so I’m trying to make a trajectory for orbiting object but the output(blue dot) is like this note: I give the object initial velocity(assemblylinearvelocity) so it orbit. the white path is from trails, I’m using the trail to see how the trajectory path should be.
what I want is the trajectory is the same as the white path. also, I want to make sure that the blue dot is enough for any type of orbit so the orbit trajectory not like this
my setup
here is the script (child of the object)
--PartA is anchored
while wait(0.05) do
--gravity effect
local arah = (script.Parent.Position - script.Parent.Parent.PartA.Position).Unit --direction
local force = 10 * script.Parent.Parent.PartA:GetMass() * script.Parent:GetMass() / (script.Parent.Position - script.Parent.Parent.PartA.Position).Magnitude ^ 2 -- Im using newtonial gravity equation, 10 is the constant
script.Parent.AssemblyLinearVelocity -= arah * force * 0.05/script.Parent:GetMass()
--trajectory stuff
fakeVel = script.Parent.AssemblyLinearVelocity
fakePos = script.Parent.Position
workspace.inst:ClearAllChildren() --clear the old blue dot trajectory
for i = 0,50,1 do
local dis = (fakePos - script.Parent.Parent.PartA.Position).Unit
local force = 10 * script.Parent.Parent.PartA:GetMass() * script.Parent:GetMass() / (fakePos - script.Parent.Parent.PartA.Position).Magnitude ^ 2
fakeVel -= dis * force * 0.05/script.Parent:GetMass() -- 0.05 is the delta time
fakePos = fakePos + fakeVel * 0.05 -- 0.05 is the delta time
local a = Instance.new("Part",workspace.inst)
a.Position = fakePos
a.Size = Vector3.new(0.1,0.1,0.1)
a.CanCollide = false
a.Anchored = true
a.Material = "Neon"
a.Color = Color3.fromRGB(87, 255, 250)
a.Shape = "Ball"
end
end
The math looks right - I think your problem is that you are assuming that there is .05 time between each “step” of the orbiting object.
Your code is basically…
while true do
wait(.05)
local time1 = tick()
--do logic that takes x amount of time to actually compute & run on the server
local time2 = tick()
end
Total time will actually be .05 + (time2 - time1)
So your blue dots are estimating a path that is actually “faster” than the white orbiting object, and therefore the blue orbit path is closer to the center object.
I would do something like this instead…
Reference: RunService.Stepped
local RunService = game:GetService("RunService")
local RATE_PER_SECOND = .05
RunService.Stepped:Connect(function(time, step)
local increment = RATE_PER_SECOND * step
--your code, use "increment" instead of your .05 constant since it is the actual time duration
end)
Saw immediately what your problem was, you are using euler integration. Gotta update the velocity first then use that to update the position (dont use body movers) also fakeVel is +=
I didn’t know about Euler integration I’m still in high school. I don’t get what you mean about updating velocity first I think I have already done it on my script. if the fakeVel += the output look like this
this is my script
while wait(0.05) do
--gravity effect
time1 = tick()
local arah = (script.Parent.Position - script.Parent.Parent.PartA.Position).Unit --distance
local force = 10 * script.Parent.Parent.PartA:GetMass() * script.Parent:GetMass() / (script.Parent.Position - script.Parent.Parent.PartA.Position).Magnitude ^ 2
script.Parent.AssemblyLinearVelocity -= arah * force * 0.05/script.Parent:GetMass()
time2 = tick()
--trajectory stuff
fakeVel = script.Parent.AssemblyLinearVelocity
fakePos = script.Parent.Position
workspace.inst:ClearAllChildren()
for i = 0,50,1 do
local dis = (fakePos - script.Parent.Parent.PartA.Position).Unit
local force = 10 * script.Parent.Parent.PartA:GetMass() * script.Parent:GetMass() / (fakePos - script.Parent.Parent.PartA.Position).Magnitude ^ 2
fakeVel += dis * force * (0.05 + (time2-time1))/script.Parent:GetMass()
fakePos = fakePos + fakeVel * (0.05 + (time2-time1))
local a = Instance.new("Part",workspace.inst)
a.Position = fakePos
a.Size = Vector3.new(0.1,0.1,0.1)
a.CanCollide = false
a.Anchored = true
a.Material = "Neon"
a.Color = Color3.fromRGB(87, 255, 250)
a.Shape = "Ball"
end
end
I was able to do this using kepler motion, rather than calculating force/acceleration/velocity/position for each frame. Like in the video @CoderHusk posted, there will always be some noise when you process motion that way. This script can be used to determine orbital parameters beforehand, then calculate true anomaly + distance as a function of time. There will be no noise affecting the path that way.
I’m still in high school. I don’t get what you mean about updating velocity first I think I have already done it on my script. if the fakeVel += the output look like this
