[Mixchump]

First of all, let me say that I'm thrilled that you're making music at 192kHz, and for those of you that have just simply said 'work at 96kHz', you don't know what you're missing. If we could do 384kHz in ProTools, even better, but for now, 192k is great. Why? Because it's not about the top end. At all. It's about the detail in everything in the time domain. Room Mics especially sound awesome, and the grainy, sandy quality that crash cymbals have at lower sample rates (and especially the 'S's in vocals) are MUCH smoother. So let's dispense with that argument right away.

I track a lot of music at 192k, and on a practical level, here's my take on it. The DSP mixer on the HDX card(s) uses a LOT of DSP for the physical outputs because each output is essentially a summing point. If you look at the IO setup, you'll see that every output is an internal bus, connected to a physical output. I’m sure there are reasons why they don’t allow us to connect a single mono track to a single physical output, but the under-the-hood design of Avid’s HDX card means that a session with 28 physical outputs also has 28 summing busses using up DSP.

The gist of it is this. Unless and until some other DAW developer can go head-to-head with the ProTools application and it’s integration with DSP and hardware at near-zero latency, we’re stuck with it. In order for me to do a tracking session at 192kHz with a basic session (Drums, Elec Bass, Elec Gtr, Piano, guide vocal) I absolutely HAVE to have a minimum of two HDX cards. The combination of creating headphone mixes, and just a couple of basic reverbs is already a DSP-maxed situation.

I highly encourage you to borrow or rent a second HDX card and continue on with your plan. I also own an HD-Native PCIe card, and I can guarantee you that it will be even more frustrating (for different reasons), and the other options (UAD Apollo, Orion, Apogee, etc.) will all be equally (if not more) frustrating in Native world…

Once you get to mix world, your options will open up substantially.

I chose to go with Merging’s HORUS interface and I use HDX-2 for tracking with a DigiLink connection, and then for mix, I switch to the Ravenna/AES67 playback engine (all native CPU with Ethernet Audio IO) so that I can access the horsepower of my Intel CPUs and I don’t get hammered by losses of voices every time I drop a Native plugin on an Aux Channel or Master Fader…. And the Native mix engine sounds MUCH better than HDX does.

Anyhow, I feel your pain, but you’re one HDX card away from being extremely happy with your setup…

Cheers, eh?