• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Introducing DSPi | A powerful, user friendly and open source DSP for less than a cup of coffee

- Three-channel stereo (mid/side)

Something like this?: https://foobar.hyv.fi/?view=foo_dsp_fsurround
That would be perfect for home cinema users with many speakers! (to interpolate between surrounds and back surrounds or atmos 4->6 in larger installations )
Yes, quite similar to that.

I haven’t been able to find it, is a Linkwitz transformation implemented?
Linkwitz transform will available within the next update, which should be available later this evening or tomorrow.

Would it be possible to use the Pico's onboard Bluetooth module (w modell) to receive an audio stream and map it as additional i2s input?
Yes, this is planned.
 
I think I eventually want to build a bass management cross-over using this. There’s a glaring gap in the market for affordable products that do this.

I actually just bought a Wondom adau1701 unit (which I saw @Weeb Labs commenting on a few years ago) for this same purpose, but a Pico + DSPi should perform better presumably? No need to use ADCs, and you can choose better DACs.

I especially want to be able to send a digital out to my active monitors and analog outs to my subwoofers, which would remove an ADC-DAC step to the monitors.

There’s so many possibilities with this. The Pico W will allow wireless adjustments which will be super useful.

anyways excellent work and progress. This is gonna have a big impact for DIY projects once boards start coming out.
 
Just a small thing.
Could you please add to the System Statistics the number of bits?
Sample rate and number of bits would be also quite useful on the DSPi Console.
 
On my Raspberry, the card works perfectly.
With aplay -L, I see several devices.

hw:CARD=DSPi,DEV=0
Weeb Labs DSPi, USB Audio
Direct hardware device without any conversions
plughw:CARD=DSPi,DEV=0
Weeb Labs DSPi, USB Audio
Hardware device with all software conversions

Both hw: and plughw: work.
Interesting. I've tried mine on 4 different Pis using Raspberry Pi Os (Trixie and Bookworm,) Moode Audio, and PicorePlayer/LMS. They all recognize the device as a USB audio device. I've used .asound.src and alsa.config to set it as the default. I've tried a variety of USB cables including OTG. Everything seems to work except the actual sound output. When I hook it up to a Window$ machine it plays music. I've tried a couple of DACs. One flashes its LED when it's not seeing the selected source. That's the error indication I get beside the lack of sound. If it didn't work with Window$ (11), I'd think I'd made a mistake in the wiring.
 
Just a small thing.
Could you please add to the System Statistics the number of bits?
Sample rate and number of bits would be also quite useful on the DSPi Console.
Yes, I will add bit depth and sample rate states to the statistics.
 
For planned I2S input coming later:
My AVR stops clocking to the DACs when source is paused.
Is it possible to have volume muted and unmuted (ramp up and down) when grabbing AVRs internal I2S? And of course be able to recover from stopped clocks?
 
Last edited:
I've been following this from the start but not been able to join in on the fun because of me having to stress out about moving houses, but now when that's more or less done and I also got temporary layoff (yey?) I will have more time :)
Not 100% sure on how the hardware side works, but I want DACs with RCA output and wonder if anyone got any good suggestions for when this is implemented in the project? Hard to find any measurements on those random boards on Aliexpress, but anyone knows if this PCM5102 or UDA1334A board would be good? Preferably with a SINAD over maybe 90dB with a flat response, but again no measurements what I can find. Will this be more or less plug and play as well? Does the Pico output power for stuff like this or do I have to hijack that somehow from the USB cable?

As I understand hardware knobs and buttons will be a thing as well, so would something like this digital rotary encoder work?
And when additional USB inputs will be available, how are those connected to the Pico?
 
Is this amount of Pico 2 CPU fluctuation normal? [brief video to demonstrate]
Yes, that's perfectly normal. The current utilization calculation is fairly simple and takes place within the USB ISR. As the SOF feedback runs and system latency varies slightly, this is represented as fluctuating core utilization, even though the actual DSP workload has not changed.

System utilization will soon be overhauled with a resource budget in place of the current wall clock approach.
 
Yes, that's perfectly normal. The current utilization calculation is fairly simple and takes place within the USB ISR. As the SOF feedback runs and system latency varies slightly, this is represented as fluctuating core utilization, even though the actual DSP workload has not changed.

