StateChunks don't update PARMLEARN and LFOLEARN when using ReaEQ

Meo-Ada Mespotine · 08-30-2020, 07:49 AM

I just noticed a strange behavior, when I add a LFOLEARN or PARMLEARN of ReaEQ to wet and bypass.

Let's say, I add a new instance of ReaEQ with factory default settings, I can add PARMLEARN and LFOLEARN to it and it will be added as, e.g.:

Code:

LFOLEARN 12:bypass 0 11 OSCMessage
PARMLEARN 12:bypass 0 3 OSCMessage

so the parameter-index is 12 for bypass.

Now, if I add another band, the parameter index for bypass moves to 15, which even can be added now as well. So if I make another PARMLEARN via UI, the statechunk can read the following way:

Code:

LFOLEARN 12:bypass 0 11 OSCMessage
PARMLEARN 12:bypass 0 3 OSCMessage
PARMLEARN 15:bypass 0 3 OSCMessage

I would expect, that the PARMLEARN-entry 12 would be updated to 15, if bypass changes from position 12 to 15 when I add a new band:

Code:

LFOLEARN 12:bypass 0 11 OSCMessage
PARMLEARN 15:bypass 0 3 OSCMessage

Is this expected and how can I find out, which one is now the real learned parameter for bypass and wet?
And is 12:bypass now used for parameter 12, which is not bypass anymore or is it used for 15:bypass as well, as it has :bypass in it's numbering?

The same problem goes for LFOLEARN.

Tested on Windows 7 Reaper 6.11x64

Meo-Ada Mespotine · 09-05-2020, 02:31 PM

Bump.

This is the last bit I need to know to finish up my ParameterModulation/Learn/Alias-functions and I don't want to work this based on false assumptions on my side on how Reaper does it currently...

Thanks in advance

TonE · 09-07-2020, 12:25 PM

Will PARMLEARN work soon correclty, hopefully? Cool, thanks for all your work.

But for bypass you can always count the number of parameters and usually bypass and wet are the last two always, wet=last parameter, bypass=second last. Not sure if this helps. Just wanted to mention. I find both parameters very useful, for quick and fun mappings, where precision does not count that much, because of phase issues or so.

Example: You bring a monster effect for just a short time, then remove it quickly again. Here nobody cares of phase issues, the effect is already monstreous.

Meo-Ada Mespotine · 09-07-2020, 12:41 PM

Yes, this is the last bit missing. Everything else works already.

The problem in here is that two or more can be associated with :bypass or :wet in StateChunks so I just need to know, if the parmindex is used or :wet :bypass as primary indicator of association.
It's important to know if I need to take care of this in my functions or rather the docs about them.

TonE · 09-13-2020, 01:55 AM

I would guess parmindex, as it is more general, or not?

Meo-Ada Mespotine · 09-13-2020, 07:10 AM

Problem is, I need to know. Guesses don't really help me. I want to avoid guessing wrong and therefore messing up the functions because of that... :/

Justin · 09-23-2020, 04:16 AM

Ah fixing this (those learn assignments are incorrect after changing the parameter count, until you save/reload the project at which point they are resolved again).

Meo-Ada Mespotine · 09-23-2020, 05:14 AM

Thnx.

Btw, I haven't checked, whether parameter linking has the same issue but I can imagine that it is the case over there as well.

Justin · 09-23-2020, 10:04 AM

Quote:

Originally Posted by Meo-Ada Mespotine

Thnx.

Btw, I haven't checked, whether parameter linking has the same issue but I can imagine that it is the case over there as well.

from what I can tell parameter linking doesn't have this issue exactly (wet/bypass links still work), but if you remove a band on the source plug-in and have parameter linking to a higher parameter it is not updated ... (fixing this!)

Meo-Ada Mespotine · 09-23-2020, 01:10 PM

Yeah! Thanks