I am wondering whether the Steinberg AU wrapper (that is an optional part of the VST SDK) takes a different approach than Juce: Is the Juce approach a “real native” AU implementation? Or is this also a wrapper approach?
Besides, I am wondering whether there might be sound implications, taking the one route or the other.
Although I don’t really know the Steinberg AU wrapper, I’m pretty sure that JUCE doesn’t use this. Instead, JUCE itself acts as a wrapper unifying the different Plugin APIs (VST, VST3, AU, AAX).
What technical explanation would you have in mind for such a difference? Or what kind of difference would you expect? I can’t see how the plugin format should make a difference in sound if the same math is applied to the same float sample buffers (which is the common approach all plugin formats use) – it’s only the way how these buffers and some control data are communicated with the host application that makes the big difference between the formats.
If I would understand the whole implementation chain of the wrapping and could confirm that this is a lossless-mapping (at least in one direction), then I would agree: there shouldn’t be a difference. But I don’t and that’s why I have to ask