Do you like... jelly candies?

[b-tag]

Hello,

Here is my attempt to render some jelly candies. The particular design uses glass2 materials featuring absorption and bump maps (who mix wrinkled and image map textures), together with particle systems for the air bubbles and HDRI lighting.

Fig. 1: Experimental render of jelly candies at ~1900 S/p.

Leaving away for a moment some easily corrigible flaws in geometry (which I noticed after few hours of rendering), do you guys think glass2 is the best choice of material, or I should use any other, say rough glass or some translucent ones?

Here is some food for your critical thinking:
http://home-and-garden.webshots.com/photo/1169714124047641748vycWfX
http://www.flickr.com/photos/8087821@N07/2489062482/
http://www.flickr.com/photos/cf_al_bs/1954508715/
http://www.flickr.com/photos/baby_jacksons/4311861351/
http://www.faqs.org/photo-dict/phrase/8087/jelly-candy.html
http://www.flickr.com/photos/scaredsquee/362328697/
http://www.flickr.com/photos/emyanmei/4087005299/

Your comments and suggestions are really welcomed!

Regards,
b.

[Meelis]

Yes i do they look very nice.

[tomb]

Brilliant!

[J the Ninja]

Glass2 is probably best......glass2-glossytranslucent mix (are you using 0.8 or 0.7.1?)

In any case, it is awesome and I'm saving a copy. How many s/px?

[A-man]

This is very good! The only thing I'd suggest is maybe a subsurf or 2 on the candies. I think I see some edges..

[paco]

They look very close to a "hard" candy - I take it thats what you were going for? If so, very nice

I must admit, when I think of "jelly candies" I think of softer things though.

[jeanphi]

Hi,

Great render, and nice composition.

Jeanphi

[pfannkuchen_gesicht]

oh man, never render stuff like this... it always makes me hungry because it looks so real

[qinjuehang]

Really realistic! Are the bubbles particles with dupliverts?

[b-tag]

Hello guys,

Wish you had a merry Christmas and some great holidays! Happy New Year!

I love your replies! It is a pleasure to read your comments and suggestions! They are really helpful!
I wish also to apologise for being slow in my reply. I am certain however, that any discussion can be more intriguing, when there's something new at hand. It's not much, but here we go.

[...] (are you using 0.8 or 0.7.1?)

[...] How many s/px?

Since there's no way I could recover these numbers straight away, the answers to your questions are just below. I am attaching the same scene, re-rendered with improved object geometry and statistics overlaid.

Fig. 1: Jelly candies re-rendered using glass2 material.

Hope you like this result as well!

[...] The only thing I'd suggest is maybe a subsurf or 2 on the candies. I think I see some edges..

Thanks, that did the trick (together with some manual edge splitting here and there). In order to keep the polycount low however, I used the LuxRender's subsurf feature. Do you guys know whether LuxRender 0.8 subsurf supports the Blender's edge creasing? I think there are still some issues around who need a little tweaking here and there...

They look very close to a "hard" candy - I take it thats what you were going for? If so, very nice

I must admit, when I think of "jelly candies" I think of softer things though.

I don't disagree at all. The kind of an unusual appearance here is provoked I believe, by the "glassy" look and "hard" reflections. So as @J the Ninja proposes, glass2 material alone might not be the best choice. Here is another preliminary result when using roughglass material however:

Fig. 2: Jelly candy modelled using roughglass material (and air bubbles).

And that picture definitely rises my curiosity about what would the result be of mixing glass2, mattertranslucent, homogenous volume and etc. materials. But that's not all the story.

As fellows of mine in biotech pointed out, apart from birefringence, absorption and viscoelasticity, jelly sweets under certain conditions also exhibit this kind of slight petroleum reflections. So what do you guys think is the appropriate way to add thin film interference? Say to mix the main material(s) also with a mirror or metal? Or there could be a better alternative?

Please bear in mind that in order to demonstrate true viscoelasticity, I need to pass the jellies through the Blender physics engine and animate the result, therefore duplicate geometry isn't quite appropriate (as much as I see things).

[...] Are the bubbles particles with dupliverts?

It is DupliObjects actually (Blender 2.55 here) - I have randomly distributed a randomly sized-down glass2 sphere throughout the whole object volume, then transformed the particle system to a real geometry. It would be nicer of course, if there was a wider Blender particle systems support on the LuxRender side (which could be of a benefit also to animations). *wink*

Thanks for your replies, fellows! I really appreciate it!

Regards,

Bogomir