making real foam on the drink - like beer or coffee - material. Not just faking it with flat photo but realistic looking foam. That would be absolutely awesome.
@Johnny Bravo I suppose links are prohibited in this comment section, but you'll find this model in cgtrader and turbosquid by searching with keyword Ganesha :)
Awesome video! Would love to see something about texturing in Substance painter and then rendering in Arnold. I can never get my materials to look as good as they do in Substance.
after 20yrs of cg that's *still wrong*. It's the latest *trend* and it's stupid. To begin with, you must have cross-polarized diffuse maps / reflection removed, in order to use that, or your rendered color will be falsely desaturated with double-reflection. I've only ever seen those once on a project, and no mid-level studio or below is going to the trouble to make them. Secondly it makes every surface in your scene cpu-intensive, with raytraced rough reflections. I've already seen single frames that take more than 1 and 1/2 f'ing days to render, with no hair / sss / volumes or anything else that's slow, just hard-surface stuff, and still had a bit of noise. At that rate, Toy Story 1 would still be rendering today. And arnold is slow enough as it is, so this is just baffling to me. *Also ignores surfaces like chalk, rust, clay pots etc.
@@axe_fx crossed-polarized photography separates light that's reflected directly from the light source ( specular ) from light that's bounced once or more in the micro-bumps of a surface ( diffuse ) ; ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Z8AAX-ENWvQ.html He shows it at 8min. 18sec. But 'he' says to use 0.5 specular, not 1.0! 🤷♂
Great stuff, thank you. Word related to the last comment you made. Yesterday I was getting really horrible, over-saturated noise/artifacts on texture. Really simple material: some metalness and lower roughness for reflections plus jpg picture for a color. I was able to eliminate them completely precisely by lowering weight in specular. I dont know how and why thats the case but I had this artifacts only with weight set to 1. Anything tiny bit lower is fine. Any ideas why it could be?
first off, jpgs are 'craaaap'. find textures on the web that aren't jpg, and don't convert them to jpg. they can cause technical problems as well as quality problems - they're only designed for final viewing, not rendering. That could be your problem right there, since jpgs do have artifacts in them from the compression process, that happened to me years ago. "some metalness" There's only a handful of semi-metallic elements in the world, like germanium, and boron or something - so for anything actually used in 3d scenes, it should really be 100% On or 100% Off, unless you're sure that a mix is what you need. Also seems that Arnold is also just adding the metal reflection to the Specular reflection too, which is wrong, and makes fireflies worse. "this artifacts only with weight set to 1" What happens when set to 1.0 is that other channels get switched off completely. Metalness 1.0 shuts off all contribution from Specular for example - so dropping the value changed something like that.
@@arvidurs thanks for the reply. Also I thought it would be easily translatable into Max, but many maps do not exist under the Maya name and many have different features.
Any way you can make a video of a animated creature like dog or monkey and have xgen fur added to a Bind skin with animation. And have fur follow along?
11:11 You can't tell people to keep specular at 1.0 and turn up roughness, just because "all light is reflected", because that is what the Diffuse channel is simulating in the first place. Surfaces don't reflect all wavelengths of light at 100% ( xcept maybe white ), and that is why we get colors. So unless you're suggesting putting all diffuse textures into the Spec channel, then don't tell people to leave Spec at 100%. Any diffuse textures based on photographs are 'already' going to have this reflection color baked into them too, unless they were photographed with the cross-polarizing method. That's like 0.001% of diffuse maps out there. How many of you photograph your own objects for texture painting, and have polarizing filters? Or even know what I'm talking about? This silly little "I'm thmart, rough-reflections-on-everything!" trend needs to end. It's loosely based on realism, yea, but so is using sss on all wood materials, or getting rid of cg lights in favor of using a super-bright filament inside of, and focused by, a reflective lighting instrument, then rendering caustics and extra trace bounces to get 'real' light. Having raytraced, rough reflections on everything in your entire scene, is also needlessly spiking the hell out of render times, and Arnold is slow enough as it is.
First of all, it's my channel, and I can tell people what I want. Now that that is out of the way, please show me any dielectric material which specualar response is non-white. Your remark on putting SSS on wood, is also wrong, as wood does not scatter light as a soft body like wax or skin. If you want to model wood, it's made up of fibres similar to hair, so it's a completely different shading model. If you listen closely what I'm saying, you will notice the phrase, "I'm using this workflow for all my basic materials". I appreciate the comment, and the sentiment you are trying to convey, but make sure to cover your bases with remarks like that. I would love to see some papers about common colored dielectrics that reflect different wavelengths. Also please have a read through this: autodesk.github.io/standard-surface/ It might help your understanding on the standard surface shading model.
Hello Arvid, happy to take your request of ideas for new videos litterally : how about data passes (depth, PW, PO, etc… ) in reflections like mirrors ? Often in 3d these are pain in the ass to make it work. Do you have any tips on it ? Cheers !
