Talk:Viewing frustum

This article is rated Stub-class on Wikipedia's content assessment scale.
It is of interest to the following WikiProjects:

Computer graphics

	This article is within the scope of WikiProject Computer graphics, a collaborative effort to improve the coverage of computer graphics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.Computer graphicsWikipedia:WikiProject Computer graphicsTemplate:WikiProject Computer graphicscomputer graphics articles
???	This article has not yet received a rating on the project's importance scale.
	This article has been automatically rated by a bot or other tool as Stub-class because it uses a stub template. Please ensure the assessment is correct before removing the `\|auto=` parameter.

Computing

	This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing articles
???	This article has not yet received a rating on the project's importance scale.
	This article has been automatically rated by a bot or other tool as Stub-class because it uses a stub template. Please ensure the assessment is correct before removing the `\|auto=` parameter.

Untitled

"Often, the far plane is placed infinitely far away from the camera so all objects within the frustum are drawn regardless of their distance from the camera." This sounds wrong (for realtime 3d graphics), as with an infinite far-plane you'd get massive z-fighting after a certain distance. Bitplane 17:41, 11 November 2006 (UTC)

Engines don't have to use standard depth sorting to get the z-order. For applications where detail is more important than speed they may do the sorting manually, and draw things in the desired order. This method has other problems obviously, but it can still be done in real time. You could also raise the amount of bits used for the depth buffer. A lot of people use 8, but 16 bits is commonly supported (AFAIK), and 24 is also available even on cards a few years old. That gives you an exponentially bigger depth before Z-fighting kicks in. In other words, z-fighting is a problem with infinite far planes, but it's not insurmountable. Prgrmr@wrk 15:08, 25 February 2007 (UTC)[reply]

Point of far plane?

Frustums were brought into computer graphics to facilitate realistic transformations, so 3D objects appear correctly. Another reason was so objects in the same line of sight would be properly occluded, far objects blocked by closer ones, taking account for distance shrinkage. Frustums were also introduced to facilitate culling, the elimination of objects that don't fit.

Transformation facilitation is aided by all six planes. A 3D object is positioned some distance from them, and a vector expresses this, which applied in matrices transforms the object into 2D space.

Near objects block distant ones because the z-axis, which extends into viewing depth, ranks objects from near to far. Only near ones are rendered.

The third frustum purpose, culling, seems evident for four of the planes - the sides, bottom, and top planes. Anything further to the side, above the top, and below the bottom are invisible, so eliminate them from further transforms. The near plane's purpose is obvious - it coincides with the screen, or another viewing surface. But the far plane, useful for positioning and ranking objects, seems arbitrary for culling, because objects beyond it aren't really hidden.

Yet distant objects are culled in most graphics engines.

It's possible that a very high angled plane has a vanishing point within the viewport. This can result in depth conflict. Yet even before this occurs, the frustum's rear plane function is violated.

— Preceding unsigned (talk) 23:28, 13 June 2015 (UTC)[reply]

Point of near plane?

Obviously in the real world there is no "near plane" - objects very close to a real camera or a real eye are not invisible (though they may appear blurry). Thus there should be an explanation given why a "near plane" is used in CGI. -- 84.63.225.161 (talk) 17:12, 1 January 2011 (UTC)[reply]

merge

I suggest merging the entire article pyramid of vision into the viewing frustum article. They are similar enough that a single article can cover both of them, with a short description of the difference. --DavidCary (talk) 19:50, 30 August 2011 (UTC)[reply]

Merge: without looking into it much, merge based on scope, size and lack of sources. Widefox; talk 12:10, 16 April 2013 (UTC)[reply]

Maybe another merge

Viewing frustum culling links to this article, but View frustum culling links to a section of Hidden-surface determination. (Neither article is very informative regarding how it is done.) Card Zero (talk) 15:54, 31 July 2023 (UTC)[reply]