Non-consumer (Pro/Live) approaches to DSP?

[Devnull]

Volume control is found all the time in pro/installed setups, commonly seen as POE wall controllers. You can do everything you want in a pro sound box with the exception of Dirac. That shouldn't be a problem because of MSO, REW and rePhase.


I currently use some BSS hardware. Literally everything is exposed on the Ethernet and RS232 interfaces, Input selection, volume control, eq, presets etc.... To control it you just have to send it a correctly formatted message. I know BSS has a white paper on integration with crestron controllers. If you're willing to write some code it shouldn't be that hard to write an ap yourself that runs on iOS or Android so you always have your "controller"

 

[AwesomeSauce2015]

Answering your original post:
I do not have much knowledge of the typical consumer DSP solutions. From my understanding, the typical way of accomplishing this would be with separate DSP units for the speakers and your subs (Say a MiniDSP 4x10 for the speakers, and a 2x4 for the subs), with some kind of preamp upstream for input switching and volume control.
Therefore, I will speak to my knowledge, which is the Pro Audio DSP world, or at least the installation-focused DSPs.

- It seems like you are looking for a system which can handle DSP processing, with 12 speaker outputs, and I presume a number of digital or analog inputs.
So that is our I/O requirement, and the limiting factor (as far as system hardware options) here is going to be the 12 outputs. If you don't already have amplifiers, you could theoretically use a digital protocol like Dante, Cobranet, or AVB, to send the audio to the amplifier(s), but these protocols do not usually run at > 48khz sample rates, and can introduce latency, along with network requirements (For example, a 5-port AVB-capable switch is like $400, Cobranet is obsolete at this point but is cheap, and Dante DSPs and amplifiers are crazy expensive).

- A good analog section for the analog I/O. This is another limiting factor. Many pro-audio DSP solutions aren't designed with the best of analog performance. From what I have heard, some manufacturers (BSS, Allen & Heath), have better analog performance than others like Biamp, QSC, etc. However, I cannot directly speak to the analog performance. What I can say, is that to me, a Biamp TesiraForte, running with line-level inputs, sounds fine, and based off some loopback measurements I performed, is transparent.

- Physical volume control / remote, with physical input switching. Pro DSPs here are going to falter, as they are usually designed to be controlled by networked remote, wall or table mounted control devices. For example: The Biamp TEC-1, various Crestron touch-panels, or sometimes physical devices like the Biamp Volume 8.
In the TEC-1's case, it can directly control a Biamp Tesira system, assuming you program it properly. Crestron touch panels require a Crestron processor to run, and the processor can control just about anything. The last category can get a bit messy, especially when you factor in proprietary control interfaces like the one found on the Volume 8.
Currently, my Tesira setup is controlled by using a small windows control surface (Biamp Canvas). I don't exactly "switch" inputs in the normal sense, as all of my inputs go into a few mixer "blocks" (In the DSP program), and I just mute / unmute them as needed, and often leave multiple inputs unmuted and use other devices to control it. For example, I have a direct USB input, and my computer's stereo analog input, into my system. The USB input is setup for zoom / teams calls, and my analog input is for music. I leave both unmuted, but usually nothing is connected to the USB input, and the volume control on my Focusrite intereface (PC -> Focusrite -> analog to Tesira), is turned down.
So basically, Pro DSPs have incredible control capabilities, and usually publish control API specifications. However, it is up to you to figure out how to control them without breaking the bank. (A word of warning: Some companies like Crestron don't allow you access to programming software, so if you get their hardware, it is effectively useless without the programming software. Do your research before buying anything!)

- Dirac Live. Nope. Not here. Biamp Tesira can run quite a bit of FIR filters, and so can many other pro DSP solutions, owing to the massive horsepower they usually have. But I have yet to see something like Dirac Live integrated into them.

- Balanced interconnects. Any good, modern pro DSP will have balanced I/O. The catch is that, at least on installation-focused models, the I/O is usually terminal connections. (See the Biamp TesiraForte, scroll to the bottom of page 2) So basically, if you have wire strippers, cutters, a multimeter, and can wire things, then it's fine. Otherwise, you should look elsewhere. One thing to consider is potentially using Dante or AVB to send audio digitally (over a network) to your amplifiers, but you should do your own research on that.

