Linux Today: Linux News On Internet Time.
Search Linux Today
Linux News Sections:  Developer -  High Performance -  Infrastructure -  IT Management -  Security -  Storage -
Linux Today Navigation
LT Home
Contribute
Contribute
Link to Us
Linux Jobs


Top White Papers

  • Is your sales and operations planning helping or hurting your bottom line? Here are 5 useful tips from the experts at Quintiq to guide you to a better...
    Download

  • On-demand Event Event Date: June 12, 2014 In the past, IT service management (ITSM) and IT infrastructure library (ITIL) basics might have been more...
    Download

More on LinuxToday


NVidia Releases Open Source OS-Neutral Driver Suite for RIVA Chips

Aug 16, 1999, 05:05 (5 Talkback[s])

[ Linux Today reader ewhac writes: ]

Several weeks ago, the linux-nvidia list (formerly the Riva Enlightenment Project) reported that NVidia released a binary-only OS-neutral driver core for their chips. Well, NVidia has released an update, and this time they're serving up source! (Note: This is not the same as the XFree86 source release a few months back; this suite is more low-level and extensible.)

NVIDIA has opened up an area on their Web site, where is contained an OS-neutral driver core. The suite is intended for developers to write their own drivers and 3D accelerators for the RIVA 128, TNT, and TNT2 chips on the platform of their choice. Materials available so far include an implementation of the NVIDIA Multimedia Architecture, provided as compilable source code, as well as several linkable object files compiled with gcc and VisualC 5.0). The source is provided under what appears to be a BSD-style license. Alas, the source code isn't very educational, as it has been run through cpp. Still, this puts it head-and-shoulders above 3Dfx's GLIDE.

Also included in the distribution are OS interface documentation, rendering primitive documentation, and source code to example programs, test programs, and sample driver implementations for Linux and Windows-NT.

My reading of the docs suggests that the NVIDIA API supports impressive levels of concurrency at a hardware level through multiple independent contexts and an object-oriented-like interface. It looks like there's enough information here to make high-performance 3D-in-a-window comparatively easy to achieve. There's also nothing preventing multi-headed support.