Take the 2-minute tour ×
Ask Ubuntu is a question and answer site for Ubuntu users and developers. It's 100% free, no registration required.

This error occurs when I try to use the optirun command:

$ optirun glxgears
[39785.949467] [ERROR]Cannot access secondary GPU - error: [XORG] (EE) NVIDIA(0): Failed to assign any connected display devices to X screen 0

[39785.949560] [ERROR]Aborting because fallback start is disabled.

We have the following error when we change nvidia-current to nouveau for driver of bumblebee.conf.

[  167.510941] [ERROR]Cannot access secondary GPU - error: Could not load GPU driver
[  167.511064] [ERROR]Aborting because fallback start is disabled.

What should I do? I am trying to get my GT650M and GTX680M with NVIDIA Optimus to install successfully. Currently optirun is not working.

My System is w110er, GeForce650M and Ubuntu 12.10.

The first GPU is intel HD Graphics 4000:

glxgears with inel HD Graphics 4000.
302 frames in 5.0 seconds = 60.259 FPS
glxgears with GTX 485M
75120 frams in 5.0 seconds = 15023.970 FPS

Optirun is working now. Is this good performance?

not use GT650m 301 frames in 5.0 seconds = 60.009 FPS

GT650m with optimus 4677 frames in 5.0 seconds = 935.319 FPS

GTX680m with Optimus 10750 frames in 5.0 seconds =2149.977 FPS

I think it is not great performance. I would like to know how to install great performance such as GTX485M.

Performance of 'nbody' cuda demo for GTX 485m is 250 GFLOPS.

Performance of 'nbody' cuda demo for GT 650m is 107 GFLOPS.

I am sad for low performance of GT650m.

I am would like to know the differnce between nouveau, nvidia-current, nvidia-experimental and 310.19. Only GPU using nvidia-current driver is working. Can I use nvidia-experimental or 310.19 driver for GT650m with NVIDIA Optimus? I ask you to help me. I would like to get advices.

share|improve this question
    
A couple of thoughts on this. Firstly, comparing the GTX 485 to the GT 650 is comparing top-of-the-range Jan 2011 to mid-range March 2012. The 485GTX has higher theoretical performance in memory bandwidth, GFLOPS, etc. (see en.wikipedia.org/wiki/… ). Additionally, the Optimus parts are working in concert with the Intel graphics. All pixels that hit the screen get written to memory and then read back out by the Intel device. That and just the general overheads of communication are bottlenecks. –  ahcox Feb 13 '13 at 20:14

1 Answer 1

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.