So some jumping-off points to consider:
Various companies make Live sound (ie, concerts, PA use, etc) DSPs, a few to look at are Lake processors, and the Meyer Galaxy (or whatever the new one is).
In the Installation world: Biamp Tesira (Nexia and Audia are old and underpowered), QSC Q-Sys, Symmetrix Prism / whatever the new one is, Allen & Heath AHM, and BSS Soundweb.
From the internet, people seem to think that the BSS products "sound" better for live music and other applications. They are also part of the Harman family which means integration with JBL speakers, and other things. I cannot attest to any of these claims, and my opinion of their product lineup is that it is a confusing mess.

So hopefully this helped you at least start somewhere!
One last thing: These companies often have free / openly available training resources, so if you are interested in learning about one they probably have resources available.

 

[gnarly]

@gnarly How do you do volume control and indication with Q-sys? Over some sort of network control? Or upstream analog preamp?

Always interested in how people make it work with the pro audio stuff.

Michael

Me too, always interested in finding out the various alternatives.

With Q-sys, given that the Core being used has a UCI license (User Control Interface-a license which almost all Cores tend to have),
just about any control element in the processing schematic can be put into a custom remote. Remotes being PC touchscreen, or any IOS device.

I use a PC touchscreen. Here's an example of relatively simple remote that has a master volume, and faders that let me play with dialing in a house curve.
The faders control amp gain to each or the individual driver sections. I could just put whatever number and type of conventional EQ's in the remote, but i like the amp gain technique better.
Point I'm trying to make is anything can go in the remote.

Here's an example of a more over the top remote, that I used when dialing in a pair of big 5-way horn speakers on top of big subs.
The volume faders work like the simple example above. Blue LA1-7 buttons make presets for wherever the faders are currently set.
Purple subcart buttons set sub's low-pass freq. Green syn11 18"s buttons set high-pass freq for the 18"s in the main horn speaker.

The green buttons: Flat FirD HC-1 HC-2 ..... Are where house curves are imbedded into the separated FIR files for each driver section.
So on the fly swapping of 6 FIR files is involved. (silent and instantaneous).

This remote was made to compare whether or not embedding a house curve into FIR files was advantageous vs section-level presets, and also to dial in sub to main xover.

Obviously most folks will never need or even want such control /experimentation capability....
But for those who might, qsys rocks ...

 

[Kal Rubinson]

@Kal Rubinson can gives us an insight to Merging as he's using one.

The Merging devices use Ravenna which is a version of AES67 network protocols. In practice, all the real processing is not in the HAPI which routes signals and does I/O. All the processing in my setup is done in Jriver which also controls volume, routes/mixes channels, DSP filters/convolutions and, via VST, DiracLive. The JriverPC uses an ASIO driver (MAD) to communicate with the HAPI via Ravenna. As long as there are only 2 devices and no complicated routing schemes, this will operate on a regular domestic LAN. Otherwise, a dedicated LAN with a managed switch is recommended.

There is a volume control on the HAPI accessible via a front panel knob or web page (i.e., via a smartphone or iPad) but it controls only the analog outputs.

In my situation, I use a mix of analog and AES/EBU outputs, so I prefer to use the volume control in Jriver.

 

[webmst007]

The latest Antelope Orion is not expensive and offers a load of features including speaker calibration that is pretty good. All the Atmos channels you need up to 9.1.6 and it can act as a monitor among lots of other features. Not sure how you go about playing all the material you might want to playback - but it woult certainly give you a lot of interest in playing around in a studio environment.
It is certainly on the radar of a few people I know in the UK to solve the need for cost effective Dolby Atmos management in studio control arrays.

 

[adam_dan]

Answering your original post:
I do not have much knowledge of the typical consumer DSP solutions. From my understanding, the typical way of accomplishing this would be with separate DSP units for the speakers and your subs (Say a MiniDSP 4x10 for the speakers, and a 2x4 for the subs), with some kind of preamp upstream for input switching and volume control.
Therefore, I will speak to my knowledge, which is the Pro Audio DSP world, or at least the installation-focused DSPs.

