Laser Light

[Dacksoldier]

how do i setup a laser light thingy that i see in all of the tests so i get a small sliver of light coming out to shine onto an object?
would i be like two planes infront of a meshlight or more complex than that

[dougal2]

yup, that's the way to do it. best way is to make a long tube with the light at the far end and the open slit at the other, it'll make the beam more parallel.

I tried an experiment with a collimating lens to try and get a perfectly parallel beam, but I must admit it didn't work too well.

[zsouthboy]

I'd like a "built in" laser light type, personally. Instead of wasting power on the tons of rays that don't make it out, just shoot mostly parallel rays in the direction of a normal.

Unless someone is able to make an incredibly realistic laser as a model that we can all share - dougal2 was the man I thought could make it happen, I know shit about lensing.

[StompinTom]

dont fresnel lenses emit ~parallel light rays? correct me if im wrong, but thats how i thought they reach so far (therefore their use in lighthouses, spotlights, etc). so what if you combined a fresnel lens with a very long tube? render time might be high, but it could work.

[Kram1032]

at least the colour of lasers is supported in indigo
You can use a peak-light for that...
I think, it's a nearly unknown feature, as I haven't seen ANY peak-light in-use...

For a "real" laser, you may need to model a diamond, or so....

Dunno, if Indigo can animate THISkind of laser...

you have to try

use the links to come around through all that can be read about laser in wikipedia


Edit: I found a better link
http://en.wikipedia.org/wiki/Laser_construction

[Dacksoldier]

ok ty cuz i tryed just a box with a slit in it and a mesh light inside of the box and it didn't really work how i wanted

[Ricky]

Laser is not only done by having parallel light rays, "Collimated" (that anyway you get only with an infinite long tunnel...), you need also coherence:

http://hyperphysics.phy-astr.gsu.edu/hb ... ig.html#c4

Without, you won't have the behaviour you expect from laser light and you can't fake them (yes, of course, with post pro... )

Look also to:
http://www2.indigorenderer.com/joomla/f ... ight=laser

Ciao,
Ricky

[Conz]

I'm testing some laser beams, too. The beam looks correct, but how can I create a mirror which reflects the beam 100%?
In this test the mirror is a phong material with exponent 1,000,000 and ior 1.0.

Or can I stop my tests - indigo can't archive this effect?

[Kram1032]

100% reflective mirror:
make a specular non-transparent material with 0 absorption and 100000+ IoR (ok, IoR100 is pretty much 100%, but higher is even better )
I don't think, you have a laser, there, btw...
It's slightly cone-ish..
a laser should be straight, which can't really be done in Indigo...

If you don't already do so, try to use "peak" light with a very sharp peak


Oh, and as tabulated spectrum should work with everything, that has RGB (if I didn't missread...), did anyone try to do the same for rgb-light? O.o

[Conz]

Thank you for your help, but it doesn't work with a nontransparent Specular.
Indigo starts to produce a lot of errors.
Could you point me to some documentation of the "peak light"?
And yes I know that the laser is only a fake and not parallel, but for testing and my needs it's more than enought.

[Kram1032]

all documentation I can give you, is in the Indigo Manual...
There weren't too many users of peak light, yet...
It's surely much easier to simply use an rgb light. But with correct settings, peak might be more real (Colour-matching to a real laser, if you have datas about the nm range of it...)

If you want to, you can also try to use your old material, but you did one mistake with it:
IoR 1 means no reflections at all

But I guess, your current results are kinda fine - you only need to bake it longer (though the green light beams look very solid...)
Did you change anything else, besides the mirror?

[Conz]

I have changed only the material. All other settings are the same. 25m rendering on 3 PCs. After such a time it should look better.

I'll try it with the phong and higher IOR.

EDIT: Yes! With a phong and IOR 100 I get a nice reflection. Thank you

Image after 6 minutes (I have changed only the mirror material):

[Conz]

Hmm ... every specular material in the beam produces some problems. I have now a little bit complexer test. The light divider are blended materials (blend01, blend02) - mirror / null. I have used 75% null and 25 % mirror to get a 50% / 50% division. (visually)

I have no specular tests online!

This test needs some more tweeking. For a better performance the room has only three sides and the room has a black diffuse material to catch all free photons. Without an object behind the SSS athmosphere I get some other render problems. (no SSS will be rendered).

EDIT: I think the blend02 needs a second blend to some translucend material.

Some more advanced laser test:

[Kram1032]

that's great!
strange that the specular doesn't work as needed...
humm...
I've one guess, why that could be:
are your mirrors solids or just flat meshes? (if they're flat, you really should make them solid... they can be very thin, but not flat)

for 50/50 transparency, it actually should be a 50:50 blend...
strange, if it isn't...

if you don't want to loose some light, up there, in the mirror that sends the beam to apparently nowhere, you should make it slightly larger (though, it looks like a nice effect, so, there might not be a need for this )

That new beam is far more straight
Do you think, you can straighten it out, even more?
I'm not sure, but I think, this is a scene, where it might be useful, to increase the maximum ray bounce depth (in the final render: before, it wouldn't make much sense, as it'll slow down the whole render)
while testing, you should lower it, though, just to get faster results

[OnoSendai]

Hi,
you're not supposed to blend null materials and specular materials, perhaps that is causing problems.
If you want a partially transparent mirror, try blending a IOR 100, high exponent phong with a null material.
Cool tests!