal chain question for analog mastering from digital source

[snaysup]

I have a few questions about mastering with analog outboard gear. Where in the chain do you insert something like an AVOCET or GRACE 902? DAW->DAC->EQ->limiter->ADC->DAW2?
DO you use the avocet dac to get out of the source computer?
Also do the coverters need to be synched to a master clock?
Can source DAW be the destination DAW?

thanks, daniel m.

[bob katz]

Originally Posted by snaysup

I have a few questions about mastering with analog outboard gear. Where in the chain do you insert something like an AVOCET or GRACE 902? DAW->DAC->EQ->limiter->ADC->DAW2?
DO you use the avocet dac to get out of the source computer?
Also do the coverters need to be synched to a master clock?
Can source DAW be the destination DAW?

thanks, daniel m.

Unfortunately, the Avocet is not a (processing) console, just a monitor controller. I don't know if the Grace has processing inserts.

So you cannot use the Avocet DAC for this purpose. I use the Cranesong HEDD for analog inserting, as it now has an excellent DAC (with the upgrade) and always has had a superb A/D. But you can use any separate high quality D/A and A/D converter you like.

Here's a chain, just like the one you drew above:

DAW-----> DAC -----> chain of analog processors -----> ADC -----> [optional digital processors] ----> back to same DAW or a different DAW if working at different sample rate.

For the lowest jitter, use as jitter-immune a pair of converters as possible. In my chain, my ADC is most stable when run on internal clock, so I run the ADC as the master clock. As you can see from the above chain, the ADC can be the master clock even if it is not in the front of the chain.

I run the ADC at 96K, usually. I use a live sample rate converter (The Weiss) to take the end of the chain back down to 44.1K for capture to the CD rate.

The DAC is on external sync, since most DACs slave to the incoming (SPDIF or AES) signal, the DAC has to be as jitter-immune a model as possible. Any converter that's slaving to external has to be jitter-immune. It's possible to find a DAC model that has internal clock and lock the main DAW to it (I believe the Weiss has that feature, and also the DCS and the TC Finalizer) but with the quality of jitter-immune DACs these days, you can slave the DAC to the DAW and not hear any difference.

Unless you're one of the lunatic fringe who claim that "you can never make a CD that will ever sound like the master, and a clone of a hard disc sounds different than the source" in which case you'll never be satisfied.

[Darius van H]

Originally Posted by bob katz

I run the ADC at 96K, usually. I use a live sample rate converter (The Weiss) to take the end of the chain back down to 44.1K for capture to the CD rate.

Bob, out of interest, why do you capture at 96k then downsample to 44?......why not just capture at 44 (if the end SR is 44.1)?

[bob katz]

Originally Posted by Darius van H

Bob, out of interest, why do you capture at 96k then downsample to 44?......why not just capture at 44 (if the end SR is 44.1)?

It's subtle to many ears, but the bottom line is because it sounds better to me done that way. But objectively, many non-liinear digital processors perform, and sound better at 96K. My ears are very sensitive to alias products, and there are far fewer in-band alias products at 96K. Processors such as the Cranesong HEDD really sound better at 96K, digital compressors do as well. The Weiss units all internally double sample, but I feel that you minimize the tradeoff by upsampling at the beginning and downsampling at the end of the chain, which turns off the up/down sampler in each Weiss unit.

So, when analog processing, the DAC feeds the analog chain, and the A/D then runs at 96K, and that feeds the digital chain, so in this case the A/D is my upsampler.

[Riccardo]

Originally Posted by bob katz

It's subtle to many ears, but the bottom line is because it sounds better to me done that way. But objectively, many non-liinear digital processors perform, and sound better at 96K. My ears are very sensitive to alias products, and there are far fewer in-band alias products at 96K. Processors such as the Cranesong HEDD really sound better at 96K, digital compressors do as well. The Weiss units all internally double sample, but I feel that you minimize the tradeoff by upsampling at the beginning and downsampling at the end of the chain, which turns off the up/down sampler in each Weiss unit.

So, when analog processing, the DAC feeds the analog chain, and the A/D then runs at 96K, and that feeds the digital chain, so in this case the A/D is my upsampler.

I feel and do the same thing, maybe is just me but at 96 it seems that whatever I do with in the analog chain is better captured, and the L2 seems to sound better. Not sure about the D/A upsampling...