Remaking monolith

[d.healey]

@dustbro Oh that sounds like a very good idea. I'll try it and report back

[d.healey]

Didn't work unfortunately, I get a big message telling me it can't find the samples.

[Christoph Hart]

Hmm, actually the idea from @dustbro should work. The original samples must be in the same location and have the same name (however if you change the length of the samples, it gets unpredictable).

But I must admit I never had to go back from Monolith to sample, since the monolith conversion is usually the last step in the process.

[d.healey]

@christoph-hart Could this be a Linux path issue maybe?

[Christoph Hart]

It definitely looks fishy because it doesn't resolve the {PROJECT_FOLDER} wildcard to the HISE project folder, but interprets it as absolute path. If this is behaviour is Linux-only is not very likely.

[d.healey]

Can someone test on another OS please :)

[Christoph Hart]

I'll try to take a look tomorrow, but I am currently rewriting the sample map handling on the develop branch so this might get fixed as side effect anyway soon...

[Dan Korneff]

@d-healey check your chat

[d.healey]

@christoph-hart While you're tinkering with the sample map stuff could you add support for different number of crossfade groups and their tables? So if I load a sample map setup for 3 dynamics and then a sample map setup for 4 dynamics the XFade tables will be correct.

[Christoph Hart]

What you can do already is changing the crossfade tables via scripting. The only thing you need to be aware is that the sample map loading is asynchronous, but there is a „preload state change“ callback which can be used exactly for this:

1. Load the sample map
2. Setup a ScriptPanel to react to preload changes using ScriptPanel.setLoadingCallback(). This can also be used to display a custom loading progress bar.
3. in the „loading finished“ callback grab the number of RR groups, and change the crossfade tables to match the amount of layers using Synth.getTableProcessor(„Sampler“).setTablePoint(). IIRC you already had a script that creates the tables programmatically

[d.healey]

@christoph-hart Excellent, that should work nicely!