ÈÐ Wïgê† Ðð† ñåmê

I am all about performance and speed.  I even go so far as to test various ways of doing things in bash (or other programming languages) to optimize the code so it runs faster or more efficiently.  I optimize services in the same matter.  With my fairly new laptop, it has one of the hybrid graphics cards – intel + nvidia.  It works with bumblebee….and cuda….for the most part.  Simply running ‘optirun command’ switches to the nvidia graphics card which gives much better performance, not only in graphics intensive stuff…..but even bash shell.

Example:  running a simple ‘time ls -la’ in a directory containing  1590 items

[codesyntax lang=”bash”]

real	0m0.095s
user	0m0.012s
sys	0m0.008s

[/codesyntax]

Example 2: running ‘optirun xterm’ and then running the same command ‘time ls -la’ in the same directory in the new nvidia xterm:

[codesyntax lang=”bash”]

real    0m0.015s
user    0m0.000s
sys     0m0.012s

[/codesyntax]

And even the glx tests are much faster:

[codesyntax lang=”bash”]

$ glxspheres 
Polygons in scene: 62464
Visual ID of window: 0xa4
Context is Direct
OpenGL Renderer: Mesa DRI Intel(R) Ivybridge Mobile 
60.131595 frames/sec - 67.106860 Mpixels/sec
31.846655 frames/sec - 35.540867 Mpixels/sec
31.377641 frames/sec - 35.017447 Mpixels/sec
31.636594 frames/sec - 35.306439 Mpixels/sec

[/codesyntax]

and:

[codesyntax lang=”bash”]

$ optirun glxspheres 
Polygons in scene: 62464
Visual ID of window: 0x21
Context is Direct
OpenGL Renderer: GeForce GTX 660M/PCIe/SSE2
153.415397 frames/sec - 171.211583 Mpixels/sec
158.583456 frames/sec - 176.979136 Mpixels/sec
161.016123 frames/sec - 179.693993 Mpixels/sec
158.156824 frames/sec - 176.503015 Mpixels/sec

[/codesyntax]

To me, this is enough of a performance improvement to use the nvidia graphics card always.  Except, there is no way to do that in the new laptop…..at least that I have figured out.