I am wanting to color the particles either with a ramp or a random function. How would I go about this?

Hi DBraun, thanks for putting together this great collection.

I'd like to contribute a component that facilitates controlling parameters using the Novation Launch Control MIDI controller. I realize that there already is a component on the same controller, but I think it can use some improvement.

The component I put together allows to:

• Map min, mid, max values for 16 sliders on 8 user pages
• Add smoothing on slider movement (similar to ease-out)
• Pick up button presses for 8 buttons on 8 user pages
• Send color commands to 8 buttons on 8 user pages

Configuring the component is done by replacing pre-populated labels and values in a DAT table for ease of use. The component can easily be adapted to other Novation Controllers.

How do you feel about this? Would this fit the purpose of this repository?

I've updated Cd to Color and that fixes the material, but can't seem to figure out the issue with sop_to_gradient_sdf_voxels

ERROR: /project1/Boids/sop_to_gradient_sdf_voxels/phong1GLSLPixel:129: 'mTD2DImageOutputs' : undeclared identifier

any tips to get that working in current TD stable?

Referenced in xoio_People_Collection_OBJ

Thanks for this great Noise COMP. Since you are using index into the Pixel Format menu of glslmulti1, in the latest 099 build I am using this is not giving the correct result. The index will change whenever Derivative adds new entries to the menu, so its not the most reliable method. To be sure of it you can use a Table DAT to export the string token for the menu entry, something which will be consistent (mostly) when Derivative adds more the the menu. I have attached a simple example of how to do that, but it still needs to be hooked up to your custom parameter menu so it switches.

Also for consideration, maybe 16-bit float is best default because 32-bit floats really hit performance hard and when people start appending extra TOP filters after this COMP they may not notice or realize that the pixel format is the reason their performance is getting killed.

curl_noise_4D_TableExport.zip

There are similar errors in kohonen.tox and kuwahara.tox, the shader complaing about using "sample" after a type declaration. Using 099 on Windows 10 and Nvidia GPU

kohonen.txt
kuwahara.txt

Thanks for porting PixelFlow into TouchDesigner - seems like an incredibly useful tool.

I'm getting a pixel shader compiling error in the sobel glsl, it says:
ERROR: 0:9: '{' : syntax error: syntax error

Thanks!

OpticalFlow.1.toe.zip

Hi David

I'm having a go at porting your voxelized lookup feature to another simulation. Basically just some regular particles in a compute shader that can collide with each other. And I noticed something strange in the get_bins component.

As you can see in the screenshot, there is a voxel that claims to be overflown with particles (in the texture its the lower left pixel), yet in the texture holding the positions, there is not a single particle in that voxel area. I'm bounding the positions to be slightly inside the bounds of the voxel world measurements. The issue doesn't happen immediately, but builds up over the course of the first 30 seconds or so.

Do you think this could be a bug of the bin counting shader?