Editing Volume rendering (section)

==Hardware-accelerated volume rendering==
Due to the extremely parallel nature of direct volume rendering, special purpose volume rendering hardware was a rich research topic before [[GPU]] volume rendering became fast enough. The most widely cited technology was the VolumePro real-time ray-casting system, developed by [[Hanspeter Pfister]] and scientists at [[Mitsubishi Electric Research Laboratories]],<ref name="phi99">{{Cite book|doi = 10.1145/311535.311563|last1 = Pfister|first1 = Hanspeter|last2 = Hardenbergh|first2 = Jan|last3 = Knittel|first3 = Jim|last4 = Lauer|first4 = Hugh|last5 = Seiler|first5 = Larry| title=Proceedings of the 26th annual conference on Computer graphics and interactive techniques - SIGGRAPH '99 | chapter=The VolumePro real-time ray-casting system | date=1999 |isbn = 978-0201485608|pages = 251–260|citeseerx = 10.1.1.471.9205| s2cid=7673547 }}</ref> which used high memory bandwidth and brute force to render using the ray casting algorithm. The technology was transferred to TeraRecon, Inc. and two generations of ASICs were produced and sold. The VP1000<ref name="VP1000">{{cite book |last1=Wu |first1=Yin |last2=Bhatia |first2=Vishal |last3=Lauer |first3=Hugh |last4=Seiler |first4=Larry |title=Proceedings of the 2003 symposium on Interactive 3D graphics |chapter=Shear-image order ray casting volume rendering |date=2003 |page=152 |doi=10.1145/641480.641510 |isbn=978-1581136456 |s2cid=14641432 }}</ref> was released in 2002 and the VP2000<ref>{{cite web |last1=TeraRecon |title=Product Announcement |url=https://www.healthimaging.com/topics/diagnostic-imaging/terarecon-its-beginning-end-dedicated-diagnostic-advanced-visualization |website=healthimaging.com |date=7 December 2006 |access-date=27 August 2018}}</ref> in 2007.

[[Pixel shader]]s are able to read and write randomly from video memory and perform some basic mathematical and logical calculations. These [[Single instruction, multiple data|SIMD]] processors were used to perform general calculations such as rendering polygons and signal processing.  In recent [[GPU]] generations, the pixel shaders now are able to function as [[Multiple instruction, multiple data|MIMD]] processors (now able to independently branch) utilizing up to 1 GB of texture memory with floating point formats. With such power, virtually any algorithm with steps that can be performed in parallel, such as [[volume ray casting]] or [[tomographic reconstruction]], can be performed with tremendous acceleration. The programmable [[pixel shaders]] can be used to simulate variations in the characteristics of lighting, shadow, [[Reflection (computer graphics)|reflection]], emissive color and so forth. Such simulations can be written using high level [[shading language]]s.