[advice requested] raytracing with three-gpu-pathtracer

[Shrilliant]

Hello!

I was looking for a way to generate very large images in my three.js project that uses three-gpu-pathtracer to raytrace images and I stumbled on dekapng where I submitted a similar issue, and the creator linked me to your library.

I'm thinking that your generator moves the camera/renders a certain part of the canvas in order to stitch the PNG together. This works well for the normal THREE renderer, though I was wondering if there would need to be some sort of waiting period to wait for a raytraced image to converge in the scene before applying it to the output png and exporting it.

Do you have any guidance on accomplishing this with your library?

Thank you!

[taseenb]

You must have found a solution already. But if you wanted to try using this library you can change the renderNormal method (and renderDepth if you need it). It is a very simple method: you can replace the render method and then wait before calling finishRender.

https://github.com/taseenb/THREE.Highres/blob/612808977e3098aa207fa7ceda1b2748640c1af4/src/highres.js#L266

renderNormal (array, size) {
    this.rt = new THREE.WebGLRenderTarget(size.width, size.height)
    this.renderer.render(this.scene, this.camera, this.rt)

    // wait here somehow until happy

    this.finishRender(array, size)
}

You can then rebuild the library with npm build or use the class directly in your project. Also, note that this is quite old now, so I am not sure it still works with the latest three.