When using the ibis model to simulate the SN65LVDS104 part in our system using HyperLynx, when simulating at fast-strong the output differential CM drops from 1.2V (what we see with slow-weak or nominal) to around 0.82V and the VID amplitude drops more than expected (should be higher with Fast/Strong, not a reflection issue we allow the signal to settle out). This doesn't happen with other TI LVDS ibis models. Is there an issue with the ibis model for this part? Or can we expect this as real world behavior.