[bograt]

Polynurb your mad

[polynurb]

As I read from another source some times ago, the laser has the advantage of staying focused also with differing distances to the projector-wall (screen). The question is if simulating that case is possible and how that can be achieved accurately.

you can use camera projection for that. but light rays won't be collimated, thus caustic reflections off bent surfaces won't be visible

[polynurb]

Polynurb your mad

you don't like disco?

[bograt]

I LOVE disco

[Frank1000]

ok great. I have Softimage. Probably you can save it as .fbx which I can import ?

Thx!

[bograt]

ok great. I have Softimage. Probably you can save it as .fbx which I can import ?

Thx!

Me? If so pm me your email...
Polynurb's solution is the real deal, but mine is a bit more simple but might be awkward

[Frank1000]

Hi Bograt, tried out to redo your test setup but so far didn't get any picture being projected yet

[Frank1000]

Polynurb,
do you think it is possible to simulate the use a specific lens in exchange to the standard lens of a laser projector and project a panoramic movie evenly on the inner walls of an 180-degree cylinder (panorama) ?

Frank

[polynurb]

*edit: forget what i wrote here.. this won't work the way i thought for several reasons. was too late that night..


-daniel

[Frank1000]

Hi Daniel,
ok sounds good for this test. Later on I need to project into a sphere and capsule (cylinder with half sphere on top and bot) as well.
Regarding laser I found this simulation picture

on Thomas' Gallery here in this forum.

Staying with this test with your suggestion, what would be the steps to achieve that ? Haven't been able to create such crisp projections as in your picture yet, how do you achieve that ? Then the next step I guess would be to model the lens and place it right.

Pls let me know.

[polynurb]

Thoma's image, i would guess, also uses a real lens setup shooting a collimated beam of light into a medium using SSS. a laser projector is still something very different.
i'd say that image rendered for a looooooong time, considering it is also produced using a much older version of maxwell w. slower cpus.

try this for a crisper image:
Radius R1 of lense 200mm /R2 infinite (flat) / diameter ~200mm / backfocal distance 365mm / "DLP" 150x150mm / distance to screen 2500-3000mm

***

i edited my previous post.. there was error in my reasoning.

however dome projection is possible.. but it is truly doodling with something not practically usable in maxwell

these systems are highly interesting though, so i'll post it in case anyone is interested:

here is some in depth infromation:

http://www.scielo.br/scielo.php?pid=S16 ... ci_arttext

http://www.freepatentsonline.com/6542184.html

which basically concludes in the need of an image like this:



this sort of remapping can be approximated by rectangular to polar filter in ps or better some HDR utility. (ps only works in 8 bit )


this has the square UV map from before:


this uses the crappy image i found online, converted to hdr.. sorry for the low quality.


-daniel