Show Us Your Bryce Renders Part 10
This discussion has been closed.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
My last video prompted questions about scale, here is my video response. Be aware it is 40 mintues long. But in case you are interested in the topic covered, here it is.
Will cetainly watch the vid, David. Initial impressions on the previous introductory vid(s) is that the 'curvature filter' - an unintended technical term - can be used to produce additional detail in terrain and models that have curved surfaces. That the curvature part - from both a concave (saucer-like) and convex (mound-like) perspective - are both related in the filtering process, the two work together. So, the key word is 'curvature' - wherever such a surface exists, this app can be applied/manipulated/processed to produce more detail on it. Wonderful research!
Jay
Thank you Alexey! The best is yet to come, right now I'm working on 3 differents scenes, trying to rise up the bar on each.
@Mermaid : thank you! Nice grass and bench render from you.
Jamahoney - thank you. Agreed, not so obvious on a terrain but without it, you'll see.
As mentioned before, I find David's video on curvature just great. I even managed to create a curvature texture template which I used to mix textures from the High Resolution Terrain sets 1 and 2. The object is an Alien Artefact by MatCreator with nice concave and convex curves. A Hyper Texture was also used for the specular and anisotropy. It stands on Art Wade's (aka Fencepost52) Seamless Backdrop, which is partly transparent. The light and backdrop shining through is from the WpH14_SC HDRI from the Hyper Texture Base set. The sun provides only specular. Rendered with soft IBL shadows.
@Horo, great results on those ribs and a very nice render also!
Well, here I am getting around to answering another of Marco Di Marios's questions. This was an intereting one to investigate but is only probably of limited interest to most Bryce users, though maybe some of the power users on here might wish to take up this gauntlet. Here we go. This is my initial stab at the workflow.
David Brinnen- Outstanding demonstrations!! And incredible results in Bryce with the pbr rendering studies
C-Ram- I love the warmth and openness of this image. Perfect scaling and placement as well as flawless lighting and texturing. Love the thickness of the grass. The clouds are perfect.
Horo- That artifact looks absolutely photo real to me, no question. That porcelain sort of finish is expressed perfectly. Along with the curvature doing incredible things as well. Doesnt look like the typical expected output from Bryce.
SlepAlex- I suspect with your type of workflow this curvature discussion is going to lead to some very cool looking materials in some upcoming renders. I can't wait to see the different ways in which you will incorporate this tool.
Mermaid- Looks like you're having fun. I think the models
In my own theorizing I feel confident that curvature driven ambient coudl help to fake sss effects on human skin. It might require a high degree of subdivision to work, but in theory it should work. Maybe ElectroElvis will beat me to it.
Thanks Horo, Jay, C-ram and Rashad for the comments.
David – thanks for the videos, the scaling one looks interesting.
Horo- very nicely rendered artefact, love the lighting.
I had fun with David’s curvature tutorials, understanding zero%, fun endless here’s my take. I know I didn’t achieve what was intended but I like the results.
Mermaid, there is a bit going on the edges there, it is quite subtle, which is not a bad thing.
Rashad, thank you. OK, curvature to fake a SSS effect. Interesting idea, I was thinking more along the lines of faking AO, so as it happens, this is an easy conversion to try out. The idea is that this looks like wax. First is the image, also supplied is just what the ambient channel is doing by being driven by curvature all on its own without any other light.
I'm currenlty trying to apply 'curvature' to a Daz head/face (a perfect subject of valleys and hills), but my ignorence is getting in the way - sigh.
Fingers-crossed, a comparison between non-curvatured and curvatured images of a head/face will eventually be produced. As with Rashad, Elvis (without putting on too much pressure on you Elvis, or need to produce results), but it must surely be interesting. The fake SSS (as opposed to 'Fake News' - hehee
), Rashad mentions sounds interesting.
Nice one, Mermaid - apologies, if I don't see this 'elusive' curvature feature the 'pro's' are talking about in your work, but it is a complex, concept area to grasp, so we amateurs aren't exactly 'seeing ' what the 'pros' are talking about. Perhaps, higher resolutions, comparisons are required.
Jay
David- It occurs to me that in the PBR video that the ambient glow output isn't as intense as it was in substabce painter. I think that it might have been necessary to give the ambient function its own channel and to have used some other image as the glow alpha instead of the specular map you used. Am I making any sense? I suspect that in the terms of the behavior of metals Bryce is probably in a good position to compete realistically in a PBR workflow. Ambient occlusion is a great idea with curvature, but this would most certainly require smoothness in the curvature we dont yet have since it would need to appear as an indirect lighting effect.
