Hi Gerard, did you executed initramfs, so that the firmware is copied to /boot? I am using kernel 3.11 without any glitches
regards Am 14.09.2013 16:25, schrieb Gerald Nunn: > Roberto, > > You can run the benchmark by doing the following: > > a. sudo apt-get install glmark2 > b. Then open a command prompt and run glmark2 > > It seems pretty stable so far (other then the artifacts on the GDM login > screen) > but I am having one issue where UVD (for hardware acceleration of video) is > not > initializing and I'm not sure why. Initially I had just updated the TURKS > radeon > firmware but it was complaing about needing SUMO for the uvd so I copied that > as > well and re-installed the kernel but no luck in getting UVD to work. > > If you want to try using the latest stuff yourself here I used the basic > instructions > here > http://askubuntu.com/questions/324733/how-to-enable-the-radeon-dynamic-power-management-feature-in-ubuntu-13-04. > Below are my steps in short form > > 1. Boot the computer in Ubuntu with the PMD NOT attached > 2. Copy the latest radeon firmware files into /lib/firmware/radeon, you need > the > TURKS and SUMO firmware (SUMO for UVD). Backup your old firmware first > 3. Install the latest kernel, I used the instructions > here > http://ubuntuhandbook.org/index.php/2013/09/linux-kernel-3-11-released-install-upgrade-in-ubuntu-13-0412-04/ > 4. Update grub to include the radeon.dpm=1 parameter > 5. Reboot, if you want the latest mesa 9.2 which helps performance a bit > install > the xorg-edgers ppa > 6. Make config changes necessary to detect PMD attached and use radeon driver > as > per here > http://forum.notebookreview.com/sony/702034-sony-vaio-z-2012-review-linux-perspective.html > > And that's it, but like I said I'm not swearing this setup is stable at this > point as I just started playing with it yesterday. Also, I tried running the > glmark2 against the Intel card and it gave me a segfault on one of the later > tests. > > I will also say while this setup feels somewhat faster then previously plus > you > get the benefit of power management, the radeon card in the PMD is still > slower > then the built in intel graphics. I used to think it was just an open source > driver issue but at this point I'm not too sure, I'm wondering if it has > something do with the way Linux manages PCI as the PMD basically extends the > PCI > system through light peak/thunderbolt. I just don't know enough to determine > this though. > > Gerald > > > > On Sat, Sep 14, 2013 at 7:30 AM, Roberto Gardenier <[email protected] > <mailto:[email protected]>> wrote: > > Hi Gerald, > > Thanks for sharing this. I currently havent touched my device in a few > months so its probably still at 3.8 in 13.04. If you could send me the > instructions on how to setup this benchmark then i'd be happy to send the > results back to you. > > Also if you could share the instructions on how to get to this stable > situation then i'd be very happy. I still was encountering problems with > the > pmd. > > Kind regards, > Roberto Gardenier > > Op 14 sep. 2013 00:43 schreef "Gerald Nunn" <[email protected] > <mailto:[email protected]>> het volgende: > > Guys, > > So I built the bullet and upgraded my Ubuntu Gnome 13.04 environment > to > use the 3.11 kernel and Mesa 9.2 to take advantage of dynamic power > management and hopefully improve the performance of games. Upgrading > went smoothly and the only graphical issue so far in light testing is > that GDM is a bit wonky looking when I go to login but still useable. > Once I am logged in the Gnome desktop is just fine. > > I'm happy to report that DPM works great and my PMD fan is not working > hard any longer, the GPU temp is down to 48 degrees versus the 59 it > was > previously. Additionally when starting a game or running a GPU > benchmark > the fan ramps up in response the GPU ramping up. Keep in mind I only > just started testing, I'll post an update after the weekend. > > Anyways, I ran glmark2 on and got the results below. I'm curious what > other folks are getting who are using the PMD so it would be great if > anyone else could report their results. I stupidly did not run the > benchmark before making the changes and now that I'm stable I don't > want > to undo everything just for that. > > ======================================================= > glmark2 2012.08 > ======================================================= > OpenGL Information > GL_VENDOR: X.Org > GL_RENDERER: Gallium 0.4 on AMD TURKS > GL_VERSION: 3.0 Mesa 9.2.0-devel > ======================================================= > [build] use-vbo=false: FPS: 348 FrameTime: 2.874 ms > [build] use-vbo=true: FPS: 1768 FrameTime: 0.566 ms > [texture] texture-filter=nearest: FPS: 1616 FrameTime: 0.619 ms > [texture] texture-filter=linear: FPS: 1615 FrameTime: 0.619 ms > [texture] texture-filter=mipmap: FPS: 1663 FrameTime: 0.601 ms > [shading] shading=gouraud: FPS: 1635 FrameTime: 0.612 ms > [shading] shading=blinn-phong-inf: FPS: 1635 FrameTime: 0.612 ms > [shading] shading=phong: FPS: 1606 FrameTime: 0.623 ms > [bump] bump-render=high-poly: FPS: 1366 FrameTime: 0.732 ms > [bump] bump-render=normals: FPS: 1752 FrameTime: 0.571 ms > [bump] bump-render=height: FPS: 1693 FrameTime: 0.591 ms > [effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 927 FrameTime: 1.079 ms > [effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 425 FrameTime: > 2.353 ms > [pulsar] light=false:quads=5:texture=false: FPS: 1478 FrameTime: > 0.677 ms > [desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: > FPS: 471 FrameTime: 2.123 ms > [desktop] effect=shadow:windows=4: FPS: 663 FrameTime: 1.508 ms > [buffer] > > columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: > FPS: 201 FrameTime: 4.975 ms > [buffer] > > columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: > FPS: 818 FrameTime: 1.222 ms > [buffer] > > columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: > FPS: 206 FrameTime: 4.854 ms > [ideas] speed=duration: FPS: 1082 FrameTime: 0.924 ms > [jellyfish] <default>: FPS: 588 FrameTime: 1.701 ms > [terrain] <default>: FPS: 132 FrameTime: 7.576 ms > [conditionals] fragment-steps=0:vertex-steps=0: FPS: 1503 FrameTime: > 0.665 ms > [conditionals] fragment-steps=5:vertex-steps=0: FPS: 1248 FrameTime: > 0.801 ms > [conditionals] fragment-steps=0:vertex-steps=5: FPS: 1504 FrameTime: > 0.665 ms > [function] fragment-complexity=low:fragment-steps=5: FPS: 1499 > FrameTime: 0.667 ms > [function] fragment-complexity=medium:fragment-steps=5: FPS: 1367 > FrameTime: 0.732 ms > [loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 1507 > FrameTime: 0.664 ms > [loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: > 1499 > FrameTime: 0.667 ms > [loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 548 > FrameTime: 1.825 ms > ======================================================= > glmark2 Score: 1145 > ======================================================= > > Thanks, > > Gerald > > > _______________________________________________ > Mailing list: https://launchpad.net/~sony-vaio-z-series > Post to : [email protected] > <mailto:[email protected]> > Unsubscribe : https://launchpad.net/~sony-vaio-z-series > More help : https://help.launchpad.net/ListHelp > > > > > _______________________________________________ > Mailing list: https://launchpad.net/~sony-vaio-z-series > Post to : [email protected] > Unsubscribe : https://launchpad.net/~sony-vaio-z-series > More help : https://help.launchpad.net/ListHelp > _______________________________________________ Mailing list: https://launchpad.net/~sony-vaio-z-series Post to : [email protected] Unsubscribe : https://launchpad.net/~sony-vaio-z-series More help : https://help.launchpad.net/ListHelp
