If you ever create that Audio level filter, would it be possible to add an option to adjust how accurate the bars are?
Meaning that you could set it, let’s say to 10 pixels, and the bars would go up and down in blocks of 10 pixels. That would prevent from producing fractions of LEDs when using a mask. Like you can see in this example:
Of course, that option would also be useful in the Audio Spectrum Visualization filter settings.
Also, there is a problem similar to the one reported above by @Namna concerning the Spectrum Color parameter.
I wouldn’t know how to code a spectrum analyzer or a VU-meter with Shadertoy and make it react to an actual music track in Shotcut. Plus, unless I’m mistaken, to use shaders, you need to be at least a bit familiar with a software like Natron. For the mere mortal, the Audio Spectrum Visualization filter is a lot easier to use in my opinion.
The new Audio Level Visualization has two types: Bar and Segment. The bar type is the same as the Spectrum Visualization Filter bar type. The Segment type is new. It divides the bar into one segment for each color specified. The segments to not have a gradient - they match the exact color specified by the user.
The “Segment Gap” setting sets the space (gap) between each segment. The “Thickness” setting modifies the width of each segment.
I also added the “Segment” type to the Spectrum Visualization Filter.
Peut-être ajouter si c’est possible un paramètre rotation. Aujourd’hui, si je veux obtenir l’effet suivant, je dois faire pivoter l’image d’origine avec SPR, puis mettre mon filtre audio level et enfin redresser l’image avec un second filtre SPR.
Maybe add a rotation parameter if possible.
Today, if I want to get the following effect, I have to rotate the original image with SPR, then put my audio level filter and finally straighten the image with a second SPR filter.