On 12/6/05, Dirk Reiners <[EMAIL PROTECTED]> wrote: > > Hi Josef, > > On Mon, 2005-12-05 at 20:12 +0000, Josef Grunig wrote: > > Many thanks to Andreas Zieringer for answering me to the "Using Vertex > > Textures" post. > > > > Now I'm a little confused on how floating point (32 bit or 16 bit) > > textures work when used in a shader. What I need is that the float > > values of the texture (in an unspecified range) can be fetched from a > > shader program (vertex or fragment) without loss of precision and > > without range change or clamping. Is this possible? Which is the right > > pixel format/type to use? Should I specify a particular > > InternalFormat? > > Float texture values are not clamped or manipulated by OpenGL, you > should be fine just using them with any float format. Note that color > values are clamped at different stages of the pipeline, so to pass the > values between vertex and fragment and to store them you should not use > the color variables. > > Dirk > >
I got it working using samplerRECT and texture2DRect in fragment shader, with GL_TEXTURE_RECTANGLE_NV target and GL_FLOAT_RGBA32_NV internal format in my code. Using Texture2D I was receiving clamped values in 0-1 range as explainded here: http://sourceforge.net/mailarchive/message.php?msg_id=11579060 Many thanks, -- Josef Grunig www.blending-life.org ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_idv37&alloc_id865&op=click _______________________________________________ Opensg-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/opensg-users
