Philipp Klaus Krause wrote:
Look at the output of Mesa/progs/tests/texwrap
With the i915 driver there's no green without a border.
Indirect rendering and the r128 driver always show
the area around the [0,1]x[0,1] red aread in green.
What is the supposed display with GL_REPEAT?
With indirect rendering there is no green grid.
With the i915 driver there is exactly when the
texture has a border. With the r128 there's always
a green grid.
GL texture borders and clamping are one area where the hardware tends to
diverge from the GL spec (typically by favoring DX usage). It's also an
area where fallbacks can really hurt performance, so you will see
drivers doing there best to get as close to spec as possible, while
allowing some non-conformance to avoid the performance penalty of a
fallback. The r128 and i915 drivers I believe take slightly different
approaches to the tradeoff.
Keith
-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
--
_______________________________________________
Dri-devel mailing list
Dri-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dri-devel
