Introduction to the SmartCross project

[Nonick]

From a strictly subjective point of view: if i would find a decent All-in-one unit on the market, i'd buy it.
Ideally some combination of: Amp-DAC-Streamer.

2channels with subwoofer out option (and highpass filter)
For example - ideal All-in-one desktop unit would be: WiiM Pro (with screen), MiniDSP Flex and Topping PA5 II in one box.

Whats wrong with those three units as separate devices? It gets messy and needs control from various places -Stream on WiiM, open app, setup MiniDSP Flex and control volume, turn on PA5, switch inputs on PA5...
Therefore "unifying" those three units in All-in-one box would be perfect. One ring to rule them all.
Features: Screen, remote control, (P)EQ, 2 Ch Stereo, RCA input, USB, Optical, BT, Wi-Fi. Power min. 50W / 8 Ohm, 120W / 4 Ohm. SINAD 95 or above (so we avoid annoying tweeter hiss)

 

[Zek]

... and durability of at least 5 years!!!

 

[chnan]

Hi
I am just another random reader. It seems like you are interested in market definition of your project.

1. Since your product is DSP based digital-cross-over+amp for speaker DIYers, you have to consider the diverse possibilities. One may need a compact 2-way with small foot-print (60W each-channel/F3 60Hz is more than enough), while another could be seeking 4-way full-scale sounds. A single box to satisfy everyone could be difficult, especially in a relative small market.

2. 2* Hypex FA253 is around 1000 USD, if you want to take over their market share, you may need to reduce the price by 50% => ie. 500-650 USD for 6 channels DSP-to-speaker-outputs. It is possible, one TPA3221 amp board from Shenzhen Sanruisheng Technology retails for $20~30 USD. If you can arrange 3 TPA3221 or TPA325x with PFFB at the similar costs plus some amazing ASR scores, you are good. On the other hands, you should be aware the chip quality and sources for a long-term business. Comments on XY LM3886 amps from TomCHR are worth reading (how come a pack of $22 components could be sold for $11?)

Otherwise, if you are competing with Arylic quality, I don't think it would be easier for small business.

3. Software perspective, I only suggest to check RME. They are the No.1 in my mind. Whatever you use: PC, mac, iphone, ipad.... simply stable full-functions.

4. just another idea, since all the functions (DSP/multi-channels amp/EQ/gain-control....) are already built-in among most HT receivers, you can simply made your rocket-DSP/FIR-kernels to overwrite the functions of one receivers, and make one HT receiver is easy for Active-Cross-Over jobs (receive 2 channels from RCA/SPDIF/eARC.. and do the FIR things for every speaker unit), just like a MCU device for a car. If you can do this, I will buy 3 of them.

Good luck developing

 

[popej]

SmartCross is an all-in-one device that combines digital processor, DAC converter and amplifier all into a single box.

Great idea. Actually I like my AVR too

 

[Roland68]

Thank you for your interest!

For HDMI please see below.

The screen of this device is only for displaying the control GUI, so it would not be useful to add an HDMI output.

But, we have a control app with the exact same interface that runs on PC and mobile phones! So if you need to control it from the other side of the room, you can just install the control app in your phone or PC, and control the device over WiFi.

And that's exactly where you're wrong.
An HDMI input with (e)ARC would expand your buyer group to include a niche area that is very important for small manufacturers.
Take a look at these threads...
https://www.audiosciencereview.com/forum/index.php?threads/hdmi-and-dacs-not-evolving.38838/
https://www.audiosciencereview.com/...ity-budget-a-v-preamplifier-processors.28500/
And pay special attention to this post...
https://www.audiosciencereview.com/...-v-preamplifier-processors.28500/post-1363618

 

[bluefuzz]

It's more like the latter for now (but multiple guys) - but we have not only solder iron, but also SMT station, oscilloscopes, signal generators, etc, and we have experienced embedded developers.

Fair enough.

But you don't have to use a Pi if you have SmartCross. You may not even have to use your NAD anymore - I'd like to know what's your typical audio source?

Obviously SC would replace the Pi, but it would also need to be able to replace the NAD as well to be really attractive for my use case. At the very least it would need to able to do Spotify Connect and stream local content from a NAS with a decent interface (not DLNA). Preferably something akin to Moode/Volumio. I could easily live without Bluetooth and wifi but wired ethernet is also a must-have.

if performance is very important to you (but that means you have to use external amplifiers).

I'm not fixated on ultimate SOTA performance but it needs to perform at least as well as my current setup in that respect. Preferably better than fully transparent for 16bit content. I don't see much point in the device if you have to use external amps or streamers to achieve decent output quality or functionality.

Could you please tell us how much do you think the "effort" is worth?

