Fixed all of the alignment issues. Richter used a lighting feature in Maya that splits light into a spectrum for his timeless gift entry that is the inspiration for this image series. To fake that effect here I have used a dome made of 80 lights with colors arranged to construct a perfect spectrum. The colors are balanced exactly to combine to white. Hope you like it and feedback is very much appreciated. Thanks.
I'm really pleased to see the Haematite put to good use here, in your image. It responds quite well to the lighting applied, thus giving you this specific cold-black titanium look. That's what I personally like about the Haem. mat. I've checked the previous versions and I've got to say this one is the best of the three. There are certain minor things in this one I'm not so fond of, but as an overall result - success, Rashad. It appears to me that you've also changed the chess board design with the edges raised up a bit. This gives me an almost subconscious assurance that a figure shouldn't fall off board if I'm to accidentally tip over some of them :) What I trully like about this final image is how you've managed to fake those caustics in a particular area - lower right - I see "defined" prismatic circles which can be very much taken for caustic spots. I also like how the board reflect some prismatics here and there. The glassy material has very good refraction and the "black pixels" is what this glass really needs in my opinion to become a "real" glass. Concerning the models, it is not visible on the opaque figures but the transparents ones - there's some distortion in the mesh, as if someone used the smudge tool in PS to edit few pieces in postwork. Right here it's a modeling issue I guess, so it's not your fault the figures appear this way. On the left side of the image there's a prismatic band that is "unmotivated" right there. I don't see what could be causing the light to scatter in spectral hues there, no refracting surfaces, anything(?)
As i said, I find this last image successful - you aimed for caustic effect in Bryce, you've got it!
Join Date: 05.26.2004 Comments: 4442
Great effect! Everything looks plausible, except the rainbows at left, as richter mentioned. We see it at the raised edge at left as well. Though it looks nice, I'd like to have an explanation where they come from. Otherwise, the black figures look excellent, the white ones glassy as you wanted them. Again, richter observed correctly that the black spots make it to glass - crystal glass with a high refraction. I like the monochrome appearance - only it is not monochrome, as we can see from the spectral colours. Good job, really.
Join Date: 01.03.2004 Comments: 2227
Interesting strategy, the effect is subtle, but that should not be consiered detremental. I'd like to say I was entirely satisfied with the result, and this is not a criticism of your skills, but of the limitations of the bryce renderer - for me, the shadows of transparent objects always look wrong. How wrong, it's difficult to say, at least until you set up something similar in real life and you see how the light is lensed through the glass/crystal. I don't know what the answer is, since I don't imagine, accurate lens simulation is going to be high on everyones list of must have features for their render software, but I would think it fairly important, because I like to see things working at a fundimental level. Nevertheless, a cunning idea and one certainly worthy of full marks!
Join Date: 12.21.2006 Comments: 667
now the only way to test if this is how it should look is to render it in Indigo :) those spectrum lights I think over did it a bit, there is just too much color there, but that is not something one can just tell
Join Date: 06.04.2006 Comments: 2622
Thanks you guys. I am glad to be moving in the right direction. The spectral banding is caused by two forces. I used a dome of lights made of about 8 different colors. I also applied a spectrum image into the transparent color channel. It is the spectrum image I think that causes the most noticeable bands, especially those along the borders of the board. Likely gat this is what you are observing. The colored lights actually produce a very subtle banding effect noticeble mostly around the shadows of the heads of the white king and queen figures at the bottom right. This seems like about the physically correct spectral banding. I decided to push it further. Also the spectral lights are great for the crystal but bad for the shadows of all other objects in the scene so I could not make them brighter. If you look to the left you can see the shadow of the black castle splitting into a spectrum also. It is subtle but still troubling. The haematite does a good job of absorbing these colors and hiding the spectral rigging. It is because of the troubling colored shadows spread among the solid models that made it clear to me that I could not rely soley on the colored lights to color my crystal, thus the implementation of the spectrum image in the transparent color channel.
David, I agree with you in that odd things happen with transparent objects in the caustics/ shadow functions. I think it is caused by the lack of GI. If the cryatal object was really bouncing the environmental light from all angles the caustics would be far richer.
The models are rounded and smooth and not sharp and angular. Had the pieces been more angular there would be alot more spectral response going on. The best way to test this lighting/material theory is to apply it to a diamond that will have enough internal reflection to split the light unpredictably. The notion of rendering it in indigo seems cool. I feel though that Indigo can be a little too generous with its indirect light levels. Due to the roundness of the edges of the models I doubt Indigo would offer up much spectral banding at all for the scene above.
Thanks for the feedback. I am done with this one for now but I think I imagine a marble variation in the future sometime.