Yeah i meant the scenes is behaving as if we have many small lightsources. I thought it was clear by how I worded it. Apologies if it wasn't. What I want to say is that much of the visible surfaces only gets hit by indirect light, as the tress and foilage are in the path. Add refraction from the water and we have missing or wrong information about many pixels.
Again I meant for this scene it is, not generally. And yes you could optimize it for example by adding more light into the scene with ghostlights, using a well lit hdri or by rotating the sun behind the character. But it would create different shadows and a different look. I agree I could have worded that clearer for the sake of giving advice to the OP. I would still use a post-denoiser at the last iterations in this scene tho.
This is where I am dissappointed. Maybe you should take your own advice.
Excerpt from nvidea manual (for reference standard monitor luminance is around 150 cd / m²):
The nominal luminance is a hint to Iray Photoreal on what is considered a “reasonable” luminance level when viewing the scene. This luminance level is used internally to tune the firefly filter and error estimate. When the nominal luminance value is set to 0, Iray Photoreal will estimate the nominal luminance value from the tonemapper settings. If a user application applies its own tonemapping without using the built-in tonemappers, it is strongly advised to provide a nominal luminance.
Recommendations: For visualization, a reasonable nominal luminance would be the luminance value of a white color that maps to half the maximum brightness of the intended display device. For quantitative architectural daylight simulations (for example, calculating an irradiance buffer or irradiance probes), a reasonable nominal luminance could be the luminance of white paper under average day light.
On what we can agree is that the render settings in this package are garbage, lol.
Wording is important, especially when you're talking to someone still quite new to the matter, it's no wonder that you find so many wrong information about rendering on the internet.
Sorry if my post sounds offensive, but I got slightly triggered by the amount of "wrong" (half true) informations. Think about it how someone new would take what you posted.
This is where I am dissappointed. Maybe you should take your own advice.
Welp, after looking for some nice reference I could quote, I have to admit that there isn't any to be found regarding Daz and iray... which, tbh I could have expected thinking about it now...
The problem is that your quote actually tells nothing about luminance, and it's also doing the same you did with your post, it's giving false/half true information in regards to render settings. They don't do anyone a favor by providing a monitor reference.
You
can use this reference if you use the standard render settings, but then again, what's the point of using nominal luminance with the standard settings?
The answer is simple: it's not that simple!
Different scenes have different render settings, different render settings need different nominal luminance settings. Taking the render settings of the OP scene as an example (which is why I mentioned 10k in the first place), it has an exposure value of 9.5, which means that a fuckton of light will enter the camera, which also means you'll need a "high" nominal luminance value (if you use it), otherwise the scene will look washed out, because all the highlights will get filtered out.
Maybe 10k is overkill for the scene, I don't know, I don't have the scene, but let me give you an example of a render I'm working on right now:
With this in mind, I could actually use lower values here, 5k for example, maybe even 1k and it would basically look the same, but imagine I'd used a sweaty body/reflective skin: all the shiny little details would simply not show up, because it would get filtered out (see how the shine on her arm disappearded?).
So like I said, 10k isn't really high, it depends on the scene and render settings, and let me be honest, the standard render settings are BS. You need to tweak them if you want to get good (better than average) results, and if you do that, you'll need "high" nominal luminance.
Hope this is understandable and clears things up.
