Someone sent me a link to CUDA from Nvidia.
After looking at some of the source code it seems that there are two
(see below) includes that makes CUDA the numerical processor.
include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <GL/glew.h>
#include <GL/glut.h>
#include <cuda_runtime_api.h>
#include <cuda_gl_interop.h>
#include <cutil.h>
#include <cutil_interop.h>
#include <Mandelbrot_kernel.h>
#include <Mandelbrot_gold.h>
Generally, how difficult do you think it is it to interface such
includes with FPK or Fortran?
Surely they can be wrapped.
Users reported a 80x improvement with Cudo on a Gforce compared to a
Opteron 2GHZ!
The test was single precission though.
Lv wrote:
That is a great reference for me to dig into, thanks.
Lord Satan wrote:
Hope this helps:
http://www.gpgpu.org/
http://developer.nvidia.com/object/cuda.html
_________________________________________________________________
To unsubscribe: mail [EMAIL PROTECTED] with
"unsubscribe" as the Subject
archives at http://www.lazarus.freepascal.org/mailarchives
_________________________________________________________________
To unsubscribe: mail [EMAIL PROTECTED] with
"unsubscribe" as the Subject
archives at http://www.lazarus.freepascal.org/mailarchives
_________________________________________________________________
To unsubscribe: mail [EMAIL PROTECTED] with
"unsubscribe" as the Subject
archives at http://www.lazarus.freepascal.org/mailarchives
