I just recently bought a HP DV6T-7000 Quad Edition. It has an Intel HD 4000 and a Nvidia GeForce GT 650M 2GB with Optimus. I read that I could use bumblebee to make optimus work, so I installed it. I also installed bumblebee-nvidia and nvidia-current from the ubuntu-x-swat/x-updates ppa. I rebooted, but when I tried to run anything with optirun, the computer would wait ten seconds or so, then do a hard shutdown. I got no log messages from bumblebee, Xorg, or optirun, either. I have purged and reinstalled bumblebee, bumblebee-nvidia, and nvidia-current. I have also turned off power management for nvidia in the bumblebee.conf file to no avail. I am out of ideas about this, and I need both graphics options. Any ideas would be much appreciated.
UPDATE: Today (6/14/12), I updated my kernel (to version 3.2.0-25) and reinstalled bumblebee-nvidia and nvidia-current (from ppa:upubuntu-com/nvidia with driver version 295.59) so the new driver would build against the new kernel. I noticed several things:
- Optirun worked (with various errors).
- I had to modprobe nvidia-current and modprobe -r nvidia-current before I got optirun to stop telling me that the card fell off the bus.
- Once I did that, I was able to run non-OpenGL applications with optirun; however, running OpenGL applications would give me the error Xlib: extension "GLX" missing on display ":8".
- Looking this up quickly yielded the following page: https://github.com/Bumblebee-Project/Bumblebee/wiki/Troubleshooting which listed my issue. Following the instructions in the section resulted in my having the same issue (shutdowns) that I posted about originally.
I also found this article, where someone said in the comments that he experienced the same issue if he did not run a cuda program before running optirun. (comment number 9). I will try this and see if it works, if it does maybe I can add a quick cuda program to my init scripts?
Thanks to everyone who already replied, I really appreciate your help.
Today (6/23/12), I switched to the xorg-edgers ppa to see if the newer drivers/xorg/etc would fix my problem. Now I'm getting the same errors listed above in numbers 1, 2, and 3. I have not tried step 4 yet because I do not want to wreck something up like I did last time (at least now non-gl applications run).
Also, CUDA is getting me nowhere due to make errors, so for now, that's out of the question.