- It seems like you are looking for a system which can handle DSP processing, with 12 speaker outputs, and I presume a number of digital or analog inputs.
So that is our I/O requirement, and the limiting factor (as far as system hardware options) here is going to be the 12 outputs. If you don't already have amplifiers, you could theoretically use a digital protocol like Dante, Cobranet, or AVB, to send the audio to the amplifier(s), but these protocols do not usually run at > 48khz sample rates, and can introduce latency, along with network requirements (For example, a 5-port AVB-capable switch is like $400, Cobranet is obsolete at this point but is cheap, and Dante DSPs and amplifiers are crazy expensive).

- A good analog section for the analog I/O. This is another limiting factor. Many pro-audio DSP solutions aren't designed with the best of analog performance. From what I have heard, some manufacturers (BSS, Allen & Heath), have better analog performance than others like Biamp, QSC, etc. However, I cannot directly speak to the analog performance. What I can say, is that to me, a Biamp TesiraForte, running with line-level inputs, sounds fine, and based off some loopback measurements I performed, is transparent.

- Physical volume control / remote, with physical input switching. Pro DSPs here are going to falter, as they are usually designed to be controlled by networked remote, wall or table mounted control devices. For example: The Biamp TEC-1, various Crestron touch-panels, or sometimes physical devices like the Biamp Volume 8.
In the TEC-1's case, it can directly control a Biamp Tesira system, assuming you program it properly. Crestron touch panels require a Crestron processor to run, and the processor can control just about anything. The last category can get a bit messy, especially when you factor in proprietary control interfaces like the one found on the Volume 8.
Currently, my Tesira setup is controlled by using a small windows control surface (Biamp Canvas). I don't exactly "switch" inputs in the normal sense, as all of my inputs go into a few mixer "blocks" (In the DSP program), and I just mute / unmute them as needed, and often leave multiple inputs unmuted and use other devices to control it. For example, I have a direct USB input, and my computer's stereo analog input, into my system. The USB input is setup for zoom / teams calls, and my analog input is for music. I leave both unmuted, but usually nothing is connected to the USB input, and the volume control on my Focusrite intereface (PC -> Focusrite -> analog to Tesira), is turned down.
So basically, Pro DSPs have incredible control capabilities, and usually publish control API specifications. However, it is up to you to figure out how to control them without breaking the bank. (A word of warning: Some companies like Crestron don't allow you access to programming software, so if you get their hardware, it is effectively useless without the programming software. Do your research before buying anything!)

- Dirac Live. Nope. Not here. Biamp Tesira can run quite a bit of FIR filters, and so can many other pro DSP solutions, owing to the massive horsepower they usually have. But I have yet to see something like Dirac Live integrated into them.

- Balanced interconnects. Any good, modern pro DSP will have balanced I/O. The catch is that, at least on installation-focused models, the I/O is usually terminal connections. (See the Biamp TesiraForte, scroll to the bottom of page 2) So basically, if you have wire strippers, cutters, a multimeter, and can wire things, then it's fine. Otherwise, you should look elsewhere. One thing to consider is potentially using Dante or AVB to send audio digitally (over a network) to your amplifiers, but you should do your own research on that.

So some jumping-off points to consider:
Various companies make Live sound (ie, concerts, PA use, etc) DSPs, a few to look at are Lake processors, and the Meyer Galaxy (or whatever the new one is).
In the Installation world: Biamp Tesira (Nexia and Audia are old and underpowered), QSC Q-Sys, Symmetrix Prism / whatever the new one is, Allen & Heath AHM, and BSS Soundweb.
From the internet, people seem to think that the BSS products "sound" better for live music and other applications. They are also part of the Harman family which means integration with JBL speakers, and other things. I cannot attest to any of these claims, and my opinion of their product lineup is that it is a confusing mess.

So hopefully this helped you at least start somewhere!
One last thing: These companies often have free / openly available training resources, so if you are interested in learning about one they probably have resources available.

Thanks, this basically sums up my thoughts on live sound stuff that were previously scattered around various tabs and forum posts. It seems like the tradeoffs are either in raw horsepower or in architecture complexity, right? I'd love to find a solution that doesn't have a PC, though a headless linux server wouldn't even be noticeable when just wanting to play music.

There definitely seems to be a trend of moving DSP in live sound setups to the "edge," meaning that doing DSP before the amps seems to be getting harder, big live sound installs seem to be moving DSP to the amplifier (which is often paired with the speaker now!?). I've always liked the look of seperates, especially for a listening room, and want to DIY most of this system, including the amps, if I can.

