Boyd Stephen Smith Jr. put forth on 1/6/2010 9:20 AM: > Technically, OpenGL was designed with network transparency in mind, but it > doesn't always work well when large textures and/or changing textures or > geometries are involved.
Waaay back in the day this was true, when an SGI, SUN, HP, Intergraph, or IBM workstation cost $20-$40K USD, and before 3D chips were common. Companies would have four engineers with X terminals sharing one workstation. In the mid-late 90's the model started to change with the introduction of the Pentium Pro, the Glint chip, and Windows NT4. At that point companies could afford to put a $10K USD workstation with better 3D graphics performance on each engineer's desk for about the same capital outlay. This trend was accelerated by the proliferation of ultra high performance, inexpensive, volume 3D graphics chips (3Dlabs, nVidia) going into the consumer space. Due to the average PC cost, OpenGL architecture shifted from a multiuser to a singe user focus. I actually recall reading a Linux announcement some years ago that mutiuser 3D OpenGL would no longer be supported, period. I think this was back in '02 or '03. To make OpenGL really scream on single user 3D chips, they had to eliminate over the network OpenGL completely, as keeping that capability would have totally hosed the rendering pipeline performance for 3D chips. -- Stan -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
