Talk:Global illumination

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Global Illumination[edit]

In real-time 3D graphics, global illumination is sometimes approximated by an "ambient" term in the lighting equation. —Preceding unsigned comment added by 217.215.68.96 (talkcontribs)

I don't agree that (first line of article) Global Illumination is a name for a group of algorithms. It is a phenomenon. And there is a group of algorithms that simulates it. Isn't that better? 213.84.65.201 (talk) 09:54, 5 October 2010 (UTC)
Sorry, I screwed up my last post.
Global illumination is very useful and would be responsible for effects such as when light enters a window, it does not just illuminate a patch of floor, but instead the whole room lights up from the "ambient" light. That is what Global Illumination does.
—Preceding unsigned comment added by 72.148.17.13 (talkcontribs)
This is a wiki, just edit your comment. But don't edit other people's comments. And sign your comments with four tilda's (~~~~)
Imroy 05:13, 27 May 2006 (UTC)
I've never heard of the "ambient" tag either, and I think it's a little confusing since many software packages already use the term to describe basic overall lighting. Maya uses it on most of the shaders to describe the minimum color of an object before light hits it, and I remember running into "ambient" or "ambience" in several other packages before most of them started using GI. Which software package in particular uses that term for GI/radiosity/whatever?
—Preceding unsigned comment added by 69.203.23.174 (talkcontribs)

The ambient term is used in many lighting models, e.g Phong shading.

To mimic global conditions an ambient element is added to give general illumination, usually as a constant value
Ig = Iaka

So the amount of ambient light (Ia) is constant for the whole scene. Global illumination removes this assumption by finding the amount of light falling on a given area or point. But it takes a lot of computation, so it's not something we going to be doing in real-time with current hardware and methods. Imroy 05:13, 27 May 2006 (UTC)

Made a bunch of edits today, trying to make the article a bit more clear. I also included some practical information about the ambient term and a link to a video that covers it in a bit more depth. I hope these edits help visitors get a better feel for the purpose of GI and its advantages. Maruchan 05:15, 29 August 2006 (UTC)

Are hard shadows part of Global Illumination?[edit]

Here's a small question from the Radiosity discussions. Someone is saying that hard shadows are considered to be Global Illumination, but I can't find a second source for this. Can anyone find a reference or paper stating that hard shadows are considered to be part of Global Illumination? Rocketmagnet 14:54, 9 January 2007 (UTC)

Rocketmagnet 14:54, 9 January 2007 (UTC)

Local Illumination vs Global Illumination[edit]

IMHO the article miss a sentence clearly explaining the origin of the Global term. That is Global was in counter position to local illumination models, e.g. the ones that you can compute without a global knowledge of the scene using only local information. ALoopingIcon 14:44, 10 May 2007 (UTC)

Hmm, interesting. There was some talk about something similar in the Radiosity discussion. Trevorgoodchild mentioned that he thought that shadows were considered global illumination. I disagreed, but no reference was found to prove it either way I think. Anyway, Googling for "global local illumination" brings up lots of stuff about the difference between global and local. I think there might be some confusion because "global" seems to be the opposite of both "local" and "direct". So, while shadows might not be included "local" lighting, does that place them in "global" as TGC suggests, or perhaps in "direct" instead. Can anyone clear this up, or anyone know of a definitive paper? Rocketmagnet 00:06, 11 May 2007 (UTC)

External Links needs cleanup[edit]

Too much promotion is happening in EL. If it is in the interest of readers to find links to 3D software, they can find such a list elsewhere on Wikipedia. Otherwise, it's obvious that people are just posting links to their favorite software packages' websites or links to packages that they are paid to promote. Ideas? Maruchan 15:58, 17 September 2007 (UTC)

I noticed that edit too, and thought it looked out of place, but didn't change it (assuming the site has a gallery, which other links also have). In general, though, a link to the official Lightwave site is probably best in LightWave. "Makers of" links are very dubious — nobody is likely to care who made it (Mental Images – makers of the Mental Ray renderer should be changed to a wikilink and moved to "See Also"). "The site has an extensive gallery of contributed images." also occurs twice in the article.
I think we should wikilink to notable free packages, notable commercial packages, and external-link to galleries. If we do that without removing information, then we'll have a much smaller list to prune. ⇌Elektron 14:16, 18 September 2007 (UTC)
OK, since it was rather obvious that so many ELs were a violation of WP:EL, I pared the list down to match the recommendations given on the policy page. I'm not even sure if linking to only commercially-available books is worthwhile and relevant (I'm sure there are hundreds of such books out there that could be shoehorned into "relevant" status), so any feedback in that area is welcome. —Preceding unsigned comment added by Maruchan (talkcontribs) 18:59, 29 January 2008 (UTC)

Internal Links need cleanup[edit]

Just as in the external links section, it seems that the wikilinks section is being used to promote commercial software (and perhaps non-commercial software as well). This sort of stuff belongs in a category listing (like "software that uses global illumination"), since it doesn't really help seekers of information related Global Illumination. I think a way to find GI software is important, but not in the wikilinks area. And probably not in a list on the page - it's embarrassing that the internal and external links sections were about to become longer than the article itself. Maruchan (talk) 19:07, 29 January 2008 (UTC)

OK, I put all the wikilinks software in a category and put THAT

Unbiased rendering[edit]

This article redirects from Unbiased rendering but it doesn't even make any mention of unbiased versus biased techniques: That's pretty terrible ;) Errantkid (talk) 16:03, 28 October 2009 (UTC)

Faked Global Illumination Section[edit]

The Faked Global Illumination section really needs some help. It's pretty much unreadable. —Preceding unsigned comment added by 12.52.185.30 (talk) 13:32, 6 April 2010 (UTC)

What scene is it in the images?[edit]

I'm wondering about the scene used to render the images, is it a common test scene? In that case, what is it called and where can I find it? I am creating a list of common 3D test models (still work to do there though...). —Kri (talk) 01:00, 18 July 2011 (UTC)

Methods[edit]

It seems to me very uninformative to show a few different ways how to rewrite an equation. Methods to me mean the method of solution. For example inversion and iteration are both used for radiosity which is a quadrature method at heart(geometric discretisation and integrate), inversion is how one would solve radiosity equations if they didn't care about efficiency at all and it's not used in any real method. Also the Statistical methods biased/unbiased. And of course the numerous approximations that usually involve discretisation in a different way like splitting radiance field into point lights for example. Three equations with incorrect comments following them is not doing it justice. -TimW