In terms of control, it seems like as long as I have logic inputs, I can make something work, and since it's all rackmount, just have patch panels to send balanced signals from DSP to amps. (This would be quite a bad project if I didn't know how to wire things haha)
Here's the pro/live options as far as I can tell.
- Biamp Tesira: Capped at 24/48Khz, TesiraForte doesn't have enough outputs, Tesira Server would work but $$$$. DSP seems to be flexible enough to do what I want.
- Lake Processors: 2 LMX48 chained together would do 24/96, dsp seems flexible enough but similar in price to above. ($$$$)
- QSC 110f: Capped at 24/48, but USB is capped at 16/48 (neither of the above have USB audio as far as I can tell). Not super expensive, and control options seem good. Seems aimed more at conference installs than live sound, but flex channels are neat. ($$)
- BSS: Just heard about these guys, looks like the OMNI or London series could meet my needs, since they go to 24/96 it seems. (OMNI: $$$$/London: $$$)
- AHM - Seems like a lot of the same capabilities as the QSC, but more live-sound focused.
- Meyer Galaxy - The 816 meets my hardware needs, but the software seems limited in terms of flexibility. Never heard of anybody running these on non-meyer systems. Seems to be the most expensive solution, but attractive for a number of reasons, as long as the software would let me do what I want. ($$$$.5)
- Symetrix: Limited to 48Khz.

It still seems like my best option is getting an RME interface, doing DSP on a PC (Maybe a Pi if I can get over my fear of SD cards, I assume CPU horsepower isn't really a concern, audio has relatively little information compared to say video), and getting a good A/D converter (oh wait there's one in the RME!) to level match the headphone output and line outputs (I wonder if there's a way to disable the front panel volume controls)

Theoretically, my rack looks like this:
1U: Mac mini or Rpi Enclosure
1U: RME Fireface
1U: Custom rackmount unit with balanced drivers (run the headphone outs in the front like patch cables), OLED, IR reciever, and "main system" volume knob.
And then 12 XLR cables out the back of the rack to amplifiers

If I got a Fireface and sent it to Amir, I assume he'd be able to run some tests on it? I've got access to all the test equipment at work, just not the expertise.

 

[AwesomeSauce2015]

There definitely seems to be a trend of moving DSP in live sound setups to the "edge," meaning that doing DSP before the amps seems to be getting harder, big live sound installs seem to be moving DSP to the amplifier (which is often paired with the speaker now!?). I've always liked the look of seperates, especially for a listening room, and want to DIY most of this system, including the amps, if I can.

Yep. Pairing amps with speakers as a package does offer a lot of benefit for the manufacturer, as they can guarantee certain performance parameters are met. IE: I can have proper limiting behavior since I know the amp's parameters, and the speaker's limits.
What I settled on for my system (Which I plan on doing a bit of a write-up about, eventually), was a Biamp TesiraForte AVB, which handles mixing and DSP processing, and a non-DSP older crown amplifier to power the speakers. It had enough I/O for my current needs, and AVB allows me to expand it if needed.

In terms of control, it seems like as long as I have logic inputs, I can make something work, and since it's all rackmount, just have patch panels to send balanced signals from DSP to amps. (This would be quite a bad project if I didn't know how to wire things haha)
Here's the pro/live options as far as I can tell.
- Biamp Tesira: Capped at 24/48Khz, TesiraForte doesn't have enough outputs, Tesira Server would work but $$$$. DSP seems to be flexible enough to do what I want.
- Lake Processors: 2 LMX48 chained together would do 24/96, dsp seems flexible enough but similar in price to above. ($$$$)
- QSC 110f: Capped at 24/48, but USB is capped at 16/48 (neither of the above have USB audio as far as I can tell). Not super expensive, and control options seem good. Seems aimed more at conference installs than live sound, but flex channels are neat. ($$)
- BSS: Just heard about these guys, looks like the OMNI or London series could meet my needs, since they go to 24/96 it seems. (OMNI: $$$$/London: $$$)
- AHM - Seems like a lot of the same capabilities as the QSC, but more live-sound focused.
- Meyer Galaxy - The 816 meets my hardware needs, but the software seems limited in terms of flexibility. Never heard of anybody running these on non-meyer systems. Seems to be the most expensive solution, but attractive for a number of reasons, as long as the software would let me do what I want. ($$$$.5)
- Symetrix: Limited to 48Khz.

