 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Here's another test scene (59596 bytes) using Nathan's photon mapping
implementation. Both reflective and refractive caustics can be seen. As
always, comments and helpful tips are welcome.
- Alan
Post a reply to this message
Attachments:
Download 'rflcaust08.jpg' (59 KB)
Preview of image 'rflcaust08.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Alan Kong wrote:
>
> Here's another test scene (59596 bytes) using Nathan's photon mapping
> implementation. Both reflective and refractive caustics can be seen. As
> always, comments and helpful tips are welcome.
>
> - Alan
Ok I will post the source for my disco ball if you share with me the
finish statement used on the reflective cube. I'll clean up the mess it
is in now and post it in a couple of hours in the binaries.scene-files
group. The ball itself is made from a mesh object so is a bit too large
for the text group.
That is some pretty stuff you got there and I think I have just witnessed
the birth of a whole new breed of raytracing. Did I mention that's real purdy ?
Mr. Nathan Kopp Sir.,
Your not going to start charging for this are you ?
--
Ken Tyler
mailto://tylereng@pacbell.net
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Can't help but love it.
If anything at all (can not resist), the reflective spheres look like
they could have used a brighter reflective caustic to be obvious.
Was the same photon density used on all objects?
Alan Kong wrote:
>
> Here's another test scene (59596 bytes) using Nathan's photon mapping
> implementation. Both reflective and refractive caustics can be seen. As
> always, comments and helpful tips are welcome.
>
> - Alan
>
> ------------------------------------------------------------------------
> [Image]
--
omniVERSE: beyond the universe
http://members.aol.com/inversez/homepage.htm
mailto:inv### [at] aol com?Subject=PoV-News
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
This is really cool! I want to point out that, though a spotlight is used,
the photon-tracing part treats it like a point light (so photons can light
up stuff outside of the cone of the spot light... that part of the code
isn't finished quite yet).
About the reflective caustics from the spheres... remember that the light
gets spread out (I can't think of an antonym for focused) very much, so
it really shouldn't be very bright (I like what you have here... it is
visible yet subtle).
-Nathan
Alan Kong wrote:
>
> Here's another test scene (59596 bytes) using Nathan's photon mapping
> implementation. Both reflective and refractive caustics can be seen. As
> always, comments and helpful tips are welcome.
>
> - Alan
>
> -------------------------------------------------------------------------------
> [Image]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Nathan Kopp wrote in message <36F562A1.73E6E9AF@Kopp.com>...
>
>I can't think of an antonym for focused
Diffused, no?
Margus
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ken wrote:
>> Ok I will post the source for my disco ball if you share with me the
>> finish statement used on the reflective cube. I'll clean up the mess it
>> is in now and post it in a couple of hours in the binaries.scene-files
>> group.
I used ambient 0 for the four object plus the floor box. Little too dark
for my tastes so I used a shadowless fill light in addition to the spotlight
shining onto the 'mirror'. I differenced the backside of the mirror during
an earlier test rendering because the reflective caustics were putting light
where I didn't want it to show when I was experimenting with additional
lighting placement.
Thanks to everyone for the kind comments. Don't use my photon mapping
settings as gospel. I have tried to read all of Nathan's posts concerning
appropriate settings but I may have changed too many items at one time and
quite possibly went the wrong way...<s>. Many thanks to Nathan for his
fantastic work on implementing photon mapping into UVpov.
I did find one shortcoming during final traces - I need more memory! I was
hitting the swap file much too soon with 64mb RAM. Here are the stats for an
800w by 600h, +am2 +a0.2, on a Pentium 200MMX.
Shadow Ray Tests: 360457 Succeeded: 63626
Reflected Rays: 3076492
Refracted Rays: 1433909
Number of photons shot: 2816131
Number of photons stored: 2098198
Priority queue insert: 287930609
Priority queue remove: 118863033
----------------------------------------------------------------------------
Smallest Alloc: 28 bytes Largest: 16384024
Peak memory used: 102953138 bytes
----------------------------------------------------------------------------
Time For Parse: 0 hours 25 minutes 48.0 seconds (1548 seconds)
Time For Trace: 2 hours 59 minutes 25.0 seconds (10765 seconds)
Total Time: 3 hours 25 minutes 13.0 seconds (12313 seconds)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Oops. You wanted the finish statement of the reflective box in my scene,
didn't you, Ken - I forgot to attach the scene source to my previous post so
here it is.
- Alan
Post a reply to this message
Attachments:
Download 'rflcaust08.pov.txt' (5 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Bob Hughes wrote:
>> Can't help but love it.
>> If anything at all (can not resist), the reflective spheres look like
>> they could have used a brighter reflective caustic to be obvious.
>> Was the same photon density used on all objects?
Yes, the same density on all objects. I played around with the density and
didn't see much difference after a certain point (but I've forgotten which
point that was). I'll venture that it *might* be okay 'as is' since the
reflective spheres aren't being directly hit by the spotlight. Still, if I
can figure out how to increase the reflective caustic just enough to avoid
an overly long rendering time I might check it out.
- Alan
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Nathan Kopp wrote:
>> This is really cool! I want to point out that, though a spotlight is used,
>> the photon-tracing part treats it like a point light (so photons can light
>> up stuff outside of the cone of the spot light... that part of the code
>> isn't finished quite yet).
I had started this scene before I saw one of your earlier posts concerning
how the photon patch treats all lights as point lights. Some head-scratching
took place when I saw the reflective caustics of the 'mirrored' box appear
across the floor surface. I also saw the effect you spoke of when decreasing
the density of the photons, it gives the impression of soft lighting effects
coming off that same box.
Thanks for your efforts, Nathan. Your comments, also, are invaluable as
it's so much trial and error for this humble user (yes, I do read all docs
on your pages).
- Alan
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Alan Kong wrote:
>
> Oops. You wanted the finish statement of the reflective box in my scene,
> didn't you, Ken - I forgot to attach the scene source to my previous post so
> here it is.
>
> - Alan
That's a fair enough swap. I just came here from posting the source
for the mirrored disco ball in the povray.binaries.scene-file group.
Thank you,
--
Ken Tyler
mailto://tylereng@pacbell.net
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
