Jump to content

Talk:Scene graph

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

It really should be its own page considering BVHs could be used for much more. —Preceding unsigned comment added by 134.197.40.199 (talk) 09:45, 27 February 2009 (UTC)[reply]


You are absolutely right here. BVHs and Scene graphs are completely different things. The nodes of a scene graph usually contain bounding volumes (amongst other informations) so they are in most cases some part of a scene graph which leads a scene graph to forming some kind of hierarchy of bounding volumes (but that is not a BVH in common sense).

The purpose of a scenegraph, whether it uses bounding volumes or not, is solely for managing a three-dimensional scene for optimized rendering. It does not necessarily consist of bounding volumes (but most often will) so it is not always a bounding volume hierarchy.

BVHs are a geometric data structure like an Octree, kd-Tree, BIH, Grid and so on and are used for the same purpose as the other geometric data structures, namely fast ray intersection (raytracing), culling and collision detection.

Bertikrueger (talk) 17:40, 22 August 2009 (UTC)[reply]

They have since gotten their own article. -- Beland (talk) 23:21, 2 December 2014 (UTC)[reply]

graph

[edit]

could do with being organised to wikipedia article style

This article goes into great detail about optimization. That's fine, except that has nothing to do with the concept of a scene graph. Whoever decides to renovate this article, keep that in mind.the1physicist 21:07, 3 September 2005 (UTC)[reply]

29th september 2005

The article did go off on a tangent about optimization so I have added a-lot of new information, expert information, given the article a decent structure and tried to keep the existing comments still valid by giving them context and examples. David Gill

Well done, David, looks like excellent work to me. Mortene 18:48, 1 October 2005 (UTC)[reply]
[edit]

please link the BVH section on this page with the octree article.
Moreover this article should be splitted on numerous pages. — Preceding unsigned comment added by 82.225.49.61 (talkcontribs) 01:43, 23 December 2005‎

The octree article has since been linked in a later section. I disagree the article should be split as it now stands. -- Beland (talk) 23:13, 2 December 2014 (UTC)[reply]

basic reference

[edit]

i would like to see a pointer to:

James H. Clark, Hierarchical Geometric Models for Visible Surface Algorithms, Comm. of the ACM, vol. 19(10), pp. 547-554, 1976 http://portal.acm.org/citation.cfm?id=360354&coll=Portal&dl=ACM&CFID=9978356&CFTOKEN=74301375

it provides the basic argumentation as well as principal applications of scene graphs. in my opinion that paper marks the invention. 217.85.114.1 19:32, 26 January 2007 (UTC)[reply]

I just added it per your request. -- Beland (talk) 23:25, 2 December 2014 (UTC)[reply]

See Also

[edit]

I've taken the liberty of adding a See Also section with topics related to the components that may be incorporated into a scene graph. I don't know how close to the guidelines these related topics are, but it's better than having nothing to turn to next, especially for somone trying to understand a Scene Graph, esp if he wishes to understand how it works.

C4Cypher 16:52, 27 July 2007 (UTC)[reply]

HowTo

[edit]

Some sections, such as C++ virtual pointers and the actual naming of coding templates are vely likely in the spirit of a HOWTO and not an encyclopedia. Please remove, and shift to a corresponding entry in WikiBooks. Fgenolini (talk) 15:31, 15 December 2007 (UTC)[reply]

@Fgenolini: I think what you're talking about may have been removed at some point, but if not, feel free to edit the article to excise anything you find objectionable. -- Beland (talk) 23:29, 2 December 2014 (UTC)[reply]

Fuzzy/lack of consensus

[edit]

I removed this text:

The definition of a scene graph is fuzzy because programmers who implement scene graphs in applications — and, in particular, the games industry — take the basic principles and adapt these to suit particular applications. This means there is no consensus as to what a scene graph should be.

This was unreferenced and seemed somewhat contradictory, since the article then immediately went on to give a pretty clear definition. Yeah, it's not always exactly the same data structure across applications, but a general class of data structures - that is, a graph, as the name implies. -- Beland (talk) 23:10, 2 December 2014 (UTC)[reply]

Example/diagram needed

[edit]

The section "Scene graphs and spatial partitioning" could use a concrete example, and probably a diagram, to explain the hybrid data structures it's talking about. In fact, the whole article could use more illustrations. -- Beland (talk) 23:20, 2 December 2014 (UTC)[reply]

leading header

[edit]

"This article is about general data structure for vector-based graphics. For the software component used in user interfaces, see Canvas (GUI)."

I dont really buy that. Firstly scene graphs are technology independent - they can be for vector based, raster based, ray traced, etc ... . Now i accept that there is cross over i the 'vector' usage there, but 'vector graphics' referes to a specific form of graphics ie OpenVG ! Also - I dont think that the concept of 'canvas' is particulary established - why not 'glyphs' instead.

I nominate for removal