Yeah, you could do it with logic inputs, if you had enough. (Disclosure, I have the Tesira certification so I will use their products for examples)
If you didn't then something like an EX-Logic would allow you to expand it.
At least on Tesira, the logic inputs can have potentiometers attached to them, which can then control things like level controls in the program. If you can run wires, then having something like that would allow a physical volume control, and you could add switches and LEDs to switch inputs.

Yes, Tesira is capped at 24/48khz. Theoretically if you are willing to have around 2ms of latency, you could use AVB to connect 2 Fortes together, but you would still not be at 96khz. One major thing, especially for me, was that used on ebay, Tesira hardware is cheap. I can get an AVB Forte for like $100, and a Server-IO for around $400-$800. The BSS processors were closer to $300, in my research.
No matter what you get I would recommend looking at the used market as the new / MSRP prices are a bit ridiculous on all of these.

I wouldn't recommend relying on any USB inputs on these devices, as they are optimized for teleconferencing. They sometimes have high latency, and at least in my experience with my TesiraForte, it just sounds bad for some reason. In a back-to-back comparison, running analog in from my Focusrite interface actually sounded better, to me.

As for BSS, since it is probably your best bet in this case, be careful when researching the products, some have DSP capabilities and others are just expanders. The issue is that unlike Biamp, the expanders aren't really differentiated from the DSP models. From my understanding, London is the previous gen product, and Omni is the current gen product, so naturally London will be cheaper on the used market. You could also possibly use BLU-Link and/or HiQnet to have digital audio to amplifiers and control. The last comment about BSS is that their programming software kinda sucks. I downloaded it and messed around with it, and it just isn't intuitive or very polished in my opinion. But, for something like this, you probably won't be trying to set up 3 separate mixes with individualized processing and routing, like I was. -- Though our resident Harman dealer(s) may have some better advice / knowledge regarding this DSP platform.

Lastly, regarding amplifiers: If you were to get a DSP platform that works well with a given amplifier, you could benefit from product synergy like having better limiting and DSP integration, as well as digital audio and control links (IE: Tesira amp 4350r, uses AVB to get audio from a Tesira DSP, and you can control it from the software). From past knowledge with the LX521 speakers, you don't need big amps to drive them, just a lot of amps, so you probably don't need the power of the big class-D pro-audio amplifiers.

The other thing regarding amplifiers is about how to turn them on or off. I want to get a networked power distribution device (Which I will control with Tesira), but some exist with trigger voltage controls as well. What you don't want to do is turn the whole system off and on at once. You'll want to leave the DSP on (they are designed for 24/7 operation), because they (at least Tesira) are slow to boot up, and turn the amps on and off as needed, with sequencing if you are also turning off the DSP.

It still seems like my best option is getting an RME interface, doing DSP on a PC (Maybe a Pi if I can get over my fear of SD cards, I assume CPU horsepower isn't really a concern, audio has relatively little information compared to say video), and getting a good A/D converter (oh wait there's one in the RME!) to level match the headphone output and line outputs (I wonder if there's a way to disable the front panel volume controls)

Theoretically, my rack looks like this:
1U: Mac mini or Rpi Enclosure
1U: RME Fireface
1U: Custom rackmount unit with balanced drivers (run the headphone outs in the front like patch cables), OLED, IR reciever, and "main system" volume knob.
And then 12 XLR cables out the back of the rack to amplifiers

If I got a Fireface and sent it to Amir, I assume he'd be able to run some tests on it? I've got access to all the test equipment at work, just not the expertise.

I can't comment on this, personally I wasn't a fan of doing DSP through a PC, and I wanted AEC and other mic processing / mixing for my system. (I have 3 mixes, a Tx mix for zoom/teams calls, a speaker mix, and a headphone mix (it adds a low-latency mic feed)). You probably don't need all that, so a PC solution might work better for you.
My only comment here is about the RME Fireface. IDK where they are getting the crazy high channel count numbers from, but it doesn't look like it has enough balanced analog outputs for your needs. (I am not counting the headphone outputs).
And if you want to make your own control interface, then more power to you!
Either way, seems like you have some good options and ideas, so if you have any questions LMK but otherwise good luck and have fun

 

[adam_dan]

The latest Antelope Orion is not expensive and offers a load of features including speaker calibration that is pretty good. All the Atmos channels you need up to 9.1.6 and it can act as a monitor among lots of other features. Not sure how you go about playing all the material you might want to playback - but it woult certainly give you a lot of interest in playing around in a studio environment.
It is certainly on the radar of a few people I know in the UK to solve the need for cost effective Dolby Atmos management in studio control arrays.

