 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Orchid XP v2 nous apporta ses lumieres en ce 2005-08-14 04:23:
>
> I did an image a while back using a 256x256 area light + radiosity. It
> didn't take that long...
>
> I have a vague feeling that "adaptive" only works properly for a power
> of two - anyone confirm/deny?
>
> Anyway... I presume the grooves are an isosurface? What about the liquid
> - what's that? Does it use some kind of media?
>
It's power of 2 PLUS 1 for optimum subdivisions. Adaptive 0 start at 2*2, adaptive 1
at 3*3,
adaptive 2 at 5*5, adaptive 3 at 9*9, adaptive 4 at 17*17...
You should start at adaptive 0 and increase if you get artefacts.
Alain
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>> I have a vague feeling that "adaptive" only works properly for a power
>> of two - anyone confirm/deny?
>>
> It's power of 2 PLUS 1 for optimum subdivisions. Adaptive 0 start at
> 2*2, adaptive 1 at 3*3, adaptive 2 at 5*5, adaptive 3 at 9*9, adaptive 4
> at 17*17...
> You should start at adaptive 0 and increase if you get artefacts.
Right. Glad we cleared that up...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> > Just about, I guess. =) My name actually comes from the Slime monster in
> > Dragon Warrior games.
>
> ...so now you're a monster? :-|
Yeah, but I only have like 3 or 4 HP so don't worry about it.
> I did an image a while back using a 256x256 area light + radiosity. It
> didn't take that long...
Without adaptive? Keep in mind that as the rays travel through the liquid,
they reflect and refract and end up hitting a lot of surfaces (the max trace
level of 5 is reached), and for every intersection the area light has to be
sampled for shadows. I just don't want to think what radiosity will add to
that (it was slow enough without the liquid part).
> Anyway... I presume the grooves are an isosurface? What about the liquid
> - what's that? Does it use some kind of media?
Yes, the grooves are an isosurface, as is the liquid, which makes use of the
first isosurface's function to "cut" the shape of the grooves (plus surface
tension) out of itself. The liquid is reflective and refractive, and has
some emission and absorption media (which is what most of the color is
from).
- Slime
[ http://www.slimeland.com/ ]
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>>>Just about, I guess. =) My name actually comes from the Slime monster in
>>>Dragon Warrior games.
>>
>>...so now you're a monster? :-|
>
>
>
> Yeah, but I only have like 3 or 4 HP so don't worry about it.
...right...I'm going to stop persuing this line of conversation now...
>>I did an image a while back using a 256x256 area light + radiosity. It
>>didn't take that long...
>
>
> Without adaptive?
:-|
Hell no!! That's *with* adaptive. ;-)
> Keep in mind that as the rays travel through the liquid,
> they reflect and refract and end up hitting a lot of surfaces (the max trace
> level of 5 is reached), and for every intersection the area light has to be
> sampled for shadows. I just don't want to think what radiosity will add to
> that (it was slow enough without the liquid part).
They only reflect and refract when they *hit* the liquid, not as they
travel through it. (Although that might make it look even better - but I
don't wanna think about how slow that would be!)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
