MLT3 / PAM3 signal generator on the demo board is working great now, I forgot to reset the OSERDESE2 during startup which was screwing up my emphasis.
The top blue waveform is straight off the board and is unrealistically sharp and has just a tiny bit of overshoot on the transition from the -1 to 0 state. The red waveform is after a 180 MHz FIR LPF roughly simulating channel loss and magnetics response. I'll add a discrete SMA LPF to the system for the real demo and it should be go…