Well if it is not a case of carbon tracking causing a cross-fire situation in the cap and rotor and you say that you can see half of the primary signal going away on your Oscope then you will have to back track it from there. Check at the power stage and see if it is going away and then check at the engine speed sensor and see what that says. That is tough, that is one of those "chicken or the egg" stories. If you are monitoring the engine speed sensor and you rev past 1500 and the frequency drops off by half why did it do it? If the engine slows down, regardless of the reason, you will see the frequency of the engine speed signal reduce. It's back to the chicken and the egg. Did the frequency slow down because of a defect in the sensor or did it slow down because something else is screwy and the sensor is just reading exactly what happened. I'd have to agree with the other guys in that I would be suspicious of the driver in the ECU that drives the iginition power stage. What is there in that circuit though that would allow it to cut out just one or two cylinders, you would think that if it was in the primary circuit that it would effect all cylinders not just one or two in a specified RPM range. I know that in the LH ecu that when it goes into injector shut off, it shuts off all four and you would think that it would be the same on the EZK side or at the very least be more random. Have you any tester units that you can plug in and see what it does, it'd be nice to have a spare power stage in this instance.

Let us know what you come up with.
Mark