Different exposure when rendering different parts of the image

[alpistinho]

How would it help to adjust both rather than just one?

I believe that after reversing the gamma for both sides they will be a multiplication away from each other.

So you would need to reverse the gamma for both sides, multiply one of them to make them equal and then apply the gamma correction again.

[FXtend]

Thanks for all the feedback
I think I've found a solution. There's a little bit of overlap between the halves so I'm thinking that if I render a few pixels within that overlap without gamma correction I should be able to set up a system of equations. I tried without this extra render but then I got eight equations and eight unknown variables whereas this method should give me only four and, well, I'm lazy
All I need to do is make sure that Auto Brightness is disabled, right?

[Dade]

All I need to do is make sure that Auto Brightness is disabled, right?

Yup.

[FarbigeWelt]

All I need to do is make sure that Auto Brightness is disabled, right?

This is right. I would render a preview, adjusting brightness manually in linear tonemapper during rendering and then stop and restart wirh final resolution.

I am very curious how your rerender will top your already fantastic work.

[FXtend]

This is right. I would render a preview, adjusting brightness manually in linear tonemapper during rendering and then stop and restart wirh final resolution.

Okay, I think you've sort of misunderstood my strategy. I don't want to rerender the full scene since I don't feel like waiting another month for my scene be finished. My idea is to figure out the gamma and gain by sampling both halves and a tiny render without gamma correction in order to set up a system of four equations that should give me the factor (gain?) and gamma value.

I am very curious how your rerender will top your already fantastic work.

Thanks I think I'll top it by having the same gamma on both halves and applying some lens flares and color correction and stuff

[B.Y.O.B.]

Have you tried what alpistinho suggested?
Reverse gamma correction is just gamma correcting with gamma = 1/2.2.

I believe that after reversing the gamma for both sides they will be a multiplication away from each other.

So you would need to reverse the gamma for both sides, multiply one of them to make them equal and then apply the gamma correction again.

[Racleborg]

I don't want to rerender the full scene since I don't feel like waiting another month

A month!? That's a very long time. Is this usual for some scene types?

[FXtend]

Have you tried what alpistinho suggested?
Reverse gamma correction is just gamma correcting with gamma = 1/2.2.

I believe that after reversing the gamma for both sides they will be a multiplication away from each other.

So you would need to reverse the gamma for both sides, multiply one of them to make them equal and then apply the gamma correction again.

That's what I'm trying to do but I need the gamma values and factors. That's what the equations are for.
I'm assuming Luxcore does gamma correction according to .
Maybe that's where I'm wrong because it still doesn't look good after solving the equations

[FXtend]

A month!? That's a very long time. Is this usual for some scene types?

That is very long indeed. Keep in mind that most of the sunlight has to go through both the fog and the water where it's refracted and dispersed and bounce around inside the pipe. I mean, interiors can be heavy but I think I've created something extreme
I don't know if this sort of render time should be considered acceptable still but that's all I know. Oh, and it's 4K.
Maybe I could have done with a little fewer samples...

[Racleborg]

Well, it looks great! So if it takes a month, then it takes a month