Multi-Channel, Multi-Amplifier Audio System Using Software Crossover and Multichannel-DAC

[SDC]

Hello @SDC, welcome to this multichannel project thread!

(I also carefully read through your posts on the "hot" Sointuva thread...)

Thank you for your info and suggestions on another way of system-wide XO/EQ configurations which would be of great help for many people periodically visiting this thread.

In my multichannel setup, as shared in my above post #558 (and @Doodski just kindly responded positively in his above post), I would like to always have safe straightforward and simple system-wide (one-stop) digital XO/EQ GUI like EKIO, and also would like to have flexible on-the-fly relative gain controls depending on the "hearing tastes" and/or "hearing abilities (especially in high Fq)" of the listeners/audiences.

Even for this, I would like to fully eliminate using any attenuators or potentiometers in analog line level signals and SP high level signals.

I would like to understand, your approach/configurations in your above post, and my multichannel approach/configurations shared through out this thread, are illustrating wide range of choices for people interested in active multichannel multi-amplifier audio system!

Thank you. For GUI there is Beqdesigner, but it doesn't work on my computer so can't say more about it...

If it wasn't for REW's accurate simulation I too wouldn't have tried Jriver for Active XO.

It takes so much time to copy and paste all the numbers from REW, if it wasn't accurate I would've just given up!

 

[Doodski]

(I also carefully read through your posts on the "hot" Sointuva thread...)

That is putting it mildly! That Sointuva thread was major mess. I wish I never read it. Unfortunately I could not keep my eyes off the spectacle that it was. It was too juicy to ignore.

 

[MCH]

Hi dualazmak, sorry to interrupt, and first of all thank you for the several learnings from your super detailed thread, that i read from time to time with admiration.

I have a question for you regarding gain (volume) control.

I understand and agree that the safest is, as you mention a couple of post above, integrated amplifiers and set a safe negative gain and then use the master upstream as your system wide volume control.

However, this brings a lot of comexity and expense to the system, for something that can be easily done in the digital domain.

Do you think it would be possible to have, after your computer that runs your player etc, a second dedicated small computer that simply takes the multichannel signals applies a "safety" negaive gain and from there run everything on power amplifiers not necessarily with volume control? (A sort of digital pass-through with gain)
The cost would be minimum and the chances something goes wrong in the two computers i guess is pretty low. The benefit would be to be able to build the system on potentially smaller and cheaper power amps.

What would be an easy and reliable way to do such a thing?

Thanks a lot

 

[dualazmak]

It takes so much time to copy and paste all the numbers from REW, if it wasn't accurate I would've just given up!

I agree, and this aspect is one of the main reasons for my preferences on simple straightforward easy-to-configure (even on-the-fly) GUI of wonderful EKIO!

 

[Doodski]

The cost would be minimum and the chances something goes wrong in the two computers i guess is pretty low. The benefit would be to be able to build the system on potentially smaller and cheaper power amps.

Not with exotic and esoteric beryllium drivers and rare Yamaha woofers. Safety is paramount.

 

[MCH]

Not referring to dualazmac's case in particular, i don't think he will ever want to go back to start from scratch anyway. But to someone starting to build a complex multichannel system, not necessarily with expensive drivers. That something goes wrong with the gain accidentally set to 0 in a computer is something most of us fear no matter how expensive our speakers are...

 

[dualazmak]

Do you think it would be possible to have, after your computer that runs your player etc, a second dedicated small computer that simply takes the multichannel signals applies a "safety" negaive gain and from there run everything on power amplifiers not necessarily with volume control?

Hello @MarcosCh, you are very much welcome!

Actually you can do it in still-digital stage on EKIO's output panels' gain controllers even on-the-fly.

For the safety reasons, I do not recommend you to do the relative gain control on-the-fly using keyboard numeric input; please use mouse wheel up-down having the mouse pointer on the gain controller bar which is much safer.

Using a second computer,,, I assume it would be another doable idea/concept, but we need to have multichannel transfer routing (by full ASIO) and software, and we also carefully measure validale solve the latency and synchronization issues throughout the two computers and multichannel DAC...

 

[SDC]

Hi dualazmak, sorry to interrupt, and first of all thank you for the several learnings from your super detailed thread, that i read from time to time with admiration.

I have a question for you regarding gain (volume) control.

