lapleca ·
Hi people!!
I would like to explain the VST that I want to do. My idea is do the FFT of a signal (for example of 1024 points). After, I want to "apply" this FFT (or FFT coefficients) to this second signal, for getting that this second signal could have a FFT similar of the first FFT calculated.
For this, I think It's necessary to do a circular convolution, (or maybe linear convolution, depending of the points), and for this reason I have been asking in the Forum about the convolution.
This process is done for example in the some commercials VST, like the "Assimilator" VST (of TC-Electronics), or the "Ozone" VST (of Izopote) and I would like to try something like that.
I am consciuos that this idea could be impossible (at least in this moment), because it requiere a lot of processing.
I have seen in the Forum an example of FIR filter, doing for Sambean (very good job!), but this use the "method of windows" for doing the FIR, but in my case, I think it is not necessary to do this method to do a FIR filter using the FFT coefficients. However, I have tried to change this example, and I have removed the "module of window" and I have done all FFT coeffients (of the my first signal) multiply directly with the signal (my second signal) and the delays that there are in the example. In this sense, I have a doubt about this..what does exactly this example of Sambean (SimpleFir2) in the end, when all delays go to the same out? is it correct and why? are there a summary in the out?? I believe that each sample when you multiply it with another sample, the result is a different sample, isn't it?
Sorry for my english, it's very poor, I hope you understand me... 🙂
Pleaseeeee..., if someone could help me with some idea about this, I am very grateful. Any ideas about this?about the FIR of Sambean? or about my VST idea?and do you think it would be possible, with all this processing?
best regards!!thank you very much!!
Tania