<p>Using gpu for number crunching and not showing pixels would be best dpne with vendor drivers. </p>
<p><blockquote type="cite">On Mar 29, 2010 8:33 PM, "jrtroberts" <<a href="mailto:jrtroberts@gmail.com">jrtroberts@gmail.com</a>> wrote:<br><br>
<div bgcolor="#ffffff" text="#000000">
This is very interesting. I will have to see what I can learn. My
issue is that I am starting to research the use of CUDA/GPU offloading
with blender. And since Nvidia seems to have the best workstation
video cards hands down, like the quadro fx , quadro plex, and tesla
series cards, I am kinda stuck when it comes to GPU integration into
the rendering pipeline. So hopefully as long as I am only using the
proprietary drivers and not trying to use anything that creates
conflicts, I should be ok. Cross my fingers anyway. And it is kind of
funny that they would stop open source support since the whole CUDA
open source side of their cards relies on the community of users to
create implementations for CUDA to be successful. Oh well, guess that
all coins have two sides.<br>
<br>
Thanks for the update. Even though Nvidia is not being very friendly,
for the kind of work I want to do they are the best option that I know
of.<br>
<br>
Joshua<p><font color="#500050"><br><br><br>Richard Faulkner wrote:<br>><br>> I'm a *YES* vote for reading on your kernel mode experiences. I'm<br>> ...</font></p></div>
<br>_______________________________________________<br>
Ale mailing list<br>
<a href="mailto:Ale@ale.org">Ale@ale.org</a><br>
<a href="http://mail.ale.org/mailman/listinfo/ale" target="_blank">http://mail.ale.org/mailman/listinfo/ale</a><br>
See JOBS, ANNOUNCE and SCHOOLS lists at<br>
<a href="http://mail.ale.org/mailman/listinfo" target="_blank">http://mail.ale.org/mailman/listinfo</a><br>
<br></blockquote></p>