For Beta testers on Win 10 x64 machines, I put the .zip file here on Dropbox:
No manual yet. I hope to push out a demo video tonight that shows some of the features. Blur is a spectral manipulator so you can get some beautiful sounds with gentle settings and some edgy sounds with the more extreme settings.
This is a Beta version. The feature set, controls, UI panel size, etc., may change as the module matures. One feature I would like to add is a phase cleanup button that cleans out residual phase offset accumulation that can occur after spectral manipulations. The residual buildup can leave the signal slightly malformed even after returning all knobs to their ‘do nothing’ positions. Using the menu to init the module clears the phase buffers but also resets all the controls so that is not a good option. Phase cleanup will appear in next release.
Just a note: if you are planning on releasing this via the Rack Plugin Manager at some point, you must remove the dist directory and its content from the repository. This also means removing it from the repository history. I would not add binary builds to your repository, but instead post them here in this thread.
So I tested it a bit today, it looks very promising I have a few ideas/thoughts about it, shall I share them here or github, or somewhere else perhaps?
I think that stereo might be useful here. A stereo input would be comfortable, but having a stereo output with random panning of the audio could be very cool.
The quantization feature is great, but maybe it will also be possible to have it quantize to 5ths and octaves.
It would also be more comfortable to see the semitones in the popup information.
The same with the buffer size, maybe having seconds could be more comfortable.
in addition to the frame dropper, it would be nice to have a button to take it from 0 to 100. Something like freeze.
And a global mix control can also be interesting, especially with a dedicated cv input.
What about the design, by the way? Will it change eventually, or will it stay more or less the same?
Anyway, I’m excited to see how it will end up being, it looks like a great sound design tool. Thanks again!
Stereo - yes, can do. Will have to trust the opinions of the beta testers and early adopters though since I am deaf in one ear and cannot hear stereo.
Quantization - good idea to move to popup menu and add 5ths and octaves as option
Buffer Size - was originally thinking that being able to dial in a very specific buffer length would be useful but it turns out to be not so much. Agree that adding that as a menu option is a better way to go.
Frame Dropper - yup, makes sense to have a “Freeze” button.
Global Mix - good idea, but a bit tricky to implement. Since the output is spectrally modified and generated from "past’ frames, mixing with current dry incoming audio tends to produce undesired (or maybe desired ) effects and distortions. Worth a try though to see what happens. Can simulate that in VCV now by sending the Blur input and Blur output to a mixer. If traditional mixing does not work well there may be some other point in the processing chain dry/wet mixing would make sense.
The UI will likely change. Once the controls get sorted as to what is on the panel and what is in the popup menu I would like to try collaborating with a UI designer to produce the final panel.
Ok cool, sounds good!
About the quantization and buffer size, What I meant by pop up menu is the parameter tooltips so the information that pops up when you hover above a certain parameter. It would be great to see the buffer size in seconds and the pitch in semitones.
Thanks again!
Regarding buffer size units – I like seeing it in # of input samples (ie, powers of 2) as you currently have it since that’s the standard for FFT-based FX (eg Unfiltered Audio Spec Ops). If you only show one I vote for powers of 2 to ensure consistency with general FFT terminology.
Note that converting to seconds depends on sample rate, which is external to the module, whereas input samples is purely internal to the module. If the menu item defaulted to seconds as the unit then this could lead to a confusing situation where you change your sample rate and it looks like the default setting in the module changed, but it actually didn’t. Probably this is why samples is generally chosen as the default unit.
To clarify, the Buffer Length knob on the panel sets the number of seconds worth of historical audio saved in the history buffer. The history buffer is like a delay line but instead of holding a list of individual samples it holds a list of FFT frames.The actual number of frames in the history buffer for a given number of seconds (in the range 0 to 10 seconds) can vary based on the FFT Frame size, the oversample factor, and the current sampling rate.
The FFT Frame Size from the popup menu, sets the number of samples in an individual FFT frame.
I think “Buffer Length” is not a good label for the knob that controls the length of the history buffer …
I was thinking of removing the Buffer Length knob and adding a menu item for it instead. The units for that menu item would be seconds. The units for the FFT Size menu item would remain as number of samples by powers of 2.
) effects and distortions. Worth a try though to see what happens. Can simulate that in VCV now by sending the Blur input and Blur output to a mixer. If traditional mixing does not work well there may be some other point in the processing chain dry/wet mixing would make sense.