Does LWF affect Reflection Values?

[braddewald]

Because vray uses a color value (0-255) for reflection strength, does an imported material (with no maps) need to be adjusted for a scene with LWF? Or is a value of 128 always 50% reflectivity? Thanks!

[BrianKitts]

kinda two different questions you have there....

 

yes an imported material would need a gamma adjustement to be physically acurate. That's why the max settings have the two selection buttons for affect color editor, and affect maps.

 

is 126 always 50% = NO....... not in LWF.

 

In LWF your color editor is not an even gradient from black to white, it's an exponential curve (i think)..... see the attached screen shot of the color selecters in max from both LWF and w/o LWF setups. Notice how LWF pushes the blacks into a smaller region, the mid grey point actually shifts up. The green line is the same grey value as the non LWF 126 which keys in somewhere around 50 approximately

 

That's my loose answer but the idea is there. I'm sure Jeff could give your a more specific technical answer....but hopefully you get the idea. After a few years now of LWF I feel I've got a pretty decent understanding behind it all, but the more I've learned..... the more I quit caring and have resorted back to doing whatever I want to get the outcome I desire. No one's ever measuring my images to hit a specific color target..... so I don't get too caught up in the details any more.

Edited February 25, 2011 by BrianKitts

[Crazy Homeless Guy]

So what would happen if I set a custom reflection map/color shader and gave it an override gamma of 1.0? When you use a normal map, displacement map, or bump map you should always override the gamma and set it to 1.0. I wonder if the same can be said for reflections?

[Crazy Homeless Guy]

I think it works.

 

The horizontal plane on the right has a gradient reflection map with the gamma overrode to 1.0. Notice how the reflection starts at the intial point of contact of the box, and appears to fade linearly.

 

The plane on the right is using a gamma of 2.2. Notice how the reflection doesn't really become noticeable right away. This would be because of the curve applied to it because of the 2.2 gamma.

 

EDIT...

 

I tried to do the displace test, which is how this was explained to me, but mine did not come out perfectly. This may be because I work on a high gaumet monitor, but I am just guessing at that,

 

So, the furthest object has a gradient applied to it with a override gamma of 1.0. In theory it should be a straight line.

 

The nearest simply has a gradient applied to it using the default gamma. Notice how it looks like the inverse of a 2.2 gamma curve.

 

This info/concept comes for a session that Pierre-Felix Breton from Autodesk taught at AU.

 

EDIT 2...

 

OK, further testing. In the first displace image I was using a gradient ramp created in Photoshop. In the second displace image I was using a gradient ramp created in Max.

 

It appear that the gradient ramp created in Max actually ignores the gamma correction, and behaves in true linear fashion. The first object has the unmodified gradient ramp applied to it, and notice it is as straight as an arrow.

 

I think this means that is you used a gradient ramp for reflections, the answer would be no. Still doesn't shed light on the default reflection parameters, though that should be an easy test to do to see what happens.

 

The second object, furthest from the camera, has a gamma'd 2.2 override applied to the modifier. Notice how it takes on the 2.2 curve.

 

SO, ...I guess it depends on how you create the reflection as to whether or not it is modified by working in LWF.

Edited February 25, 2011 by Crazy Homeless Guy

[CliveG]

Boy I'd love to understand what you're saying here... but I'm like Homer simpsons dog. "blah blah blah LWF blah blah blah Gamma 2,2" (this is a compliment not a criticism by the way).

 

I believe Travis and Brian have, in the past, both expressed a similar sentiment that they're a bit "over" the whole LWF thing and just want to do what it takes to get the right end result. Has this all now just got too complicated. The goal for me is to simply have light travel more (realistically) through my interior scenes, requiring less (false) supplemental lighting....

 

LWF sort of achieves this but really - hasn't there got to be an easier way than all of this adjustment when it comes to bump, reflection, displacement and material maps. Is this not a yawning opportunity for those savvy plug-in people to give us a plug in that sorts this all out for us?

 

Or is it that there seems to be no - one correct answer and methodology when it comes to LWF (or none that I've gleaned from all of these sites) that this plug-in hasn't materialised yet?

[Crazy Homeless Guy]

So, I have cont'd to look at this. I have been curious about how the gamma works on the displacement since hearing Pierre's segment on making sure the normal and bump are set at 1, instead of 2.2.

 

From what I can tell the reflection behaves the same as the displace, normal, and bump. This makes sense once you think about it. These are all items that would be taken into more or less during the math calculations of a rendering, and not a direct feed to what is being displayed on the screen, or being loaded as a bitmap. Which is what the gamma settings in the preferences control.

 

Meaning that our brick texture will need to be adjusted because it was created in the sRGB or whatever color space. It will need its gamma adjusted to work within the gamma corrected environment.

 

The color inputed at a level that deals with the direct math of the reflection does not need adjusted to compensate for gamma. Hence, 127.5 in the reflection slot = a 50% reflection in Vray.

 

The attachment shows a Gradient Ramp set to the solid color of 127 gives a base reflection. I say this because my test above showed the gradient ramp to be the most precise in creating a precise line which is what should happen in a gamma 1.0 environment.

 

The second series of boxes show the Vray grayscale of reflection set to 127. The result not discernible from the gradient ramp.

 

The third reflection in the image shows a solid color set in Photoshop at a 127 solid color and then brought into Max with the gamma overrode and set to 1.0. Notice it behave exactly the same as the first 2.

 

The fourth image shows the same solid color at 127 bitmap when the gamma is left to the native setting, which is 2.2 in this case. It behaves as you would expect a bitmap to behave.

