Hello. I am trying to create a Fiber Optic fiber that can bend and sway. I have tried a cylinder with Cycles Glass as it’s material with an IOR of 3.2. It’s OK, but the strand (cylinder) doesn’t have that subtle glow that a real fiber does.
I have also tried a mix shader with Glass & Translucency applied. Still not what I was looking for.
Thanks for placing this in the right spot. I am rendering in Cycles but am open to a Blender Internal option if there is one. None of my machines work with the GPU acceleration.
3.2? Where did you get that? Fiber Optic cable has an IOR of somewhere around 1.6ish in the core, 1.5ish in the outer section. 3.2 is huge, somewhere along the lines of of something like liquid gallium I think.
EDIT: Ah, I see. I think you googled “fiber optic index of refraction” and saw the wikipedia result where it says “Index of Refraction 3.2”. The 3.2 is a sub-section number of the article, not the material IOR itself.
Yep. Your guess was right. I Googled it and that was the number I saw. I am using Cycles, so I have a cylinder I stretched & subdivided. I gave it a Subsurf Modifier and a Curve Modifier so I could bend it slightly, then gave it the Mix Shader with Glass & Translucency as the two shaders being mixed. Inside, at the base of the cylinder, I have a filled in circle with an Emmission material.
At the IOR setting of 3.2, the light is strong when it reaches the far end of the Cylinder. At 1.6 it is pretty weak. I am testing settings between those two values now.
It’s going to be difficult to do without a spectral renderer because of the complex light interactions, probably something that would be better faked in Cycles.
Cool. That answer helps a lot. One idea I had was to place the Circle with the emission shader on the tip of the fiber. Then give the shaft of the fiber a, emission shader but set the value pretty low.
I tried making “fiber optic” in Cycles when I wanted to know if it were possible (even in theory) to send an IP package from a cube to another.
High strength emission in one end, the camera in the other and a curvy tube between. Just a glossy material without roughness was enough to get the light to go trough, but you have to remember to increase light bounces from the light paths.
But yeah, faking it would probably be much more efficient.
I got intrigued by the premise, so I did my take, basically it is a mix of everything already said above, I just read it thou, lol.
Anyhow fun to experiment, for the image I made the fiber more plastic “just because” if you put it straight and choose the colors correctly you’ll get to your reference image in no time.
Fake is the way… wink.
Cheers
Kubo
A little update, I’m guessing something is wrong on the hair setup 'cause it’s too tangled even thou I did a little combing.
I got myself a new wallpaper just the same.
Btw the file is setup in cycles for GPU but you can switch it to CPU and the render should look the same.
Marco’s idea to mix glass and emission was also my idea. It took me quite some time to render my very simplified version of the image at the beginning of the thread. (Some silly guy thought it would be fine to have hundreds of glass fibers… errr… That’s me.) Any way…
I just added some glow on the tips in post-processing and I increased the contrast with the GiMP. Still, it doesn’t hide the fact that I would have needed the double of samples.
It’s 100% Cycles… and 100% fake too. There is a second material for the bright tips. The first material makes all the rest.
I used some ramp shading to add a tiny bit of light along the edges of the glass fibers, I also added some gloss to make them shinier. The math nodes with the color ramp make it so that the lower end of the fibers is mixed with light too. It’s supposed to imitate what happens when some light comes out of the holes in the foot of the lamp with the fibers. It was supposed to be subtle… but the final result isn’t that much so.
Last one, I wanted to roast it a little more, as Kaluura’s the images needed some more samples, the first test was around 50, the second was 100 and this last one was cooking for 45min and 1200 samples. And it might use some more. Straight from cycles without comp.
I like Kaluura’s look of the fiber better, mine I think is even less subtle the light coming from the strand origin but just as soon as I start it to add a little light it rapidly overtook the whole fiber. Mine are thicker than they should also, Kaluura is more spot on. I might have to model it more realscaled 'cause once I started playing with the focus it was a pain to get the right one since there was no real world correlation.
Also Rickyblender with either one you (basically with the approach that everyone has mentioned in the thread) can fake that lamp, I’m guessing by using more and thinner hair and a single blueish tone with double ramp on the tip.
WOW. Nice work. I will try several of these ideas and see which works best. My final use for this is to have the fibers create a logo (sort of like this photo). Thanks for all the inspiration!