I've heard not so great things about Antelope drivers and support. Wouldn't be an issue if I didn't want to do linux eventually.

My only comment here is about the RME Fireface. IDK where they are getting the crazy high channel count numbers from, but it doesn't look like it has enough balanced analog outputs for your needs. (I am not counting the headphone outputs)

Haha welcome to the world of audio interfaces - they are counting digital outputs, over ADAT/AES etc. as part of that channel number. That number is just the number of channels on the matrix mixer, you'd need more hardware to actually get that many ins/outs. I think I can use the headphone outs and use THAT corp. driver chips to drive balanced cables to the subs. If not, I need to figure out how to either get 4 channels out over ADAT, or use the 2 ADAT interfaces as SPDIF and use two dacs. This introduces latency on the subs, which I will need to delay the main system to match.

 

[Devnull]

I have to respectfully disagree with AwesomeSauce2015. I like the BSS software because it's completely free form. I get to choose what processing blocks are in the signal chain and the order that they are presented. The BSS units boot fairly fast, within a few seconds of my AVR, and are ready to go when the delayed trigger on the AVR starts the last amp.

 

[AwesomeSauce2015]

I have to respectfully disagree with AwesomeSauce2015. I like the BSS software because it's completely free form. I get to choose what processing blocks are in the signal chain and the order that they are presented. The BSS units boot fairly fast, within a few seconds of my AVR, and are ready to go when the delayed trigger on the AVR starts the last amp.

Like I said, if I was to put in the time to learn it, it very well could be good.
I think my main situation is that I learned open-architecture DSP with Biamp software, and Tesira is a natural progression from the Audia / Nexia software. Hence, I am just really familiar with Biamp's software. FWIW, Tesira is also open-architecture (Free form, you choose the processing layout, etc). Depending on your needs it can be incredibly useful. Neat to know that BSS hardware boots quickly. Which model processor do you have? I wonder if networked audio and/or AEC hardware influences the boot times.

 

[DWPress]

Just to chime in - a silicon Mac mini will never engage the fan. I've had a M1 and currently have an M2 and have never heard the fan and it sits 2 feet from me. Hang Loose Convolver can handle XO and EQ for all your channel needs and still possible to use something like DLBC if you're interested.

It is possible to link 2 Okto Dac8 Pros together giving you 16 channels and a remote but you'd need to add an ADC. I use a MSB ADC and which inputs AES into the Okto, software loopback allows me to play my turntable anytime, don't even need to push a button.

 

[AwesomeSauce2015]

Just to chime in - a silicon Mac mini will never engage the fan. I've had a M1 and currently have an M2 and have never heard the fan and it sits 2 feet from me. Hang Loose Convolver can handle XO and EQ for all your channel needs and still possible to use something like DLBC if you're interested.

I just remembered that.
At least on my case, with the Tesira hardware, (and probably many other pro DSP solutions), it does have a fan. It is audible, but I have other things in the same room that are louder, so to me it isn't an issue. However, for OP, if you are going for an "ideal" listening room, using hardware with no fans, or that will never engage its fans, is probably a good idea.
Same issue goes for pro audio amps. Some have fixed speed fans, and others have variable fans. The variable speed fans are obviously better here, but it depends on your use case. In my case, the fans ramp up at around 100w/ch... So basically they never ramp up, but they are audible, just not very audible in my room.

 

[adam_dan]

Just to chime in - a silicon Mac mini will never engage the fan. I've had a M1 and currently have an M2 and have never heard the fan and it sits 2 feet from me. Hang Loose Convolver can handle XO and EQ for all your channel needs and still possible to use something like DLBC if you're interested.

It is possible to link 2 Okto Dac8 Pros together giving you 16 channels and a remote but you'd need to add an ADC. I use a MSB ADC and which inputs AES into the Okto, software loopback allows me to play my turntable anytime, don't even need to push a button.

I need to sit with that thought for a bit, the idea of a fan bugs me quite a bit, but I do like the hardware. In my current setup, my turntable noise floor is HIGH, I think because of disgusting mains power in the walls that my system picks up, so I couldn't do always-on loopback without introducting audible noise.

