Massive performance Difference between studio and live game

When running all my tests in studio for my ocean simulation, the performance I’m getting is quite good.

about .9ms for the entire cascade update code, a massive improvement from my previous 4ms results. Now this is wonderful, but when going to test this in a live game…

The entire cascade update is taking roughly 10 times longer to compute, both versions are running on the client so there shouldn’t be such a massive difference… The only thing I can think of that would be doing this is my use of native code generation, but my previous version of the ocean simulation logic did NOT have this performance difference. I have no idea what could be causing this, any ideas?

1 Like

a 10x increase is quite a lot, plausibly explained by native code generation, but that does seem fishy. another factor i would like to bring up is how the client and studio has/had a different number of threads used for parallel scripting, but recently this hasn’t been the case https://devforum.roblox.com/t/win11-desktop-player-showing-raised-8-taskscheduler-workers-instead-of-previous-3-cap/4357901

yea It’s super weird, what make it more confusing is that a previous place of mine, with about the same logic is running smoothly between studio and client, it’s only this new place here where I’ve noticed this huge change, I’m really not sure what the issues is here, I thought maybe it was because I’m using connect parallel for my cascade update but switching to to the normal connect didn’t change anything, I’ve pretty much ran out of choices as far as fixing this goes.

1 Like

studio allocates more cores to running it than Roblox player

This project looks clean. How long did it take to make. What are your future plans for Massive performance Difference between studio and live game? I’m so excited to know. I would like to see more videos of this. Keep up, hero!