I made reflections in reflections… for 5 mins… accidently… so yes, let’s start from begin!
i just was trying to render transparent part, and accidently created reflections in reflections. Yes.
You’re supposed to be running your code in Heartbeat for stuff like this. I have a feeling it isn’t based on the FPS in your video. If you’re gonna go down this approach I’d also suggest you use blurs of pixels instead of individual frames.
This code runs full on client, and how im know HeartBeat is for servers, so im using RenderStepped.
Wdym? “I’d also suggest you use blurs of pixels instead of individual frames.” what is that blurs?
No, from my personal experience doing that exact same thing on the client using HeartBeat made it faster. Idk why…
These blurs are when you get a blurry circle as an image and use it as your pixel. Effectively allowing your resolution to be a lot lower while also not getting that retro look. You can then overlap this with surface guis blur they get to make it seem more natural.
One could also take that knowledge that surface guis create blur at low studs per pixel to create a pixelation effect.
I think one GUi element has limit of 64.000 elements in it, so maybe you can do like 6 Frames each has 64.000 elements (pixels), never tried, but you can
You’re kinda capped at the quality you’ll get with rays. You are min-maxing really GUI elements and ray casts. Some quick fixes you could do now is, cache your Color3’s and use Vector3’s for color logic. Do not render directly to pixels as frames, but use the method I told you with the blurred circle. If you want quality at a crisp performance then I would recommend using UIGradient compression.
It’s not inaccurate, it’s a very old method where since you can store 8 colors per gradient you make a 1x8 ui gradient thing. this lowers gui usage by 800%. You’re still sampling all pixels it’s just you are rendering it more efficiently.
So, it’s efficiently? Im gonna rewrite my whole raytracer, and maybe use this method to render, but don’t know, i think is more easier to use frames, when u can work with single pixel and etc.