Consistency Sorting 2D Issue

,

Hi,

Before I spend a lot of time trying to make a simple project for this issue, I just want to see if it’s a known issue.

2021.2.0f1

The issue happens only when useConsistancySorting is ticked.

I know this information I am giving isn’t very helpful, but it might give a clue to an issue already reported.

I have a scene with a lot of physics2D objects. Here are the issues:

  • If I let the game just sit there, the frame rate will decrease to very low over 15 minutes.
  • If I disable parts of the scene, the frame rate will never decrease.
  • If I disable the parts of the scene, it will decrease in less than 5 minutes to 3 fps. If I then disable a few parts of the scene while in play mode, the fps will jump back up.

I know that really doesn’t give any clues, but again, it might help with an already reported issue?

If not, I will upload a reproduction scene and post the case #.

Here is the profiler. I doubt this gives any insights, but just doing it in case it does.

Feel free to submit a simple reproduction scene and post the case number here.

I suspect this is fixed by your recent bug report and is indirectly related to the same issue though here.

Only way to know is for you to submit the project and I can confirm.

It is similar but in my instance, I am using multithreading along with sorting.

I spent a good hour or two trying to get this going. For certain reasons I couldn’t get a bug report. I will just see if its still an issue once those bugs are pushed.

However, if you can still grab that water 2D glitch I reported, that project has the same issue but you have to wait 40-50min for the fps to drop.

No, I didn’t mean your symptoms in the report; I mean the subsequent fix and the reason why things were going wrong.

Well I might be able to do that at some point. I’m actually on annual leave so I shouldn’t really be doing anything. :wink:

1 Like

I will add that it’s very unusual for it to open happen like this because it’s running at least 50 times a second so taking 50 minutes then suddenly taking off is very odd. I would look at something other than just pure timing. Look at the 2D physics area in the profiler too. It shows a whole bunch of metrics there including contact count.

I will also say that in the profiler, the physics is already taking a HUGE amount of time well before that (around the 33ms mark).

Noted! Thanks for pointing this out :slight_smile: