Sorry for the late reply. Busy times here.
We're following this issue closely and I'd like to share with you some of our findings and measures being taken.
The issue is related to the audio driver and how it manages the power of the CPU when the audio codec is in use. There is a unwanted current discharge at each buffering cycle that is "contaminating" the input circuits. Because the audio buffering is done at every 128 frames and, as we run at 48kHz sampling rate, the frequency of the current discharges is at 375Hz. Harmonics of this frequency - 187.5Hz, 750Hz, etc - also appear. If you enable "development mode" in the GUI - type "enable_dev_mode()" at Chrome's Developer Console - and change the buffer to 256 samples you can notice the noise frequency change.
As many interference issues, the overall result is a sum up of multiple factors that range from circuit component deviations to the actual set-up and cables hook up.
The USB connection, although not related the audio, has a strong impact on the noise depending on your computer USB implementation and also on the way that the computer and the Duo are hooked up to the electrical mains. When using some laptops, disconnecting it from the AC power and running on the battery completely changes the noise behavior.
We were working on the improvement of the audio driver in order to improve the power handling but right now, this task is temporarily paused. We are using a forked version of the Linux kernel called sunxi and, because of it's forked nature, it proved to be unworkable due to the never ending "fix here, break there" cycle.
That said, there is an ongoing effort to move to the mainline Linux kernel (the official version) and, as soon as the migration is done, the audio driver improvement will start moving again.
In the meantime, here are some measures to be taken in order to minimize the issue.
1 - The "contaminated" circuit is in the pre-amplifiers analog section. Changing the Gain stages - Menu/Volumes and Gains - makes a huge difference so the users that require a setting different than Low Gain should try to find the best balance between the Gain Stage, the Fine Gain and also on the internal volumes of the Pedalboards.
2 - Due to the location of the circuits in the PCB, the Input 1 is more vulnerable than Input 2. That said, Input 2 should be preferred for users that require higher gain settings
3 - Try different USB connections and ports between the Duo and the computer
I hope this info helps.
I'd like to remind that we consider audio quality to be one of our top priorities and I'd like to assure all users that we are constantly working towards offering the best sounding device we are able to.
Best regards to all members of the MOD Community