I have never been able to hear any difference using different inputs cables to the SM, but based on customer feedbacks, I took some measurements. The jitter distribution plots are all very similar, so instead I post the frequency spectrum of these where you can see differences. All use my reference output cable:
Cheap BNC input cable:
First spike at 1MHz is fairly small.
Plastic Toslink input cable:
The center peak is about 1/2 the height of the cheap BNC, but it is about twice as wide. Some new wiggles at 300kHz.
Belden 1694A input cable:
Belden is about the same as the Toslink.
Empirical Audio Reference BNC used as input cable:
This is significantly better. Center spike around 3MHz is not very tall and very thin. 1MHz spike seen in the other plots is gone. Noise floor much lower but there is a large spike at 6MHz that the other cables do not have.
It's clear from these that the jitter at higher frequencies is definitely affected by the quality of the input cable. The difference between the $20 cheap BNC and my $499 reference BNC cable is not significant though. I will do some listening tests to see if I can hear any difference.