• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. 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!

Sigma ADAU1701 Review (DSP Board)

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
46,928
Likes
266,148
Location
Seattle Area
This is a review and detailed measurements of the Sigma ADAU1701-2In4Out DSP board from @3eaudio . It was kindly loaned to me by a local member. I think it costs $18 for the main board (?) The version I have has the tone control and USB interface options as well.

This is what the whole setup looks like:
ADAU1701 Sigma DSP Review digital crossover evaluation board.jpg


I was pleasantly surprised to find balanced outputs on such a low cost solution. I used the tone control board to get access to analog inputs for testing. I did not want to install drivers and such (which seemed pretty messy from the documentation) so did not use the USB dongle. As such, I had to power it using my lab supply which I did with 12 volts (range is 7 to 12 volts).

Here is a closer shot of the main board:
ADAU1701 Sigma DSP Review digital crossover evaluation board USB.jpg


The bluetooth board add-on seems quite messy but I assume this is supplied by someone else and not Sigma.

Sigma ADAU1701 Measurements
There is no way to defeat the tone controls that I could see. And unfortunately there is no center neutral position forcing me to spend quite a bit of time to get more or less flat frequency response:
ADAU1701 Sigma DSP Measurements frequency response.png


Once there, I left them as is and ran the rest of the tests starting with our dashboard:

ADAU1701 Sigma DSP Measurements.png


Anything higher than 1.1 volt would cause clipping even though datasheet states that is the max level. Unfortunately that limit means that we can't get our nominal 4 volts out, nor can you use a standard desktop DAC that outputs 2 volts as is. You need to lower the volume on it to avoid clipping.

Overall performance -- considering that is the combination of ADC and DAC -- is not bad for something like this. You are distortion limited. If you lower the output even more, it gets a bit better:
ADAU1701 Sigma DSP Measurements distortion versus level.png


Dynamic range barely makes it past 16 bits:
ADAU1701 Sigma DSP Measurements dynamic range.png


Finally, a comparison against a commercial solution, the minidsp DDRC-88:

ADAU1701 Sigma DSP Measurements distortion and noise versus level.png


Definitely not in the same class.

Conclusions
I can see the appeal of this board: two channel input and four channels out makes for a nice crossover solution for active speaker design. Performance is good enough for DIY/hobby solution. The audio elitist in me wants something better :) but you can decide for yourself. I am not going to pass judgement on it one way or the other.

------------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
 
Last edited:
@amirm
If you use the USB dongle, I suppose the ADC is going to be skipped - correct? If so, it's a pity not to take measurements that way as well, since for a 4-way xover solution this would be the way to go. Datasheet report:
2 ADCs: SNR of 100 dB, THD + N of −83 dB
4 DACs: SNR of 104 dB, THD + N of −90 dB
Which could improve the SINAD quite a bit. The datasheets / application notes talks also about a linux driver, which means that this with a Raspberry could act as a complete source (player, xover and DAC) for a 4way custom speaker system.
Seeing the issues with drivers & co, what about setting up a virtual machine just for these kind of tests?
BTW, thanks again for this website and your work!
 
@amirm
If you use the USB dongle, I suppose the ADC is going to be skipped - correct? If so, it's a pity not to take measurements that way as well, since for a 4-way xover solution this would be the way to go. Datasheet report:
2 ADCs: SNR of 100 dB, THD + N of −83 dB
4 DACs: SNR of 104 dB, THD + N of −90 dB
Which could improve the SINAD quite a bit. The datasheets / application notes talks also about a linux driver, which means that this with a Raspberry could act as a complete source (player, xover and DAC) for a 4way custom speaker system.
Seeing the issues with drivers & co, what about setting up a virtual machine just for these kind of tests?
BTW, thanks again for this website and your work!


USBi only for programming ?

Only digital input is I2S ?

http://www.3e-audio.com/wp-content/uploads/2017/01/DS_EDSP-1701-24-A_Rev1.0.pdf

1612162183278.png
 
OK, now I see... This is a module likely coming from China (I found on AliExpress which seems exactly the one reviewed, code of the board EDSP-1701-24-A) and not the official dev module, so it all will depend on the board we are getting. At first glance the chip seems capable to directly act as a 4 way x-over. I need to dig on this a bit more...

-edit-
ha, now I see the block diagram of the chip - only 2x DAC are on board, the chip is going to act as a digital 8x IN and digital 8x OUT without DACs, so back to the tested board.
 
OK, now I see... This is a module likely coming from China (I found on AliExpress which seems exactly the one reviewed, code of the board EDSP-1701-24-A) and not the official dev module, so it all will depend on the board we are getting. At first glance the chip seems capable to directly act as a 4 way x-over. I need to dig on this a bit more...

-edit-
ha, now I see the block diagram of the chip - only 2x DAC are on board, the chip is going to act as a digital 8x IN and digital 8x OUT without DACs, so back to the tested board.
No. There are 2xADC 4xDAC on board
 
Nice work @amirm, thanks. Interesting to see another set of measurements of an ADAU1701 board. This one appears to be fine apart from the indefeasible tone controls.

You didn't happen to measure distortion/noise per frequency did you?
 
I have one of these, it's not an official sigma dsp dev board (sigma is just the family of chips) but a 3E audio product. When you program the chip it's a free form dsp that runs a kind of flow chart with programing elements so you can make the tone controls do anything you want (or just ignore them). The adc/dac are not very good as they are on the same chip as the dsp but i2s is exposed on the headers so you can use other adc/dac chips. I added another stereo dac to get 6 outputs.
 
Why would anyone attach tone controls to a DSP unit? It doesn't make any sense.
The tone controls are implemented in the default DSP program. The pots feed low resolution ADCs in the DSP chip that control parameters in the program. If you use SigmaStudio and the dongle to reprogram it you can use them as you see fit, subject to the limitations of the DSP.
 
Last edited:
HIFIBERRY DAC+ DSP another nice board and supports both sigma studio programming and simple PEQ bands loading
 
Might as well get a MiniDSP 2x4 HD, higher cost, but it's a much better unit. Thank you @amirm
 
I hope SMSL, Topping, Loxjie, Schiit take notice of this!

The usual 2-channel DACs are almost a commodity. USB/SPDIF/Toslink, remote and optional XLR, Headphone amp ob Bluetooth. Performance is flawless, so why pay more than the cheapest one with SINAD over 100dB?

Meanwhile I need a high performance DAC that can handle stereo monitors +2...4 subwoofers using MSO. And then suddenly the only option left is miniDSP. I like miniDSP, but unless you pay a LOT more than a good stereo DAC for a miniDSP SHD then miniDSP is quite poor on the DAC quality side. So I dream about more alternatives to miniDSP 2x4 HD that have SINAD at least 100dB. XLR connections are a big bonus and the 3-band Equalizer is AWSOME!!! (see Shiit Loki and JDS Labs Subjective3 for examples, but all digital DAC version would be even better and cheaper).
 
Back
Top Bottom