I understand and agree that the safest is, as you mention a couple of post above, integrated amplifiers and set a safe negative gain and then use the master upstream as your system wide volume control.

However, this brings a lot of comexity and expense to the system, for something that can be easily done in the digital domain.

Do you think it would be possible to have, after your computer that runs your player etc, a second dedicated small computer that simply takes the multichannel signals applies a "safety" negaive gain and from there run everything on power amplifiers not necessarily with volume control? (A sort of digital pass-through with gain)
The cost would be minimum and the chances something goes wrong in the two computers i guess is pretty low. The benefit would be to be able to build the system on potentially smaller and cheaper power amps.

What would be an easy and reliable way to do such a thing?

Thanks a lot

It is possible to go

pc1-avr-audio interface-pc2-audio interface-amp-speaker

using pc2 as fixed setup with volume protection


But ain't cheap. Pro systems for multichannel mixing costs a ton! Starting from Trinnov/JBL for Dante output.

 

[Doodski]

Not referring to dualazmac's case in particular, i don't think he will ever want to go back to start from scratch anyway. But to someone starting to build a complex multichannel system, not necessarily with expensive drivers. That something goes wrong with the gain accidentally set to 0 in a computer is something most of us fear no matter how expensive our speakers are...

I had expensive rare speaker drivers and one day I touched a RCA cable and BaMO BaZZaP! The tweeter was instantly blown. I had safety measures in place but the static electricity got me anyway. In active system always have a backup plant and safety in mind. Not extra PC's for volume control.

 

[dualazmak]

Let me add that I prefer "as short and simple digital path as possible" avoiding possible inclusion of any digital noises...

I had expensive rare speaker drivers and one day I touched a RCA cable and BaMO BaZZaP! The tweeter was instantly blown. I had safety measures in place but the static electricity got me anyway. In active system always have a backup plant and safety in mind. Not extra PC's for volume control.

Yes, as you know, I always have protective capacitors in the SP high level signals going into my treasure Be-SQ, Be-TW and metal-horn-ST!

 

[Doodski]

Yes, as you know, I always have protective capacitors in the SP high level signals going into my treasure Be-SQ, Be-TW and metal-horn-ST!

I was running a pure direct system with no XL and Xc reactance for the sake of signal integrity and phase relationship but I got bitten.

 

[3ll3d00d]

8. Two HP/LP is one LW HP/LP filter. But following REW EQ simulations thoroughly for acoustic LW filter is the most important part.

I recommend using https://yabb.jriver.com/interact/index.php/topic,129609.0.html for configuring jriver DSP in an active setup in order to get a convenient way to access multiple filter types, an easier way to handle routing, to visualise the filters and to enable import from REW or MSO. Not sure why it didn't work on your computer though, free feel to log something on GitHub about it if you want to try to get it working.

Edit: I did also push a new release recently to add that MSO support and fix some things that broke after mc29 came out

 

[3ll3d00d]

btw on this subject of this thread and recent posts, I do use a Motu 1248 (augmented by a rme ff800 connected via ADAT to get a few more output channels) with jriver using both peq for xo (for low latency use) and acourate for jriver sourced content.

 

[dualazmak]

btw on this subject of this thread and recent posts, I do use a Motu 1248 (augmented by a rme ff800 connected via ADAT to get a few more output channels) with jriver using both peq for xo (for low latency use) and acourate for jriver sourced content.

Hello @3ll3d00d, thank you for your visiting here.

I am happy to know that you use MOTU 1248 in your multichannel digital active setup. I also heard that @mdsimon2 confirmed and validated that FOCUSRITE Scarlett 18i20 2nd gen (and 3rd gen) (CS4272 [x4] + CS4392), as well as MOTU UltraLite Mk5 (two ES9026pro), can be used as more than 8-CH multichannel DAC;
- Information exchange is in progress after this post on this remote thread regarding FOCUSRITE Scarlett 18i20 2nd gen (and 3rd gen) (CS4272 [x4] + CS4392) as well as MOTU UltraLite Mk5 (two ES9026pro) ; it looks both of them support simultaneous sync 10-channel DAC processes.

You would please allow me asking you how many XO/EQ-ed channels can MOTU 1248 receive through USB ASIO for sync DAC processing in 192 kHz sampling rate?
How many independent "physical" analog out channel (inclusive of its headphone out, if available) can MOTU 1248 provide you?
And, which DAC processor chip and how many of it in MOTU 1248?

