Please forgive me if this should be obvious, but it seems to me that the mod costs a MASSIVE amount of performance, a lot more than I think it should. I'm not familiar with the inner workings of this mod, but I will state my obervations.

On my rig with an i7 7700K and GTX 1060, I get over 900 fps on an empty superflat world at 32 chunk render distance with Sodium installed.

When this mod is introduced, the framerate is below 60, even on worlds without a dimension stack, also with Sodium installed. This is also at 32 chunk render distance.

Both of these results are at stock speeds.

These results seem really off. I was expecting a linear decrease in performance from adding dimensions to the dimension stack, but I was expecting such a performance degradation, and was definitely not expecting the mod to affect the framerate in a world without a dimension stack.

To me, it seems like there is something very wrong here, but please feel free to correct me.

I forgot to add, this is on minecraft version 1.16.1 with the experimental build of Sodium that is compatible with Immersive Portals.

What's the fps with only IP installed

It's pretty much the same. Sodium seems to be having no effect on performance when Immersive Portals is installed.

Is the FPS of IP lower than vanilla? If not, it's my sodium fork's issue.

And the fps stablizes after entering world 3 minutes. Fps may randomly fluctate. And computer temperature can also greatly affect FPS. Some software running in the background can also influence FPS.

I noticed the fluctuation. All of that went away after all chunks were loaded. That was expected behavior.

The framerate with Immersive Portals and Sodium is definitely significantly higher than vanilla.

However, on a superflat world without a dimension stack at 32 chunk render distance, the framerate increases by about 9 times when Immersive Portals is removed (from about 103 fps to around 920-950 fps while standing still, and from around 71 fps to 650 fps when moving). Something smells fishy here.

The expected behavior is that the empty superflat world without a dimension stack would have the same performance with or without Immersive Portals installed since no portals exist in the world.

Even if there was a portal in the world, like a dimension stack, the expected performance cost of this would be 30% of the framerate (I remember seeing that value somewhere, and I think it was the mod wiki). Cutting the framerate to a ninth of what it was before is a much higher performance cost than what is expected.

I know the game doesn't need to run at 900 fps to be enjoyed. I'm just running the game with an uncapped framerate to collect performance stats and noticed the huge performance difference between Minecraft + Sodium and Minecraft + Sodium + Immersive Portals.

Note: All my framerate results are with your experimental fork of Sodium installed. Great work on that by the way.

Basically just installing Immersive Portals causes a massive drop in performance. I believe other people had this issue, but nobody ever opened an issue on github about it.

If IP without Sodium is not slower than vanilla then it's not a bug. I am not interested in figuring out the weird interaction between IP and sodium.

I understand that. It's just that the framerate drops immensely just by installing the mod vs having Sodium by itself. I'm not trying to be a bother here. It's just weird that Sodium by itself performs great, but introducing Immersive Portals into the mix kills performance, even on a world with no portals.

I'll see if I can do some experimentation this week to see what's causing the performance deficit. I'm sure you have other things you'd rather be worrying about.