I remember complaints in the Reaktor Forum about the lack of an EventOutput on AudioCoreCells.
I don´t think that this is really needed in R5, but for sure it is the interface to event-based instruments that have been built before Core had its release.
In general terms: If you want to do some work in AudioCore and trigger an event-based instrument with the results, this is what you need.
It looks a little complicate and reliable distinctions of cases will offer simpler solutions; but if you want an AudioCoreCell-Output to behave EXACTLY like a PrimaryLevel-EventOutput there is no other way to do it without using even more CPU-power.
It consists of two parts: One inside the AudioCoreCell and an "Adapter" outside. To connect them you need THREE CellOutputs (after having decided if it is float or integer) and THREE virtual cables. If you open the ensemble structure and the CoreCell you can easily see how to do this. Just connect the corresponding I/Os.
The example shows the "worst case scenario" being managed: Single Events with the same value are to be transported to the Cell´s outside; sometimes at slow rate (or even not at all), sometimes at sample rate.
By the way: Having this you won´t necessarily need my AudioCoreSonde in order to find out what is happening inside the cell anymore.
You can have a single direct bus from anywhere inside the Cell to the AudioCoreCellEventOutput and feed Mr. Cristopher List´s EventWatcher with the signal which is to be analysed through it.
Turn down the SampleRate if the ensemble crashes Reaktor.
...feel a little like opening Pandora´s box...
May the force be with you, Gerald List.