> On Nvidia hardware that is also used for display, there is a hard limit > of 5 seconds of kernel execution time.
OK?! That seemed to be it. FYI I have now split the problem into subproblems and the final solution is easily build from the subresults. Thanks for the info! _______________________________________________ PyOpenCL mailing list [email protected] http://lists.tiker.net/listinfo/pyopencl