Since I used to use OPPO Sonica DAC (one ES9038pro in it, I still keep Sonica DAC!), I rather prefer ESS DAC processors, ES9028por, ES9038pro and ES Sabre32. My OKTO DAC8PRO has one ES9028pro in it. I am very much looking forward to the announced debuts of new generation ESS DACs as well as AKK DACs installed in new multichannel DAC units in the near future.

How can you measure and adjust/establish 0.1 msec precision time alignments of actual room air sound between all the SP drivers in your multichannel setup?

 

[3ll3d00d]

You would please allow me asking you how many XO/EQ-ed channels can MOTU 1248 receive through USB ASIO for sync DAC processing in 192 kHz sampling rate?
How many independent "physical" analog out channel (inclusive of its headphone out, if available) can MOTU 1248 provide you?
And, which DAC processor chip and how many of it in MOTU 1248?

it has 12 1/4" balanced dc coupled TRS line out at the back + 2 1/4" stereo headphone out at the front. I've never used the latter & they are spec'ed differently so I'm not sure how they behave. I had the RME already so using that rather than headphone out seemed simpler.

Google tells me it uses 2xES9016 DACs for the 12 TRS outputs. I think the 1248 was part of the initial wave of AVB models which motu have not refreshed since then as far as I'm aware. It's not a concern for me personally but for sure it's getting towards 10 yrs old now. I think the useability of these models is v good (remote ui via a browser, robust multiclient asio driver, good output levels and clean sound).

the only downside I've seen is the way the PSU is built in in a way which means if it fails, you have to swap the entire device. RME devices are more expensive up front but a mile ahead on serviceability given that you can swap the PSU yourself. To be fair, the local motu service agent did swap the device at a reasonable price but ultimately I'd be happier buying RME 2nd hand vs Motu for example but motu is generally cheaper as a result (but not cheap in absolute terms, just cheaper than an RME).

To put that in concrete terms, I bought the motu in 2016 and had to get the PSU replaced in 2019. I bought the FF 2nd hand in 2014 (no clue how old it was at that point, fairly sure "quite old" is the answer) and had to buy a new PSU for it in 2018.

How can you measure and adjust/establish 0.1 msec precision time alignments of actual room air sound between all the SP drivers in your multichannel setup?

I have generally used typical acourate approaches for in room time alignment, this is mostly about the subs though as my speakers are of my own design so I have my own quasi anechoic measurements to determine alignment within a single speaker as part of crossover design. This is a predominantly HT setup btw so 3 way LCR + passive surrounds + 2 sub channels (all DIY).

 

[dualazmak]

Hello again @3ll3d00d, thank you for your really in-depth introduction and descriptions of MOTU 1248; the information you kindly gave above would provide another multichannel DAC option for many people periodically visiting this thread.

I am very happy having this thread as general and comprehensive one-stop showcase in active digital multichannel multi-driver multi-amplifier audio systems.

 

[Holmz]

Thank you. For GUI there is Beqdesigner, but it doesn't work on my computer so can't say more about it...
…

Reading that just about that just about made me spit out my coffee laughing.

 

[mccarty350]

I want to thank every participant in this thread, it's been off the charts informative and especially Dualazmak for his stellar documentation and efforts to keep the sticky posts in this thread current!

 

[SDC]

Reading that just about that just about made me spit out my coffee laughing.

And I have to drink coffee everytime I fix my PEQ. Beqdesigner, openDRC, nothing works invalid rendering whatever that is. However dirac and audiolense works so into the labyrinth the solution goes...

 

[dualazmak]

And I have to drink coffee everytime I fix my PEQ. Beqdesigner, openDRC, nothing works invalid rendering whatever that is. However dirac and audiolense works so into the labyrinth the solution goes...

Thank you for your further info...

At least in my setup, I am very happy keep using EKIO as "OS-system-wide" XO/EQ center providing really safe (even on-the-fly parameter changes), very flexible, straightforward, robust, simple and well-designed GUI operations.

Just for all of our info;
- EKIO uses IIR filters; cascade 2nd order direct form II biquad in 64 bit floating point: #138-#142
- EKIO (using IIR filters) gives no audible post-ringing nor pre-ringing at all, at least in my XO/EQ configurations: #143, #181