What is wrong with my voxelization script?

Help and Feedback Scripting Support
, , , ,
1

I tried to make a destruction script, but it became absurdly laggy. By that i mean freezes for no reason, even with low part count. Freezes with high (200+) and low (4) part count are almost same, i don’t know what to do.
Here’s the script. It’s only a concept right now, so it’s just a legacy script.

local function subdivideByCubes(part: BasePart, cubeSize: Vector3, resizeToPart:boolean) :(Folder)
	
	local handler = Instance.new("Part")
	handler.CFrame = part.CFrame
	handler.Size = part.Size
	handler.Parent = part.Parent
	part.Parent = nil
	local cubesize = cubeSize or Vector3.new(1, 1, 1)
	local xCount = math.round(part.Size.X / cubesize.X)
	local yCount = math.floor(part.Size.Y / cubesize.Y)
	local zCount = math.round(part.Size.Z / cubesize.Z)
	if xCount ==0 or yCount ==0 or zCount == 0 then
		warn("One of the sides is way too small for the cubes with size:", cubeSize) 
			warn("Part size:", part.Size)
	end
	
	local cube = Instance.new("Part")
	local startPosition = part.Position - ((part.CFrame.XVector * (part.Size.X / 2 - cubesize.X / 2)) + (part.CFrame.YVector * (part.Size.Y / 2 - cubesize.Y / 2)) + (part.CFrame.ZVector * (part.Size.Z / 2 - cubesize.Z / 2)))
	local folder = Instance.new("Folder")
	
	for Index = 1, (xCount * yCount * zCount) do
		cube.Size = cubesize
		cube.Anchored = true
		cube.Position = startPosition + 
			(part.CFrame.XVector * (Index % xCount) * cubesize.X) + 
			(part.CFrame.YVector * (math.floor(Index / xCount) % yCount) * cubesize.Y) + 
			(part.CFrame.ZVector *( math.floor(Index / (xCount * yCount))) * cubesize.Z)
		
		cube.Parent = folder
		cube = Instance.new("Part")
	end
	script:Destroy()
	folder.Parent = workspace
	handler:Destroy()
	part:Destroy()
	return folder
end

task.wait(5)

subdivideByCubes(script.Parent, Vector3.new(4, 4, 4), false)
2

I’m not sure if you’ve solved this by now, but it is because your for loop increases in complexity on the order of O(n^3) with the size of parts, if you make the part twice as long on just a single axis, that loop will take twice as long to complete. You can imagine its worse if you also scale up on the other two axis.

I have actually been developing a system to make these kinds of destructions optimized and simple, see the thread here if you are still interested:

1 Like
3

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.