Chaos Corona for 3ds Max > [Max] Tutorials & Guides

OFFICIAL Q/A: sRGB, gamma, color pickers, linear workflow, ...

<< < (6/6)

TomG:
Remember that the ACES OT checkbox does NOT make the image into ACES colorspace - it only applies the ACES "sweetener curve" to the image. The output EXR is not then in ACES colorspace.

maru:
@VIZSET as Tom explained, the ACEST OT operator in the Corona tone mapping stack is something like an advanced version of a LUT, which just makes your image look nice.

Could you please explain what exactly is your problem? You are saving an image from the Corona VFB, loading it into Davinci Resolve, and you are getting different colors?

VIZSET:

--- Quote from: maru on 2022-11-28, 14:50:06 ---@VIZSET as Tom explained, the ACEST OT operator in the Corona tone mapping stack is something like an advanced version of a LUT, which just makes your image look nice.

Could you please explain what exactly is your problem? You are saving an image from the Corona VFB, loading it into Davinci Resolve, and you are getting different colors?

--- End quote ---

Hello!
Yes, problem with different colors when I save an exr from VFB and import it to DaVinci (when ACES OT on or off - it's doesn't matter). Importing the same exr to photoshop goes correctly.

maru:

--- Quote from: VIZSET on 2022-11-29, 09:36:43 ---
--- Quote from: maru on 2022-11-28, 14:50:06 ---@VIZSET as Tom explained, the ACEST OT operator in the Corona tone mapping stack is something like an advanced version of a LUT, which just makes your image look nice.

Could you please explain what exactly is your problem? You are saving an image from the Corona VFB, loading it into Davinci Resolve, and you are getting different colors?

--- End quote ---

Hello!
Yes, problem with different colors when I save an exr from VFB and import it to DaVinci (when ACES OT on or off - it's doesn't matter). Importing the same exr to photoshop goes correctly.

--- End quote ---

I did some quick checks:

Rendered a colorful scene similar to yours (strongly colored lights, objects, enviro).
Also added an invisible box casting shadow to see a dark gradient (this is great for testing sRGB vs 2.2 gamma - you can see the difference in shadows).

Opened the VFB as a reference.
Saved as 32-bit EXR.
Opened in Chaos Player with "color space" set to "default" - the shadows are visibly darker.
Opened in Photoshop - the shadows are visibly darker.
Opened in DaVinci Resolve 18.1.1 (the Fusion tab) - by default the image is clearly displayed in linear color space and gamma 1.0. Hence the huge change in the overall brightness and color appearance. I tried "fixing it" to look the same as in Chaos Player and PS so I clicked LUT at the top of the viewport and selected VFX IO > Linear to sRGB. I did not measure this, but in my opinion it looks exactly like the result from Chaos Player and PS.


What happens (I am not an expert so I could make some mistake here, but it should be correct):

Corona VFB:
3ds Max is not color managed.
Corona uses Adobe Wide RGB color space internally (you can change it in devel/experimental stuff rollout - https://support.chaos.com/hc/en-us/articles/4528523195537).
Then the image is displayed in the VFB with gamma 2.2 (this is specified in 3ds Max settings).

Chaos Player and PS:
The EXR is opened and displayed with sRGB gamma which is slightly different than gamma 2.2 - this is why there are small differences mostly between the darkest areas.

I also tried to get exactly the same result in DaVinci as in the Corona VFB, but I don't know DaVinci well enough. I *think* the way to do it would be:
- load the EXR
- set input color space to Adobe RGB (I am not sure if this is correct, I could not find this option)
- then set output gamma to 2.2 (not sRGB)

I hope this helps.

piotrus3333:
@VIZSET
just tell DaVinci that exr from Corona is lin sRGB.

@ Corona Team
how about calling a spade a spade? ACES OT is Corona's version of lin srgb to display srgb transform from ACES OCIO config. you slapped "aces" name on it after all. this would simply help to avoid confusion.

Navigation

[0] Message Index

[*] Previous page