Jamahoney- Fake new is better than no news I suspect!! AS far as what the experts are looking at with curvature, David and Horo are on a completely different level than I am. What I find compelling about curvature is that it analyzes the angles at which faces point toward one another to determine if there should be curvature applied. Curvature doesnt yet work with gradients as I can tell, so the transition will always be sharp. A polygon is either "curved" shaded or it isnt. Some blending and mixing between zones is all this feature needs to reach acceptable maturity.
Mermaid- I never finished my previous comment. I meant to say that the models look great including the grass. I like the wall textures as well.
Oh! That IS interesting! Very good!!!!!
Rashad (apologies to Horo and David in this now-topical/complex 'curvature' area - your works/contributions = excellence), but that's about the best explaination/description of 'curvature' that makes sense. Super research to ALL - keep going, really love reading discoveries/contributions
Jay
Aye, the lighitng conidtions and the handling of ambient vary. I considered multiplying the AO with the diffuse in PSP8 and dropping that into ambient instead of the glowing output - indeed there are quite a few ways that this can be re-jigged depending on what properties you want expressed most in your material. Given that the basic requirement is to understand a PBR workflow, I fully expect that given a nudge in the right direction as far as a starting point is concerned most people who are capable of exporting a set of effect maps are equally capable of pluging them into appropriate channels. The key feature, which I think would elude most casual users of Bryce is the means by which it is possible to express the roughness data.
As to balancing curvature against it's shortcommings, it is fun but fiddly. So long as the effect required is not too radical or the model too course, the faceted nature of the response can be lost in the texture. The wax was a good test, since there is not much texture to hide the facets in. Hence the subtlety with which it was employed. I'd have probably put in more contrast if I thought I could get away with it.
Yes that is a good way of explaining it, I will have to steal that for future use. Also consider that it does not appear to be mesureing the angle of adjcent faces (as this would remain constant when the object was scaled), but at a set distance that does not scale with the object. So that as objects get larger, the measureing distance falls relitive to the surface and so the surface often appears less measureably curved. And also that the output is average of a product of measurements along different axis. Hence all the trouble over trying to decide if a saddle is convex or concave or flat - but some arbitary decision has to be made and depending on the flow of the geometry and the type of object you wish to prortray one choice may prove better than another. Ultimately it is something that needs to tinkered with until you get a feel for what is happening and you end up with a hard drive that is filled with scenes that look like this...
David - interesting video and AO emulation experiments.
Rashad - thank you. Interesting remarks about curvature.
Mermaid - vases looking nice.
We're not yet there with understanding curvature completely. It appears that also the camera angle is taken into consideration. Two simple metaball objects (Bryce trees, wood only, no foliage) and "default" curvature in channel C as in the video. The objects are the same (copied) and equally left and right of the camera. One has the camera (almost) isometric with FOV 1 degree, the other 80 degrees. At least half of the objects ought to show the same pattern.
Camera angle... Hmn... that is interesting. Well we don't know how curvature goes about inspecting the surface, it can't be exhaustive, so it must base the probes it makes around some convetion. Either the object orientation, or the world orientation or perhaps this suggests the measurement of the surface it triggered by the ray comming from the camera?
Edit, also... trees... metaspheres... not the most stable of subject matter to begin with. Will have to test this finding with other geometries.
It is difficult without knowing how the render engine works. I was of the opinion a texture is "baked" to the object. It appears the angle with which the ray from the camera hits the object is also taken into consideration. Metaballs, yup, a bit out of the ordinary but then, why not try it? In fact, I wanted to do something with them but then got stuck and started to look closer.
With exception to Time (an essential Einsteinian ingredient
), this 'curvature' thingy - space, object, light, lensing (virtual) effect - may be a TOE (graphic-wise) 
Interactivity (3D/4D): another option to be explored/developed.
Jay
What I found with the metaball trees (I failed to spot the channing of the response unlike yourself) was that it was very difficult to set them up with much of a transition zone. The curvature response seemed to be bordering on binary. It is always the case that one thing leads to another. I lost hours after discovering that if you use anistropy, with only one of the axis mapping enabled and everything else unticked it is possible to generate a powerful negative response in the diffuse - the effect is unpredictable and unstable with standard Bryce primitives but is far more obvious with imported mesh objects. In spite of a lot of testing, I couldn't "do" anything with it. It is more or less beyond control and overpowering.
Edit: What's a TOE Jay?
