do you think terragen could do this?

[Tangled-Universe]

Yes.... it can fake it, but it doesn't look as good as the real thing.

And radiosity is probably very important too

I don't really understand what radiosity is, but I always thought it was the same as Global Illumination.
It's just another way of solving it. Isn't it?

- Terje

Yes indeed, it is a type of GI, only another algorithm.
It should be quite easy to implement in the TG2-renderer, but it will probably very very slow.

Like you I think these renders indeed could benefit from subsurface-scattering or maybe of clever usage of the Lambert shader like PG said.

@PG: I can't find the image you refer to, it wasn't very long time ago, but if I remember correctly it was done by adding a reflective shader layer beneath the waterplane.

[PorcupineFloyd]

You can try faking radiosity with experimenting with the colour of light that is reflected from surfaces. For example make specular colour from leafs a bit green and from the wall a bit gray etc.

[darthvader]

the plant models in tthose images look spectacular

[RogueNZ]

Ahh i see 

[Tangled-Universe]

Well, took some time to see if I could improve the lighting on models in TG.

I used a mixture of GI, AO and fill lights for this (AO and fill lights are very soft/weak) so the main effect is still GI (strength on surfaces= 2.5 e.g.)
Rendersettings were: detail 0.8, AA 8, GI 1/1, prepass on and cubic B-spline with bloom.
The leaves have a translucency of ~12. Tried very high reflectionvalues with AA bloom settings to get that diffuse look. Also added some haze to add to this effect which completely failed...lol
Looks ok though in overall, but not any better than this: http://img41.imagevenue.com/img.php?image=29104_TU-A-Brand-New-Day_122_877lo.jpg

Shall look if I can come up with another approach.

Martin

[Volker Harun]

The approach is a team-question. Take someone who is perfect in setting up a scene, another person who is a magician on shaders. Get somebody who will do the atmospheres - and last not least, get a couple of people for setting up the lighting.

I doubt that a single sun with GI will do within reasonable rendertimes. You'll need fill-lights, spots etc. -TU is on a very good way. But he'll need some help.

My 5 cents,
Volker

[Matt]

And radiosity is probably very important too

I don't really understand what radiosity is, but I always thought it was the same as Global Illumination.
It's just another way of solving it. Isn't it?

Pretty much, yes. Everyone wants to call it something different, and terminology is confused all the time. Global illumination in TG is just called... Global Illumination. The "radiosity" algorithm is not really a practical algorithm for this kind of scene.

Matt

[PG]

Radiosity's normally for small indoor scenes with sharp contrast between light and shadow. Normal illumination creates very hard edged shadows whereas radiosity allows for softer shadows and allows colour of surfaces to be scattered to nearby surfaces. Matt's right, it won't work properly here.

@Volker: Don't know if you were reffering to my method but I used 2 suns with their visible disc turned off and diameter cranked up to have the light more widely spread while not creating a grainy effect from haze. GI was set to 0

[Tangled-Universe]

Radiosity's normally for small indoor scenes with sharp contrast between light and shadow. Normal illumination creates very hard edged shadows whereas radiosity allows for softer shadows and allows colour of surfaces to be scattered to nearby surfaces. Matt's right, it won't work properly here.

@Volker: Don't know if you were reffering to my method but I used 2 suns with their visible disc turned off and diameter cranked up to have the light more widely spread while not creating a grainy effect from haze. GI was set to 0

I might try that to see if it makes a difference. Why did you turn off GI? And how much did you cranck up the diameter, to about 50 or so?

Martin

[PG]

renders quicker let's see sun diameter...nope. 19.5 and 15

[Matt]

Radiosity is one way to approximate global illumination. Terragen uses another. They are both trying to achieve (or approximate) the same goal - i.e. realistic global illumination. Radiosity actually lacks the ability to simulate some important aspects of global illumination, though. It only considers diffuse (Lambert) reflections. The reason I said Radiosity is not appropriate for this kind of scene is because there's a lot more than Lambertian surfaces in it, and in general there are better global illumination algorithms than Radiosity, although Radiosity is *possibly* still worth considering for some architectural visualisations where view-dependent detail in your global illumination is not so important and if you only have diffuse (Lambertian reflectance) surfaces. But I think you would only want to use it if the geometry were not very complex. There really aren't any advantages that I can think of.

What's confusing is that Vue calls one of its global illumination modes "global radiosity", but I don't think it uses the actual "Radiosity" algorithm.

[PorcupineFloyd]

Very interesting Matt.

About Vue - GR really looks like radiosity, considering how colours from surfaces are able to blend on others and the patterns they make if there are too few samples.

Is there any source for in-deep explanation of GI methods floating somewhere over the internet? I'd like to read something about it (other than Wikipedia).

[PG]

There's some more here. But I'm not sure if it's all radiosity or not. http://www.gamedev.net/reference/list.asp?categoryid=40#180

[PorcupineFloyd]

Very interesting reading. Thank you very much indeed

[dandelO]

Haha! This line is great! ...

Consider a simple room with only four walls, a ceiling and a floor. Can you see it in your mind's eye? You better not be able to; I haven't specified a light source yet. :-)

Great information page, cheers!