 

[DWPress]

so I couldn't do always-on loopback without introducting audible noise.

It's not always on, it's in software - like a virtual cable on the computer. The software is from Rogue Amoeba and it is called "Loopback".

I turn off my phono preamp when I'm not using it and, as I said, the TT phono is run into an ADC then sent to Okto via AES then to Mac via USB for processing THEN back to the Okto via USB for distribution. I run eight channels all in software on the Okto but it has recently occurred to me I could also use my old Minidsp 4x10 to potentially add channels since it has AES output that could also come into the Okto. Haven't played with that possibly though.

The Okto really is a versatile piece of kit if you can utilize the digital inputs and run it in AES/USB mode which opens it up to utilizing many pro audio processing gear as well.

I put up a post a couple weeks ago about the MSB toys here.

edited for more info

 

[adam_dan]

It's not always on, it's in software - like a virtual cable on the computer. The software is from Rogue Amoeba and it is called "Loopback".

I turn off my phono preamp when I'm not using it and, as I said, the TT phono is run into an ADC then sent to Okto via AES then to Mac via USB for processing THEN back to the Okto via USB for distribution. I run eight channels all in software on the Okto but it has recently occurred to me I could also use my old Minidsp 4x10 to potentially add channels since it has AES output that could also come into the Okto. Haven't played with that possibly though.

The Okto really is a versatile piece of kit if you can utilize the digital inputs and run it in AES/USB mode which opens it up to utilizing many pro audio processing gear as well.

I put up a post a couple weeks ago about the MSB toys here.

edited for more info

The okto seems awesome, had my eyes on it for a while when this system was only planned to be 8 channels. My concern is being able to synchronize the two devices with the latency I need, since them being even 1 or 2 samples out of lockstep could introduce phase issues into my system, at least that's what I'm worried about. They don't have the ability to output/input word clock like some of the pro audio offerings.

I think my best option right now is the Motu 16A and a linux box (since I'll do volume control in DSP). Maybe I'll get everything configured on mac, just since linux audio can get MESSY, though at that point I might as well use a mac mini, especially if the fan doesn't turn on. If it gives me issues, I'll return it and save up for an RME interface.

 

[DWPress]

You could contact Pavil @ Okto and ask about the current ability to link units, the information I'm remembering is a couple years old now and firmware has improved functionality. I know that the word clock coming into the Okto is dictated by the first AES input and subsequent AES inputs conform.

 

[Fledermaus]

Sort of solved this, on a smaller scale (2-ch audio, active XO), with a Xilica Solaro processor (DSP, 96 kHz) with 2 GPIO add-on cards, attached to an Arduino w/IR sensor, relay card and motorized pot.
Once your preset is done, you just need a remote to turn the system on/off, select sources, and control the volume - you can even set the volume with a knob, last century way

 

[Geert]

It still seems like my best option is getting an RME interface, doing DSP on a PC (Maybe a Pi if I can get over my fear of SD cards

You can easily connect an NVMe SSD drive to a Pi via a dedicated HAT board or USB adapter. The Pi 5 even has a PCIe interface for it.

 

[Geert]

Maybe I'll get everything configured on mac, just since linux audio can get MESSY

It takes some learning, but once you understand Linux ALSA it gives you the most control over the audio path.

 

[Devnull]

Like I said, if I was to put in the time to learn it, it very well could be good.
I think my main situation is that I learned open-architecture DSP with Biamp software, and Tesira is a natural progression from the Audia / Nexia software. Hence, I am just really familiar with Biamp's software. FWIW, Tesira is also open-architecture (Free form, you choose the processing layout, etc). Depending on your needs it can be incredibly useful. Neat to know that BSS hardware boots quickly. Which model processor do you have? I wonder if networked audio and/or AEC hardware influences the boot times.

I'm back to my original combo of a 9088iis and a 9008. The 9008 is there because I was playing with some extra FIR filters and needed some extra processing power. I had a blu-160 for a while but traded it for a Browning A5.

The blu-160 was a little nicer than the 9088iis. 98khz vs 48khz and the ability to retrieve the configuration over the network are the two prominent differences that I remember. I can't speak as to how hardware config impacts boot time but IIRC boot times were under 20 seconds or so for the 160.