Thoughts on the Raspberry Pi 400 as a psychophysics platform?

Hi everyone, I am looking into setting up a new psychophysics system for my lab. I am curious about the recently-released all-in-one Raspberry Pi:

I am interested in using it to set up cheap, easily replicable (and transportable) testing setups. I see some people at least are using PsychoPy on previous Raspberry Pi models, but I thought I would see if anyone knows of any major obstacles I am heading toward? Is compatibility on Raspberry Pi similar to that on an ordinary computer?

I can see the amount of RAM being a possible stumbling block. I tried something similar previously with Octave/Psychtoolbox on the MintBox Mini 2:

The two issues I ran into were: i) insufficient memory (for the specific study, I think for most studies it would have been fine), and ii) if the device breaks (which mine did temporarily) there’s no obvious way to recover the data from it. The relative advantage of the raspberry pi would be that the SD memory card is swappable.