Don't apologise for lack of knowledge - I would not be doing this if I wasn't discovering new things.
You have two things going on - the RF noise leaking into Hugo's ground plane, then creating havoc with the analog electronics. Random RF noise creates noise floor modulation (the noise pumps up and down with the audio signal) making it sound harder and bright - if it gets really bad, the treble will sound grainy and aggressive. The second problem is signal related audio frequency noise; this is at its worse at 0 when all the bits changeover from all 0's to all 1's (and 1's to all 0's) as the signal crosses through 0 and goes negative. This small current gets into the ground plane and corrupts the analogue signal, making the fundamental linearity worse - so a -120dB signal will now be say -119dB. These very small changes have the audible effect of confusing the brains ability to perceive depth of sound-stage. Depth of image is a hobby horse of mine, both from the listening POV and from the understanding POV (why is reproduced audio so bad at reproducing depth? what are the engineering factors?). Now I know that fundamental linearity is important to depth perception, but to what levels is it important, and are there other factors?
Anyhow, I digress. The Olimex will eliminate the LF noise currents. As to RF, it has a few pF of capacitive coupling, but optical will be perfect from this POV. But I think a few pF is good enough isolation. On my tests, I have a Olimex on the SD, and another USB card fed via optical. I can't hear a difference between them when the source is the same, so if you are hearing something, then I think perhaps something else is at play.
Thanks for your reply. So can we just replace Olimex with optical, as the later can play hires to get the same quality of music?