That's very difficult to say but from my perspective probably not more than the used price of a NAD C658, a Motu M4, an RPI4 and 4 UCD180s ... ;-)

So if you need to control it from the other side of the room, you can just install the control app in your phone or PC, and control the device over WiFi.

Which then makes the display and volume knob redundant. If there's going to be an app/web interface I don't see any point in a physical volume knob or display. They just add cost and complexity.

 

[RandomEar]

Interesting project and I like the "hackability" aspect. I would, however, consider a line level subwoofer output with configurable low-pass as mandatory for a device like this. Otherwise, integrating a sub becomes a real hassle and essentially requires one with high level inputs and/or lots of tinkering. I also think that offering a smaller model with only 2 amp slots and 4 outputs would be a welcome addition in the future.

 

[Deleted member 48726]

Which then makes the display and volume knob redundant. If there's going to be an app/web interface I don't see any point in a physical volume knob or display. They just add cost and complexity.

I'd much much rather have it all in one with the touch display than another app among the myriad of other household devices with their dedicated apps. And a volume knob is just.. Nice to have.

 

[dualazmak]

SmartCross is mainly targeted at stereo speakers.

If so, you would please aware of the concept and possible policy; "right person in right place" approach/arrangement(s) for multiple-amplifier(-module) utilization. I mean, for example, woofers need rather high power input, midrange drivers need modest power input, tweeters and super-tweeters need very small power input but need High S/N excellent THD+N values.

Even Greg Timbers uses "reasonable and budget" Pioneer Elite A-20 for compression drivers (super tweeters) in his extraordinary expensive multichannel stereo system with JBL Everest DD67000 which he himself designed and developed (ref. my post here).

I assume My DSP-based multichannel multi-SP-driver multi-amplifier fully active stereo audio project and the latest system setup thereof would be of your reference and interest.

 

[bluefuzz]

I'd much much rather have it all in one with the touch display than another app among the myriad of other household devices with their dedicated apps. And a volume knob is just.. Nice to have.

Obviously we all have our preferences. I just know I have never touched any of the buttons or controls on my NAD C658 in the four years I've owned it. And the screen, which can't be turned off, is just annoying sitting right under my tv ...

 

[Deleted member 48726]

Obviously we all have our preferences. I just know I have never touched any of the buttons or controls on my NAD C658 in the four years I've owned it. And the screen, which can't be turned off, is just annoying sitting right under my tv ...

A screen that can't be turned off would annoy me too. And you know it's not gonna last with it on all the time.

 

[SmartCross]

Thank you for all your ideas and suggestions!

Hi
I am just another random reader. It seems like you are interested in market definition of your project.

1. Since your product is DSP based digital-cross-over+amp for speaker DIYers, you have to consider the diverse possibilities. One may need a compact 2-way with small foot-print (60W each-channel/F3 60Hz is more than enough), while another could be seeking 4-way full-scale sounds. A single box to satisfy everyone could be difficult, especially in a relative small market.

2. 2* Hypex FA253 is around 1000 USD, if you want to take over their market share, you may need to reduce the price by 50% => ie. 500-650 USD for 6 channels DSP-to-speaker-outputs. It is possible, one TPA3221 amp board from Shenzhen Sanruisheng Technology retails for $20~30 USD. If you can arrange 3 TPA3221 or TPA325x with PFFB at the similar costs plus some amazing ASR scores, you are good. On the other hands, you should be aware the chip quality and sources for a long-term business. Comments on XY LM3886 amps from TomCHR are worth reading (how come a pack of $22 components could be sold for $11?)

Otherwise, if you are competing with Arylic quality, I don't think it would be easier for small business.

3. Software perspective, I only suggest to check RME. They are the No.1 in my mind. Whatever you use: PC, mac, iphone, ipad.... simply stable full-functions.

4. just another idea, since all the functions (DSP/multi-channels amp/EQ/gain-control....) are already built-in among most HT receivers, you can simply made your rocket-DSP/FIR-kernels to overwrite the functions of one receivers, and make one HT receiver is easy for Active-Cross-Over jobs (receive 2 channels from RCA/SPDIF/eARC.. and do the FIR things for every speaker unit), just like a MCU device for a car. If you can do this, I will buy 3 of them.

Good luck developing

> A single box to satisfy everyone could be difficult, especially in a relative small market.
Yes, that's the reason for the flexible output modules design.

> Software perspective, I only suggest to check RME
Will do, thanks!

> make one HT receiver is easy for Active-Cross-Over jobs
This sounds like a MiniDSP, if I understand correctly.

