On-board GPU for display?

Just wondering: Is there a way to use my on-board GPU for displaying while the PCI-Express GPU computes?

If I change the primary display driver to onboard, NVIDIA X-Server still lists the PCI-Express Card, but nvcc tells me about missing files. Path is set as usual

Yes, configure X to use the on board, use the script in the release notes to load the nvidia driver and create the /dev entries.


I have an Intel on-board graphic card and a PCI-Express Nvidia GT 520 operated on Fedora 13, Dell Vostro 220s. When using the Nvidia card for both display and Cuada, it works fine. However I’ve failed in trying to totally use Nvidia card for CUDA, leaving display mission for the on-board Intel card.
To do that, I plugged the cable to the Intel card, tried to set the Init Dislay to both Onboard or PCI-Express. I also set the initial display to non-X mode: init 3.
After booting to non-X mode, I used the script provided by Cuda:


/sbin/modprobe nvidia
if [ “$?” -eq 0 ]; then

Count the number of NVIDIA controllers found.

NVDEVS=lspci | grep -i NVIDIA
N3D=echo "$NVDEVS" | grep "3D controller" | wc -l
NVGA=echo "$NVDEVS" | grep "VGA compatible controller" | wc -l

N=expr $N3D + $NVGA - 1
for i in seq 0 $N; do
mknod -m 666 /dev/nvidia$i c 195 $i
mknod -m 666 /dev/nvidiactl c 195 255
exit 1

to load the Nvidia driver.
At this time the modprobe give an error message:
“FATAL: Error inserting nvidia (/lib/modules/…/kernel/drivers/video/nvidia.ko): No such device”

Could anyone show me how to handle this problem.
Thank you very much in advance.

Have the same Problem with a DQ57TM Desktop Board and an i5-650. The problem seems to be that either the IGE or the nVidia card is enabled. The script you quoted cannot work, as lspci does not find any nVidia device if the IGP is used. The other way around, it seems if the nVidia card is enabled, the IGP is totally disabled.

I posted at Intel Support Community Forum, maybe somebody there knows if it is possible to enable both.

I do believe that use the on-board graphics card for display should be a key requirement for scientific machines that are now using GPUs to perform scientific calculus. However, its been pretty much observable high amounts of machines using a GT series to display, which could actually being used for computing.

Can someone here explain me that purpose? I mean, why there are so much people using GT series to display instead of buy a motherboard with a GPU on it?