How to use FFTConvolver with HISE?
-
Hello everyone!
This is my first post on the forum. I've been following the development of HISE for quite some time now and I really like everything that's going on here. I'm a musician, not a programmer, but thanks to the HISE community, I started learning programming and I like it.
My question may be trivial, but I don't understand how to use HiFi-LoFi/FFTConvolver in HISE.
Maybe someone could explain to me step by step how I can use it?
Thanks a lot! -
@Shelest well FFTConvolver is a C++ library - so whilst it is possible to integrate it into HISE - I suggest that you are better starting with something a little less ambitious...try starting with the existing Convolution reverb already in HISE...
-
The convolution engine is actually this library.
Just use the convolution reverb or the convolution node.
-
@Christoph-Hart said in How to use FFTConvolver with HISE?:
The convolution engine is actually this library.
Ah, so it's an integrated convolution engine... I thought it was an alternative to Intel IPP for convolution optimization. Unfortunately, I still haven't found a way to install Intel IPP on my system. It's not possible in my region. Maybe there are other alternatives? Or am I missing something?
Thank you very much. -
@Lindon Thanks for the advice! Of course, I'll practice on easier examples. I've been watching what's happening for a while. I write code on my own, study scriptnode modules, try to experiment, do something non-standard. Sometimes I get interesting results, which in turn for me, as a person far from programming, is a magical miracle! Hise is a wonderful creative tool that I'm increasingly drawn to, which I really like and all of you guys, thank you very much for this!
-
@Shelest the alternative for IPP (on Windows) is to just use the fallback FFT implementation which is the default if you don‘t enable IPP.
You leave a big chunk of performance on the table doing so but the functionality is the same.
-
@Christoph-Hart Yes, thank you, Christoph, I understand. Unfortunately, in this case, the convolution loads the processor very heavily. Never mind, maybe in time I will be able to install the Intel library in my system, but for now I will practice without the convolution.