And that's exactly where you're wrong.
An HDMI input with (e)ARC would expand your buyer group to include a niche area that is very important for small manufacturers.
Take a look at these threads...
https://www.audiosciencereview.com/forum/index.php?threads/hdmi-and-dacs-not-evolving.38838/
https://www.audiosciencereview.com/...ity-budget-a-v-preamplifier-processors.28500/
And pay special attention to this post...
https://www.audiosciencereview.com/...-v-preamplifier-processors.28500/post-1363618

Thank you for suggestions and links. HDMI eARC seems be too complex for now, but we will definitely consider this option in future versions.

Fair enough.



Obviously SC would replace the Pi, but it would also need to be able to replace the NAD as well to be really attractive for my use case. At the very least it would need to able to do Spotify Connect and stream local content from a NAS with a decent interface (not DLNA). Preferably something akin to Moode/Volumio. I could easily live without Bluetooth and wifi but wired ethernet is also a must-have.



I'm not fixated on ultimate SOTA performance but it needs to perform at least as well as my current setup in that respect. Preferably better than fully transparent for 16bit content. I don't see much point in the device if you have to use external amps or streamers to achieve decent output quality or functionality.



That's very difficult to say but from my perspective probably not more than the used price of a NAD C658, a Motu M4, an RPI4 and 4 UCD180s ... ;-)



Which then makes the display and volume knob redundant. If there's going to be an app/web interface I don't see any point in a physical volume knob or display. They just add cost and complexity.

> I could easily live without Bluetooth and wifi but wired ethernet is also a must-have.

A USB-to-Ethernet dongle can be used on SC.

> not more than the used price of a NAD C658, a Motu M4, an RPI4 and 4 UCD180s
I think the final price will not be out of your range

A screen that can't be turned off would annoy me too. And you know it's not gonna last with it on all the time.

> Which then makes the display and volume knob redundant.
> A screen that can't be turned off would annoy me too.

You will have the option to automatically turn screen off (or darker) if unused.

If so, you would please aware of the concept and possible policy; "right person in right place" approach/arrangement(s) for multiple-amplifier(-module) utilization. I mean, for example, woofers need rather high power input, midrange drivers need modest power input, tweeters and super-tweeters need very small power input but need High S/N excellent THD+N values.

Even Greg Timbers uses "reasonable and budget" Pioneer Elite A-20 for compression drivers (super tweeters) in his extraordinary expensive multichannel stereo system with JBL Everest DD67000 which he himself designed and developed (ref. my post here).

I assume My DSP-based multichannel multi-SP-driver multi-amplifier fully active stereo audio project and the latest system setup thereof would be of your reference and interest.

I appreciate your effort on your project, and I will study your system design.

It's somewhat difficult to use "right person in right place" approach, because different amplifier modules expect different power supply voltages. We will ensure our amplifier modules have enough performance to drive tweeters.

Interesting project and I like the "hackability" aspect. I would, however, consider a line level subwoofer output with configurable low-pass as mandatory for a device like this. Otherwise, integrating a sub becomes a real hassle and essentially requires one with high level inputs and/or lots of tinkering. I also think that offering a smaller model with only 2 amp slots and 4 outputs would be a welcome addition in the future.

The "Amp" slots can have DAC installed in it to provide line level subwoofer output, with filters applied by DSP.

 

[dualazmak]

I appreciate your effort on your project, and I will study your system design.

It's somewhat difficult to use "right person in right place" approach, because different amplifier modules expect different power supply voltages. We will ensure our amplifier modules have enough performance to drive tweeters.

Thank you for your response to my points which I understand, and this would be always a kind of important "issue" if you (we) would seek for all-in-one unit including multiple amplifier-module in it.

Incase if we will be able to use your planned "SmartCross" as DSP-based multichannel pre-amplifer, I mean without using multiple amplifier-module in it, then we users will have flexibility of amplifier selection depending on our preferences; the all-in-one concept, however, will be abandoned in this case. Furthermore we users will pay much money for such "non-operating" amplifier modules in it. This is a kind of inevitable dillenma.

Even if we would accept such all-in-one unit, the "protective measures" including protection capacitors (like I implemented in my setup) for our treasure midrange tweeter super-tweeter, should/would be fully end-users' responsibility, and you may better to announce this precaution to (future) users of your planned SmartCross.

 

[SmartCross]

Thank you for your response to my points which I understand, and this would be always a kind of important "issue" if you (we) would seek for all-in-one unit including multiple amplifier-module in it.

Incase if we will be able to use your planned "SmartCross" as DSP-based multichannel pre-amplifer, I mean without using multiple amplifier-module in it, then we users will have flexibility of amplifier selection depending on our preferences; the all-in-one concept, however, will be abandoned in this case. Furthermore we users will pay much money for such "non-operating" amplifier modules in it. This is a kind of inevitable dillenma.