Ooops, sorry, TOE represents - in the physics/astro world (and now, in Bryce's exploration/curvature-world
a 'Bryce Theory Of Everything' B-TOE): himself, Stephen Hawking, is surely observing Bryce activity.
Jay
A Theory of Everything sounds ambitious, to explain everything... wouldn't you need another universe at least as big as the universe you were tyring to explain to keep that theory in? Or is it a lot less ambitious than it sounds. Like it might explain some fundimental things about physics that are of interest to accedemics but it won't solve any of lifes real mysteries like how there's always a teaspoon left in the washing up bowl no matter how hard you check it only appears when you empty the bowl - kind of thing.
David - yes, I've also observed that there is almost no transition zone, no matter how wide it is set in the DTE. However, earlier tests showed a small transition zone near the ground plane and when I moved the tree up, it disappeared. Hiding the ground plane and moving the tree down also made the pink transition zone disappear.
Anisotropy, interesting stuff. I haven't detected the negative response though I tested all combinations - including the Circular Mapping Easter Egg - and have them documented here: https://horo.ch/docs/mine/pdf/Anisotropy.pdf
Jamahoney - getting a toe into the TOE
Pheew/Wow...David, you truly are an explorer/developer/programmer of OUR times. TOE...or B-TOE - so, where shall we begin...hehee
Horo...I'm such a clutch...sorry that I sometimes come across as priviledged, obnoxios, but I'm honest...and certainly ignorent - in a sort-of-ways.
Jay
David - that anisotropy experiment is again funny. Spheres with sizes 20.48, 21.00 and 22.00 behave differently, also if the location is changed. Using a white HDRI sphere to light, there's no difference between IBL Specularity 0.01 and 1000, -0.01 and -1000, but at 0. No specularity but different red-black patterns. IBL Quality also changes the pattern.
I tend to agree with you on this, David. If Copernicus was right, in that there is nothing special about this particular point in space or time; then on a "first pass" of thought experimentation it would seem that a theory of everything could be reduced to a theory of a few things...the things that we can see and test directly. However I observe that what takes up a lot of space is establishing the "proof" that the theory we've come up with really does encompass everything...every single question that could ever possibly be asked. That requires infinite testing, as well as infinite processing power, as well as infinite bandwith to upload the questions and download the results and infinite memory to store the information once extracted. In addition to those infinite number of tests is that the tests individually must be controlled. That means that one would need to treat the entire universe as a control group, thus the necessity to add a second universe to be the non control group, and to do that you need yet an additional third space large enough to hold the both universes upon which all these questions are being posed. Spaghetti in the brain.
The theory of everything will turn out to be the theory of nothing, as it is the nothingness of this universe that we have the greatest need to define.
Aye, and also the black generated is not just black it is a "black hole" black. As can be shown by testing with TA. I've yet to find a hypertexture setting to overcome it. It might be the ultimate negative hypertexture...
Images, regular and TA showing how the light is mopped up by the dark areas.
Nothing... nothing might not be possible, like infinity, it might be a thing that only maths can aspire to expressing. I mean from our own observation, the one certaintly we can have about the universe at large is that it is "something" and not nothing. If it spawned from nothing, then that raises the question of a motivator. But perhaps that's just a bad way of looking at things. If time is a human impression, like colour for example, and not a "real" actor in the universe at large, then it is simply wrong to suggest that there is anything other than the present. What we think of as time is an illusion of conciousness, the way we experiance the eternal unalterable present of the universes reality. Which places us firmly in the uncomfortable camp of determinisitc reality. So I had to write this. You had to read it. And we are way off topic, but I can't help that, the universe made me do it.
Oh dear - sorry...for opening up the whole multie-multie TOE thingy on a graphics forum (a bit like mentioning politics and Trump on Twitter
, or God on an Atheist channel 
).
Forgive, we will forever be going down this TOE route 'til...say...DAZ...changes to 64-bit. I'd love to engage, but hundreds, if not thousands - posted on other TOE-related (never non-Bryced) sites/vlogs - is that repetition/understanding never satifies. Explaination, yes...but it takes time (ooops, that Hawking's term
). Apologies - sounds like a cop-out....
Again...the above 'curvature' thingy (your wondeful research, experimentation and detailed/interest David, Horo and Rashad) is simply fascianating - an area that we Brycers are loving, I bet.
Jay
Well I won't continue in that vein if Rashad doesn't, but I suspect Rashad loves a good debate and he's certainly good at it so it seems a shame to miss out on his argumet tallents.