Grant Warwick- Mastering Vray: Material Ideas Needed.

[JezUK]

Just to chime in quickly. I didn't get the Lesson 5 part 2 email either but I got the one afterwards about the material contest.

Same here - material contest but no L5P2

[Wilyman]

Today's e-mail seems to be working, for me at least.

Grant, is there any chance the ice shader that you showed way back towards the start of this thread will be making an appearance in the next material pack?

Cheers,

John

[ChrisS]

...Grant, is there any chance the ice shader that you showed way back towards the start of this thread will be making an appearance in the next material pack?

Cheers,

John

+1

Thanks,

Chris

[bennyboy]

Does anyone know what the deal is with V-Ray RT when rendering GW's material scene? The reflections are all wrong...even with the bump map disabled (which apparently is not supported). Falloff maps are supported, but I'm sure in the way he's using them. He mentions RT in Lesson 6 but then says he doesn't like, but doesn't specifically mention RT not supporting his fresnel technique. Got a massive GPU and it would save me a few lifetimes to be able to use it with complex materials.
Been searching all over for answers to the general topic of "falloff" and "reflection" with respect to RT, but it's tricky since the search engine on this forum negates "rt" from all searches - and there's no mention in the RT forum. Anyways, would be nice to be able to tweak these materials via RT, but they don't look right...I just am not quite sure what it is that's not working, since RT is supposed to support falloff maps.

[soulburn3d]

Hey Grant, just watched that free video you uploaded. Good stuff. I'm always interested in seeing how other professional approach making materials. Lots of "My god, I can't believe he does that THAT way" mixed with "Wow, that's a really cool way of doing that, I wouldn't have thought of it." And thanks for slamming the max built in colorcorrection, you said exactly what I feel, that thing needs to be replaced ASAP by someone who knows what they're doing.

The biggest thing I learned was man do we need the GGX material. Your workaround for achieving the multi-highlight look certainly looks better, but it shouldn't have to be that complex. Really hope Vlado gets GGX in there next version.

Oh, and BTW, some of us at the professional studios ARE adding that extra level of detail

Anyways, cheers for posting the video!

- Neil

[davision]

Bennyboy: I can't be of much help, but I'm definitely interested in the answer. (I haven't seen Lesson 6 yet so maybe my response below is worthless).

I'd suggest making a super simple scene and material with a real high contrast environment (like black and white stripes or something) and rendering it with both RT and regular Vray and look for disparities. If the problems are crystal clear then put the scene in the Problems forum as a fresh topic -- instead of this behemoth thread.

I'd also like to use VrayRt GPU for Grant's stuff -- but it doesn't support Composite or Bercon or VrayDirt etc -- so it seems like you'd have to sacrifice a lot of techniques when using it unfortunately. I'm crossing my fingers that RT starts to handle some of the more complex maps soon and then it will truly become the dominant render engine. Right now, it seems like it forces artists to make too many compromises.

[davision]

Neil: I'd love to hear what aspects of Grant's workflow make you think "I can't believe he does that THAT way". I am not asking in an argumentative way -- just curious.

I personally find his techniques pretty logical... but admittedly they're not always super-quick (to create and to render).

And I agree on the GGX -- that would be great if Vray could have that built-in.

[soulburn3d]

I'd also like to use VrayRt GPU for Grant's stuff -- but it doesn't support Composite or Bercon or VrayDirt etc -- so it seems like you'd have to sacrifice a lot of techniques when using it unfortunately. I'm crossing my fingers that RT starts to handle some of the more complex maps soon and then it will truly become the dominant render engine. Right now, it seems like it forces artists to make too many compromises.

Completely agree, from my perspective, all of the current realtime engines are great if you're only using bitmaps and nothing else, but for real shading, none of the engines are compatible with enough maps in their host application. Hopefully that'll change one day, but until then, software rendering for me.

- Neil

[soulburn3d]

Neil: I'd love to hear what aspects of Grant's workflow make you think "I can't believe he does that THAT way". I am not asking in an argumentative way -- just curious.

Well, one example was already mentioned in the comments on the video, using the vray light material to manually see the result of a single map. I got so tired of doing that manually I wrote a script to do it for me. But I agree with Grant, sometimes old habits die hard Also, he uses curves for a lot of color correction. Absolutely nothing wrong with that, it gives you a lot of control, just very different from my approach (which is using Cuneyt's colorcorrect).

One of the things I thought was really interesting was using the dirty spec map in the glossiness instead of in the spec strength. I rarely add variation to the specular roughness of an individual material (usually because my materials have enough layers, each with their own specular roughness, that the variation comes naturally from the layering process), but I may try out my dirt maps in the glossiness to see if they give notably better results.

Overall I agree with pretty much everything in the video, Grant knows his stuff and is doing the right stuff for the right reasons.

- Neil

[jacksc02]

guys check out this link for what is supported in RT -
http://docs.chaosgroup.com/display/V...orted+Features

[Pixelcon]

Guys the GGX model is already in the nightlies and it's great

[bennyboy]

Davidson, good suggestion, thanks!

[soulburn3d]

Guys the GGX model is already in the nightlies and it's great

- Neil

[davision]

Neil: I also wonder about using a map to control specular glossiness (I haven't done that in the past very often) and then again in the bump map.

Degree of glossiness is essentially a function of smoothness -- that's why a granite counter top looks a lot different than raw granite in a quarry -- then why would you apply a map to glossiness and then again to bump? It seems like you would be doubling up on the same effect. Obviously it looks quite good as Grant has done it. I'm asking more about the theory. It seems like specular glossiness should be mapped to handle very fine (microscopic details) and bump should handle the coarser ones. But both slots don't seem like they should be used to handle both micro and coarse details (again, in theory).

[joconnell]

You're quite right about bump really being the driving factor of glossy or sharp reflections but it's at such a fine level, it's not something that we could really do with a bitmap or pixels. If you had a bump map fine enough to capture this level of micro facetting, vray's anti aliasing would just blend this into a flat grey value as it's well below sub pixel sizes. Neil did a tutorial or a run through showing using a really fine bump map as a way of getting glossy reflections I think? Glossiness should be approached as flat areas of colour to set the level of "focus" for the main body of the material with variations painted into the map for parts of the material that have either become duller or more polished depending on the level of wear. Bump as you say should be used for surface undulations at a human visible level, unless you're doing macro shots. You could probably figure out how much of an in or out effect is the bump having on the surface and then decide whether it should be glossy value (macroscopic), bump (surface level / depending on scene scale) or displacement (large scale / silhouette breaking).