Talk:Screen Space Ambient Occlusion

From Wikipedia, the free encyclopedia

This article is part of WikiProject Graphic design, which collaborates on graphic design-related subjects on Wikipedia. If you would like to participate, you can edit the article attached to this page, or visit the project page, where you can join the project and/or contribute to the discussion.
??? This article has not yet been rated on the assessment scale.

Please rate this article and leave comments here to explain the ratings and/or to identify the strengths and weaknesses of the article.

This article is within the scope of WikiProject Computer science, which aims to create a comprehensive computer science reference for Wikipedia. Visit the project page for more information and to join in on related discussions.
??? This article has not yet received a rating on the assessment scale.
??? This article has not yet received an importance rating on the assessment scale.

Contents

[edit] Propose for deletion

The page seems a blatant advertisment for the Crytech engine. It uses a lot of WP:PEACOCK terms without saying how it really works. It does not cite any reliable source. The page has been edited mostly by a single user who has also edited crytech related pages. On the web the term of this page is connected in most of the case with the abovely cited engine. Moreover the technique was presented not in a game but in a conference:
Real-Time Ambient Occlusion for Dynamic Character Skins
Adam Kirk Okan Arikan
Symposium on Interactive 3D Graphics and Games (I3D) 2007

I propose this page for deletion. ALoopingIcon (talk) 15:33, 18 December 2007 (UTC)

[edit] comment on proposal to delete

I am a game developer with no connection to Crytek at all, and I'm new to wikipedia. While the article is brief - and the idea of screen based ambient occlusion has been presented before, there is no doubt that Crytek IS the FIRST to impliment the technique in realtimne in a game environment. The information presented I would not describe as peacock terms at all, and while it doesn't explain the technique clearly it is a reasonable approximation of crytek's siggraph paper.

It seems too harsh to delete the article (which is in no way an advertisment for Crytek) and I would suggest it stays - though it could do with expanding by someone able to explain the technique more clearly. From my research there are several ways of generating SSAO - each with their own advantages/disadvantages and visual artifacts. Perhaps I'll have a go myself. --81.179.107.202 (talk) 22:34, 18 December 2007 (UTC)

[edit] Additional comment on proposal to delete.

I'm also a software developer with no connection to Crytek, and also new to Wikipedia. This article being flagged for deletion was enough to finally get me to register. :)

The prior poster is correct - although this article is light on details, the information provided outlines the technique laid out in the paper, and isn't an advertisment. I'm going to pull the request for deletion.

While there are multiple ways to performing SSAO, virtually every article I've run across has cited Crytek's Siggraph paper as a starting point. This technique is quite new, and finding additional information on the web is a bit difficult. I don't doubt in the next year that a **lot** more information will become available.

It would be great if this topic were expanded! For example, there's no mention on the use of surface normals or blurring the resulting AO map, which seem to be common to methods I've seen.

It would be useful to cite some prior art as well. For example, there's Luft's 2006 paper "Image Enhancement by Unsharp Masking the Depth Buffer" that covers a **very** similar process. There's also a very similar technique described as "Ambient Occlusive Crease Shading" here: http://www.shalinor.com/code.html

Dcuny (talk) 11:00, 20 December 2007 (UTC)

you are quite right - it would be fantastic if we could get a page describing the evolution of this technique. i know some good papers and websites that are covering some very early techniques in this area. also, i still think if there should be a overall category for this, since SSAO is just one technique resulting of all the research done here. Craycrone (talk) 08:49, 10 April 2008 (UTC)

[edit] Extension to SSAO - HSAO

if the article stays it could get expanded by this new technique shown at the GDC08 by NVIDIA: Horizon Split Ambient Occlusion, HSAO http://developer.download.nvidia.com/presentations/2008/GDC/GDC08_Ambient_Occlusion.pdf

developed by http://www.msainz.org Miguel Sainz and two co-authors

Rouslan Dimitrov and Louis Bavoil

also, i would suggest a new category for the entire subject itself as it is mentioned in the paper from nvidia, Real-Time Depth Buffer Based Ambient Occlusion as it describes the overall technique and not a specific one like ssao or hsao. Craycrone (talk) 08:50, 10 April 2008 (UTC)

[edit] SSAO in Leadwerks

I am not advertising or anything, but Leadwerks Engine has had SSAO working as a fragment shader since before Crysis was released. The engine just has not been released to the public until recently.

I believe this article is more of an ad for Crytek, but anyways, Crytek did not INVENT SSAO, there were college papers on the technique, so I do not believe the Crytek Shader Programmer should be credited...

I will post another comment with shader source code to show how the shader works. It is basically reading linear depth of the buffer and darkening/lighting pixels based on the depth (no matter what distance)

Tylerp9p (talk) 13:56, 5 May 2008 (UTC)

Vlad3D (talk) 07 May 2008

I removed statement about invention from this article so people don't kill each other :). About shader sources - they are available for anyone as part of Crysis public demo. —Preceding unsigned comment added by Vlad3D (talkcontribs) 10:48, 7 May 2008 (UTC)

My apologies. I just have a pet-peeve for people claiming to invent things that have college papers/research done on them previously. I was about to post something about the guy who claimed to have invented HDR for games, but then it was fixed by someone else.Tylerp9p (talk) 13:42, 22 May 2008 (UTC)