Snap Aux Iterates twice at Ensemble Open, why?

Hey Everyone,

I just realized that the Snap Auxiliary and Iteration modules both self-iterate twice upon ensemble initialization. Does anyone know why? It's weird...

 

snap value and self iterate produces only one event at initialization:

i think you have something connected at the inputs which produces additional events?

ciao herw

 

Well maybe it's just a bug, because look here. There's nothing else in the ensemble, I created a new one just to test this phenomenon that was occurring in my project.

I reset Event Watcher before initializing and also tested it with just a Counter module hooked up to a numerical readout.

 

couldn't confirm

 

freaky.

 

A General System Reset behaves normal on my machine too. The idx outport of the snap array module initializes only once. I remembered that a Sample Rate Reset does a double initialization! But testing this on the Snap idx outport gives me the same answer as before: initialization once only...

edit:
OK, double initialization on Ensemble Load Reset here!
---
All those different Initializations are very confusing.
If i sum up, you have to distinguish between:
- Ens Load
- General System Reset (Power Button / "Run Audio"))
- Sample Rate Reset
- Switch Reset
- Debug Mode in Structure

They all behave different to each other in some points (Debug Mode not tested by me).
That is really ill! Until now i thought Ens Load and General System Reset behave the same... So, i think it is really time for a FULL description of the Initialization(s). I thought, initialization is relatively easy, not your enemy but your friend.
So, maybe you found a forgotten "bug" on the iterating modules or that behaviour is essential for some logical reasons. But this double init doesn´t make it easier. Maybe it´s only unnessesary, or maybe it really can lead into different working structures (EnsLoad vs. GSR) and with this into problems.

 

I did some more testings with my ACEW (Audio Core Event Watcher).
The ACEW is in progress for version 2 (on UL) but not finished yet. But i think it logically works correctly. First, the ACEW shows in this simple example, that the Eventwatcher is NOT capable of showing Init Events of step 2 of the GSR-initialization algorythm (pdf downloaded?) ! Else use the Init Debugger for that.
In the first example, the Ens Load Initialization of a snap array of size 4 is shown (idx Out).
In the second example the GSR Initialization of the same structure.

The idx outport is connected to inport 1 and a const=1 to inport 2 of ACEW and Eventwatcher. In both examples the Eventwatchwer doesn´t show the first (synchronous) GSR init event, which is marked in orange in ACEW (row "t state"). This is the moment, when the constant gets initialised. But that is not the point. Look into the first example with double init: The second init appears at a time of 577 samples after Ens Load (marked as "Switch Reset"). This means, that the Ensemble is still running when the second init "appears"! 577samples after Ens start is very much! I ask me, why that second init doesn´t come immediately after the first one (=0 samples)

 

Hey QuietSchoolboy. What you are saying sounds interesting, but I don't actually understand fully what you are talking about. I don't know how to program in Reaktor Core yet, so I don't follow you when you talk about these deeper initialization signals.

From what you have tested, what do you think the problem is that all my Snapshot Value modules are firing twice?

 

hello lemming,
there´s no need to program in core to understand. i´m just talking about different initialization behaviours according to the init "trigger" shown in my first post.

The snapshot modules fires only twice when loading the testensemble. just restarting the reaktor audioengine only fires once here.

have a look for the old audio core event watcher in the UL. there are some restrictions and bugs in some cases, but it could open your eyes when testing initialization behaviour.
its intention was to test events in audio core cells, but its also very useful for checking primary events. just feed the core part inputs with primary events.

mark

 

After further testing, i noticed, that the Snap Value Array idx output don´t initialize twice in the old Reaktor 5.6.2 on Ensemble Startup... (but in actual 5.8.0)

What are they doing?