System utilization will soon be overhauled with a resource budget in place of the current wall clock approach.
Thank you. Normally I don't see the cpu level change.

Looking forward to the Windows updates.
 
I soldered together a voltage divider and have a working coax output. It works perfectly with my Window$ machines. I've tried using it as a USB sound card with my Raspberry Pis and had no luck. The Pi recognizes it as an ALSA device . aplay-l shows it as an available card and you can select it in alsamixer. Using VU meter simulator software or Camilladsp I can see that the Pi music player app is creating output. Is this working as designed? I understand it if it is. I was just hoping to be able to use it with my Pis.
Problem solved. Once again it's user error. I wondered about using only the non-signal side of the voltage divider as the ground. For the heck of it I connected a ground terminal from the Pico to the ground on the RCA jack along with the the non signal side of the voltage divider. It works great. Now I wonder why the former configuration worked with USB output from a Window$ machine. I'm a history major whose formal electronics education came from high school physics over 60 years ago. I've fooled around with simple electronics projects and assembled small kits successfully, but I'm nowhere near an EE or electrician.
 
Problem solved. Once again it's user error. I wondered about using only the non-signal side of the voltage divider as the ground. For the heck of it I connected a ground terminal from the Pico to the ground on the RCA jack along with the the non signal side of the voltage divider. It works great. Now I wonder why the former configuration worked with USB output from a Window$ machine. I'm a history major whose formal electronics education came from high school physics over 60 years ago. I've fooled around with simple electronics projects and assembled small kits successfully, but I'm nowhere near an EE or electrician.
The most likely explanation here is that your Windows machine's USB GND is connected both to the Pico's GND and to the power supply's earth ground, which may have been completing the signal path to your DAC. The RPi power supply was almost certainly double insulated and so there was no earth ground connection.
 
Problem solved. Once again it's user error. I wondered about using only the non-signal side of the voltage divider as the ground. For the heck of it I connected a ground terminal from the Pico to the ground on the RCA jack along with the the non signal side of the voltage divider. It works great. Now I wonder why the former configuration worked with USB output from a Window$ machine. I'm a history major whose formal electronics education came from high school physics over 60 years ago. I've fooled around with simple electronics projects and assembled small kits successfully, but I'm nowhere near an EE or electrician.
Are either of you using something like this?

1773175332522.png


The optical works but the coax does not. Similarly, I'm basically a paint by numbers electronics person so any advice is welcome and appreciated.
 
I've been following this from the start but not been able to join in on the fun because of me having to stress out about moving houses, but now when that's more or less done and I also got temporary layoff (yey?) I will have more time :)
Not 100% sure on how the hardware side works, but I want DACs with RCA output and wonder if anyone got any good suggestions for when this is implemented in the project? Hard to find any measurements on those random boards on Aliexpress, but anyone knows if this PCM5102 or UDA1334A board would be good? Preferably with a SINAD over maybe 90dB with a flat response, but again no measurements what I can find. Will this be more or less plug and play as well? Does the Pico output power for stuff like this or do I have to hijack that somehow from the USB cable?

As I understand hardware knobs and buttons will be a thing as well, so would something like this digital rotary encoder work?
And when additional USB inputs will be available, how are those connected to the Pico?
The PCM5102 is quite a reasonable option but it is important to bear in mind that many of the AliExpress boards fail to achieve the IC's specified performance ceiling due to suboptimal layout and filters.

When I make available custom DSPi boards, I will be making use of one or more TAC5212 codecs which have previously been implemented on boards such as this one. These are inexpensive (~€10), require few peripheral components and offer both well behaved DACs, ADCs and a headphone amplifier.
 
Last edited:
Are either of you using something like this?

View attachment 516726

The optical works but the coax does not. Similarly, I'm basically a paint by numbers electronics person so any advice is welcome and appreciated.

Have you tried disconnecting the +5V power supply?Because this connection (SPDIF RCA) doesn't need it.
I have the same card, I'll try the same thing as you tomorrow if I have time.
 
For my part, I'm currently developing a small PCB with 4 outputs and one optical SPIDF input... and I've never designed a PCB in my life... and even if something much more complex is probably planned by Troy or others, this DSPi project is stimulating! ;)
 
Back
Top Bottom