Editing Shadow volume (section)

== Optimization ==

* One method of speeding up the shadow volume geometry calculations is to utilize existing parts of the rendering pipeline to do some of the calculation. For instance, by using [[homogeneous coordinates]], the ''w''-coordinate may be set to zero to extend a point to infinity. This should be accompanied by a [[viewing frustum]] that has a far clipping plane that extends to infinity in order to accommodate those points, accomplished by using a specialized projection matrix. This technique reduces the accuracy of the depth buffer slightly, but the difference is usually negligible. See 2002 paper "Practical and Robust Stenciled Shadow Volumes for Hardware-Accelerated Rendering", C. Everitt and [[Mark Kilgard|M. Kilgard]], for a detailed implementation.
* Rasterization time of the shadow volumes can be reduced by using an in-hardware scissor test to limit the shadows to a specific onscreen rectangle.
* Single pass is achievable by using shader built-in variable for detecting front faces instead of two separated passes for culling each the other,combine with replacing stencil buffer with arithmetic blending on color buffer.
* [[NVIDIA]] has implemented a hardware capability called the ''depth bounds test'' that is designed to remove parts of shadow volumes that do not affect the visible scene. (This has been available since the [[GeForce]] FX 5900 model.) A discussion of this capability and its use with shadow volumes was presented at the [[Game Developers Conference]] in 2005.<ref>{{cite web|last=Lengyel|first=Eric|title=Advanced Stencil Shadow and Penumbral Wedge Rendering|url=http://www.terathon.com/gdc05_lengyel.pdf|work=Game Developers Conference 2005|publisher=2005|accessdate=18 October 2012}}</ref>
* Since the depth-fail method only offers an advantage over depth-pass in the special case where the eye is within a shadow volume, it is preferable to check for this case, and use depth-pass wherever possible. This avoids both the unnecessary back-capping (and the associated rasterization) for cases where depth-fail is unnecessary, as well as the problem of appropriately front-capping for special cases of depth-pass{{Citation needed|date=February 2019}}.
* On more recent GPU pipelines, [[geometry shader]]s can be used to generate the shadow volumes.<ref>{{cite web |url=http://developer.nvidia.com/node/168 |title=GPU Gems 3: Chapter 11. Efficient and Robust Shadow Volumes Using Hierarchical Occlusion Culling and Geometry Shaders &#124; NVIDIA Developer Zone |website=developer.nvidia.com |access-date=12 January 2022 |archive-url=https://web.archive.org/web/20110516024500/http://developer.nvidia.com/node/168 |archive-date=16 May 2011 |url-status=dead}}</ref><ref>{{cite web|last=Stich |first=Martin |title=Chapter 11 "Efficient and Robust Shadow Volumes Using Hierarchical Occlusion Culling and Geometry Shaders" |url=https://developer.nvidia.com/gpugems/GPUGems3/gpugems3_ch11.html |work=GPU Gems 3 |publisher=nVidia / Addison-Wesley |accessdate=18 October 2012 |author2=Carsten Wächter |author3=Alexander Keller |archiveurl=https://web.archive.org/web/20110516024500/http://developer.nvidia.com/node/168 |archivedate=May 16, 2011 |location=archive.org |year=2007 }}</ref>
* On systems that do not support geometry shaders, [[vertex shaders]] can also be used to create shadow volumes by selectively extruding vertices that already reside within GPU memory.<ref>{{Cite web|url=http://developer.amd.com/wordpress/media/2012/10/ShaderX_ShadowExtrusion.pdf|title=Shadow Volume Extrusion using a Vertex Shader|last=Brennan|first=Chris|date=|website=AMD|access-date=2018-02-14}}</ref>