Even if we would accept such all-in-one unit, the "protective measures" including protection capacitors (like I implemented in my setup) for our treasure midrange tweeter super-tweeter, should/would be fully end-users' responsibility, and you may better to announce this precaution to (future) users of your planned SmartCross.

The modular design of output (see the structure diagram) means you can choose amplifier (TPA3255 / MA12070P) boards to provide speaker level output; or DAC boards (PCM5242 / CS43198) to provide XLR output, to be used with external amplifier.

We plan to make this customizable. When you purchase this, you can choose DAC or Amp modules (or a mixture of them) to match your desired configuration. If you like to have your external amplifier, you choose DAC; if you prefer all-in-one, you choose amplifier. The only thing you cannot do, is to have different amplifier modules on one machine (due to power supply voltage mismatch). DAC modules can be installed together with amplifier modules.

> Furthermore we users will pay much money for such "non-operating" amplifier modules in it.

This won't happen. You can simply choose not to install output modules on some slots at all.

> the "protective measures" including protection capacitors should/would be fully end-users' responsibility

Yes, that's an important point.

 

[dualazmak]

The modular design of output (see the structure diagram) means you can choose amplifier (TPA3255 / MA12070P) boards to provide speaker level output; or DAC boards (PCM5242 / CS43198) to provide XLR output, to be used with external amplifier.

We plan to make this customizable. When you purchase this, you can choose DAC or Amp modules (or a mixture of them) to match your desired configuration. If you like to have your external amplifier, you choose DAC; if you prefer all-in-one, you choose amplifier. The only thing you cannot do, is to have different amplifier modules on one machine (due to power supply voltage mismatch). DAC modules can be installed together with amplifier modules.

> Furthermore we users will pay much money for such "non-operating" amplifier modules in it.

This won't happen. You can simply choose not to install output modules on some slots at all.

> the "protective measures" including protection capacitors should/would be fully end-users' responsibility

Yes, that's an important point.

Thank you for your clarifications which I understood well.
Yes, it would be very nice that we will be able to choose "no amp-module" model.

 

[danstahl]

I've been toying with similar ideas. It would be great for a company to do for multi-channel processing/DAC what Buckeye did for amps.

From a UI/DSP perspective; I've created a rust-based "wrapper" for Camilla with what I think is a pretty simple UI which simulates the experience one would expect from a typical AVR. You are welcome to fork/contribute.

GitHub - danielhstahl/avprocessor

Contribute to danielhstahl/avprocessor development by creating an account on GitHub.

github.com

 

[Jungstar]

Super cool project. Dont take all the "manufacturer" BS too hard. I think there is missing a lot of "digital DSP" options out there. This might be out of scope for feedback. But could I optimize DSP filters for different positions in the room? Sometimes I wish, my speakers would ("turn", and) adjust balance and timing and adjust the DSP to match an offset place like out dining table that is to the side of the sweetspot. Ideally I would want my speakers to notice, where I am sitting and adjust with a small sensor, but "2 settings" would be cool

Regardless, I think there is missing digital alternatives to the MiniDSP. They have no competition.

 

[Jungstar]

Great idea. Actually I like my AVR too

well, AVRs are known for sounding, well, not very good....
See here how they struggle to get past 100 in SINAD

 

[Jungstar]

I've been toying with similar ideas. It would be great for a company to do for multi-channel processing/DAC what Buckeye did for amps.

From a UI/DSP perspective; I've created a rust-based "wrapper" for Camilla with what I think is a pretty simple UI which simulates the experience one would expect from a typical AVR. You are welcome to fork/contribute.

GitHub - danielhstahl/avprocessor

Contribute to danielhstahl/avprocessor development by creating an account on GitHub.

github.com

Did you make a longer post about this one? I think the Mini DSP has zero competition. I have so many old android phone and old laptops, that I could easily repurpose one as a streamer-DSP player.

 

[danstahl]

Did you make a longer post about this one? I think the Mini DSP has zero competition. I have so many old android phone and old laptops, that I could easily repurpose one as a streamer-DSP player.

I have not. I was considering using an Okto or Topping 8 channel to go along with a RPI, but I may end up just getting the Minidsp HTx. Unfortunately I've had a hard time testing the software since I don't have a USB DAC currently. It should work with HDMI out of a RPI but my processor isn't able to play anything over 4 channels from my RPI for some reason.

That said, my code should be basically useable if you want to fork it or tweak it. The UI is straightforward (at least IMO ) and generates camilla compliant configurations. The only potential gap I can think of is the need to map decoded channels to camilla input...right now I assume that a multichannel decoder magically knows that input 0 is the left speaker, for example. Up to 4 channels my RPI did map it correctly.