Im sorry, but I need to learn the LWF method...

[dlparisi]

I guess what's confusing you is that the input gamma in the max settings isn't really a multiplier but a divider. Meaning if you want to get the same result in a texture without setting it globally you have to use .4545 multiplier (and you have to use it regardless in Vraycolor maps). You're right about the .255 thing though. But you still have to take it into account even though the sky has been fixed. But this is more up Lele's alley.

I don't believe I'm confused about this. The input gamma is just what it says it is - a gamma adjustment, it's neither multiplication or division, but applies a gamma curve to the incoming image. Setting the multiplier values in the material simply multiplies all of the image values by a given number (in your case .4545).

If you follow Lele's tutorial, you set the diffuse color to black and set the difuse spinner to 25.5 which effectively multiplies all of the diffuse map values by .255. With this method a gradient ramp that went from 0 to 1 would now go to 0 to .255. In contrast a gamma curve would keep the min/max values of the ramp at 0 and 1 but would adjust the mid grey point depending on the gamma value. A gamma curve is not the same thing as a value multiplier.

[Shimakaze]

Aaah I see what you're talking about now. Problem is you don't have any choice unless you know how to specify the exact gamma curve since there is no actual gamma control. So changing the output multiplier is the closest you get.
In 1.5 there is gamma correction built into the Vraycolor map though which would remove the need for the .4545 thing altogether. Haven't had time to do any tests yet though.
Just curious, how would you go by doing it? If you needed to use two textures, one which was in sRGB space and one that was in linear space (without gamma correcting in a different program first)?

[dlparisi]

Using straight Max, you could probably get a decent approximation by using the color map options under the output setting for the bitmap (either by plotting input and output points that correspond to a 2.2/.4545 gamma or guestimating a bezier handle on the first point). Otherwise you're best bet is to probably use some plug-in like the ColorCorrect map which has a spinner for a gamma value.

[DaForce]

Problem is there is MANY different ways to effectively use/implement LWF.

Just a matter of finding one that suits you and work your workflow and sticking to it.

In the end .. it doesnt really matter what you do as long as the final product looks good

[Shimakaze]

Problem is there is MANY different ways to effectively use/implement LWF.

Just a matter of finding one that suits you and work your workflow and sticking to it.

In the end .. it doesnt really matter what you do as long as the final product looks good

Well said

[Clifton Santiago]

Yes. Though I think it helps to learn the easiest way, get a handle on it, then explore the other options.

I think this is the reason that many people haven't learned it yet, because they lose the plot when reading the LWF posts, where LWF veterans are discussing only the finer details (which is always fun for us of course), rather than the basic concepts.

Uh, that sounded kind of pedagogical. Sorry.

Really though, at Siggraph and Vismasters this year, I assumed everyone was using LWF, but I discovered this was not the case.

[stevesideas]

I was completly stumped about the LWF method, so many people with slightly different views and thoughts on the process. The funny thing is, the way I was carrying out my lighting workflow before wasnt that far from it anyway.

So as long as I follow the steps in previous posts it works out fine. I can't see any reason for using any other variant on the method now that my images look the way I want them to.

[Shimakaze]

Yea it doesn't need to be difficult in most cases. Maybe we should have a video tutorial and sticky it

[stevesideas]

I think maybe a video or even a scene file for people to download that has it setup already...maybe the video can refer to it.

[ene.xis]

http://www.chaosgroup.com/forum/phpB...&highlight=lwf


u can t imagine how much knowledge is at the reach of one click at this forum!

[Shimakaze]

That one doesn't explain "why" you do those things though. Nor why you should/should not use gamma correction for the Vray output.
But better then nothing I guess.

[stevesideas]

Ok, can i just chuck something else in..

Basically, when Im using my digital SLR outside and I need a shot where the ground and the sky are both exposed correctly, I basically use 2 exposures from the raw image and composite them together in photoshop.

Would I do the same with vray. If im creating an interior scene and expose for inside, it would correctly be overexposed for anything outside..so basically I just lower the exposure or vray camera settings and then composite the 2 together in photoshop. Is this correct ?

[Clifton Santiago]

Personally, when I'm photographing, I try to get the exposure correct of my subject matter, and all the rest looks turns out however. If I'm focusing on a figure against the sky, I want them to be properly exposed, even if the sky is blownout, and I'll adjust my light metering mode accordingly, usually spot, or centre-weighted for that.

I would only do the double exposure composite thing if you want to make them look like older arch viz renderings, before we had proper exposure control. In an interior photograph you would normally expect to see the exterior overexposed, with the interior exposed correctly, unless one was focusing on an exterior garden or something.

But, in the end, its whatever looks best, rather than what is photographically correct. Or, as the case may be, what the client thinks looks best...

[instinct]

well...that's what tonemapping is all about, no ?

[Clifton Santiago]

I use tonemapping to make things look more photo-real, rather than less. And really, with VrayPhysicalCamera the role of tonemapping is reduced.