 

I am not sure how all this relates to choosing color in a LWF space. I use the CoolPicker plugin exclusively to set colors in Max, which helps in seeing the color being selected, but it still could use some more development in flipping between gammas to give a better sense of what is going on.

 

Or something like that.

Edited February 27, 2011 by Crazy Homeless Guy

[Crazy Homeless Guy]

LWF sort of achieves this but really - hasn't there got to be an easier way than all of this adjustment when it comes to bump, reflection, displacement and material maps. Is this not a yawning opportunity for those savvy plug-in people to give us a plug in that sorts this all out for us?

 

Or is it that there seems to be no - one correct answer and methodology when it comes to LWF (or none that I've gleaned from all of these sites) that this plug-in hasn't materialised yet?

 

It can be confusing but I think it is easiest when you simply think about why the gamma 2.2 exists in the first place. It has to do with a very outdated reason that unfortunately is so embedded in our pictures and movies that there is no feasible way to remove it.

 

As I understand it, the original tube televisions needed an image to be adjusted to a gamma 2.2 for the tube technology to be able to display images. This technology was used since the advent of the tube tv, and really only became old with the introduction of LCD monitors that did not need a gamma adjustment of 2.2 to display a picture. You could simply tell it what to display.

 

But, since the gamma 2.2 was relevant for so long there is no way to remove it from the process entirely unless you work in a closed environment and never interact directly with the real day to day world of imagery.

 

I am gleaming this information off past reading and studying so I can't remember the exact specific's or where I gained the info but more than likely it was either Jeff Mottle or Pierre Felix. They are both extremely knowledgeable when it comes to how color works in the digital environment. The only thing I can do is disseminate what I understand from their passions.

[CliveG]

Sorry I shouldn't have bumped into the thread, with a side discussion. It's a serious question Brad posed and I'm grateful that you guys have the time and the understanding to chase the answers.

 

I'm pretty much up to speed with the reasons we have this "problem" but I guess I'm sort of feeling that by now - LCD type monitors are ubiquitous and we shouldn't still be wrestling with what, I would consider a legacy issue. Our software providers should be onto this.

 

Can you summarise your conclusions then: Am I correct to say that it would seem that we'll get more accurate map performance with the max maps rather than PS (though there is a gamma adjustment in PS so I guess we'd need to know how yours is set?) and that we should maintain gamma 1.0 on all of these type of maps then? Or.....?

[Crazy Homeless Guy]

It is perfectly fine to use a bitmap as a reflection map, it simply needs to be overrode with a gamma of 1.0 when loaded. Otherwise you will not have as easy of a time predicting how it is going to behave.

 

As far as accuracy, I think it would be very negligible so I wouldn't worry about it. The majority of time I actually prefer to use hand painted reflection maps because I feel I have more control over what they are doing. Perhaps I simply have not invested enough time into procedurals.

 

____

 

 

I wonder how software like Maxwell handles this? ....Are they gamma correcting everything for you behind the scenes? If so, it sounds like that is what you are looking for.

 

I am not sure grandfathered is the right word as it implies it is something that is being included to keep functionality. The use of gamma at 2.2 is so embedded into our daily lives that it is not something that can be removed.

 

Think of every screen you have in your house, TV, Monitor, Cell Phone, Camera LCD, Tablet, digital picture frame, iPod, etc. are all more or less set up for a gamma of 2.2.

 

The vast majority of digital images have a gamma of 2.2 burnt into them. These are everything from the textures in your library, the images coming out of digital cameras, the movies produced by Hollywood, and so on.

 

It is simply to embedded in our daily life for it to be removed. And simply to few of us care about its existence in the first place.

 

As far as removing gamma from the workflow all together. I think you can, but I also think you would find yourself more or less in a shell that wasn't able to easily communicate visually with computer outside of your own.

Edited February 27, 2011 by Crazy Homeless Guy

[braddewald]

Thanks for all the input guys. I think the question seems a lot more complicated than it is:

 

Say I have a chrome material from an evermotion scene WITHOUT LWF. Instead of a map in the reflection slot, it is JUST a color value--let's say it's 128. Because LWF shifts the dark values to the top of the color picker, that value would need adjustment if I chose to use this material in a scene WITH LWF.

 

SO: Would I need to change that value from 128 to 55 so that the reflection color value is gamma corrected?

 

OR: Does vray simply look to the 0-255 range and assign a percentage value for refelction strength? (0=0%, 128=50%, 256=100%)? Because that would mean that a value of 128, no matter how shifted the color values are, is always 50%.

 

That's really the question: does vray use the NUMERIC color value to derive reflection strength or the ACTUAL color value?

 

Thanks again, everybody.

[Crazy Homeless Guy]

Say I have a chrome material from an evermotion scene WITHOUT LWF. Instead of a map in the reflection slot, it is JUST a color value--let's say it's 128. Because LWF shifts the dark values to the top of the color picker, that value would need adjustment if I chose to use this material in a scene WITH LWF.

 

Use 128. The color used for the reflection does not shift according to what gamma you are using.

 

Though, I will say that 128 seems low for a highly reflective material like chrome. I would expect something closer to 230-340, where the closer you get to 255, the more reflective the material becomes.

[CliveG]

Ha ha Travis.... I guess you're right, 2.2 is probably intertwined to such an extent it's difficult to unwravel -but I'm definitely the wrong bloke to look at every screen I have in the house!

 

I'm the luddite who only has two of that list of yours: Monitor and Camera! That's no TV no Cell, No I-anything.... and all the more liberated for it!