Hi,
yes, if you build your freeDSP-aurora from scratch you need the xtag2 and an. USB-to-Serial converter as stated in the GettingStarted.pd on GitHub.
The Ready-to-run version does not need any additional programming hardware.
Raphael
Hi! I stumbled upon this project, and I'm very new to the DSP world. This might be a basic question but I couldn't find anyone asking it: What would be the best way to get samples in and out of this DSP? Suppose I'd want to make a little drum machine / sequencer out of this board. It doesn't seem like it has a whole bunch of ram.
Could I hook it up to a SoC with a bunch of ram, and have it stream digital signals in and out of there? Which interface would be best suited for that? It looks like it could do it over USB, that seems straightforward, but might not be the most reliable way?
So the idea would be: I'd record a sample from a microphone, stream it into the DSP, do some fancy effect on it, then stream it to the SoC which records it into RAM, allocating it into some sequencer slot. Then the SoC streams the sequencer track(s) back to the DSP which can do some post-processing and output it onto its line output.
Could I hook it up to a SoC with a bunch of ram, and have it stream digital signals in and out of there? Which interface would be best suited for that? It looks like it could do it over USB, that seems straightforward, but might not be the most reliable way?
So the idea would be: I'd record a sample from a microphone, stream it into the DSP, do some fancy effect on it, then stream it to the SoC which records it into RAM, allocating it into some sequencer slot. Then the SoC streams the sequencer track(s) back to the DSP which can do some post-processing and output it onto its line output.
Tinco, the best way would be to use a SoC that supports i2s (if stereo is sufficient) or tdm8 (if up to 8 channels are required) and connect it to the expansion header. If your SoC does not support it you can go via USB Audio but you have to make sure that your SoC (or it's OS) supports USB Audio Class 2.
mhelin: Yes that is possible. You have to add then the EEPROM for the DSP program storage.
mhelin: Yes that is possible. You have to add then the EEPROM for the DSP program storage.
Good, noticed you have also specified the SPI EEPROM part on the BOM. I don't know how it works but it seems you can program it using SigmaStudio even using I2C programmer as it's the 1452 which actually programs the SPI EEPROM.
That is actually what the USBi does.
Although, solved it a bit different and therefore, I don't need to populate the EEPROM: I store the dsp program on the SPI flash of the ESP32-WROOM module, because I want to be able to change the dsp program at runtime without additional hardware.
There is indeed a 6 channel version of the DAC but it's not much cheaper. It is usually more easily available though (from Digikey at least).
The ADC is 8 channel (also a 6 ch one is available with the same pinout) but you can drive your 2 ch input to all channels for better SNR, for an example left channel to ch 1, ch 3, ch 5 and ch 7 and right channel to ch 2, ch 4, ch 6 and ch 8 and the in ADAU1452 code sum the channels and divide by 4 so that you will get about 6 db better SNR than using a single channel. This is what some folks do with the AK4458 or ESS DACs. The is also a little bit better (SNR 121dB for single channel) AK5578 which is pin compatible, so you could get 127 dB SNR in 2 ch use from it.
Yes, no problem. Leave the unused channels unconnected or do like mhelin said regarding parallel use.
Last edited:
Are you going sell fully assembled PCB (for example at even elements Shop), or they only will be available for Kickstarter funders ?