I don't have a Femto clock version of the M1 and was expecting zero difference between using it and the AK100 as the transport. However, much like I had with the Hugo, the best results I got sonically were with the coax input. In the Hugo's case, it was with Soundaware's D100PRO. I have no idea why, unless one of the devices has an output that isn't bit perfect. If not that, the only thing I could speculate is that the Hugo and Mojo's electronics have to perform less timing recovery and the internal electronics end up generating less noise themselves.