Why is UEFN physics so poor compared to UE physics ???

This isn’t a bug so I’ve added it here.

I was simply trying to move a cube and push other things that’s it. In UE its perfect looks realistic and is truly awesome. In UEFN its nothing like UE.

I’m not trying to be negative but i don’t understand why if its setup and working in UE and UEFN is somewhat built from UE then why cant UEFN just have UE physics. From what I’ve had a play with currently and comparing to UE I’m disappointed that a simple mechanic of a cube pushing some other shapes doesn’t work cleanly.

I like the idea but i think this physics has been built from the ground up and thus the problem is its not complete in comparison to UE which has had years of work.

Again I’m not trying to be negative but needed to share my opinion on this as its like UEFN is just getting worse with partial things added, bugs, instability ect

i feel you cannot make an experience that needs to have things controlled as the objects can randomly just fly about ect. Such a shame

Hi AshTag,

I’ve been using UEFN physics, in my recent experience they have been ‘smooth enough’ - but I’ve only used it in context of cubes/spheres and single-player.

When you mention ‘Other shapes’ what sort of collision shapes are you talking about?

As for why there is a difference, I imagine there are some raw performance limitations (*harshly glares sideways at nintendo switch 1*) also I wonder about the multiplayer-replication aspect of physics objects…

If you want to post some repro steps for a simple physics interaction setup that seems particularly broken, that might be helpful. (If only to serve as a benchmark for general future physics improvements)

I’ve been trying to create a platform that pushes tokens off (Disc Cylinder Shape)

In UE it works fine and looks so nice in UEFN idk if I’m doing stuff wrong but its deffo a step backwards. Are you saying cubes ect are better to use ?

Its like the collision from the moving platform and the disc cylinders isn’t reacting instantly and sometimes the objects get pushed under the grid plane.

Ill keep having a play about and do some vids showing the diff

I think the main issue currently is when pushing a physics object it can move down into the -z axis into another bpp/static mesh ? Ive made it more stable by tweaking settings and not moving the platform until after 5 seconds from game start seems to be more stable then.

I cannot enable the z axis blocker as it should move in the z but how do you stop something with physics from going into another mesh ? ive tried a blocking volume diff collision presets ect

The grid plane is relatively thin by default, so thickness might help for that. (If a disc gets temporarily pushed beyond the half-way point of something, it might favor the ‘void-side’ when it tries to readjust.)

As far as better/best to use - I’m just favoring the simple collision complexity shapes.

I did a quick setup and the discs (25cm tall) seem stable. (Blocks pushing the discs back and forth.)

Is it cool if i can send some snips/vids on discord ? if not i can upload to yt ?

I’d say upload then post link to vids here. I’m about to leave for the weekend, but I’ll be back on Monday.

1 Like

nws will have a play then over the weekend and share what i have by Monday.

ty for the help sir

1 Like