You have to run glxspheres with vglrun. Otherwise, it will use the software OpenGL renderer in TigerVNC, and that will be the bottleneck, not TigerVNC's image pipeline.
Also realize that the methodology is not just running GLXspheres and reading the output of that benchmark. You have to run vglrun -sp glxspheres -fs to ensure that frame spoiling is disabled in VirtualGL, so that VGL is drawing only the frames that get rendered. You then have to measure the frame rate from the client using TCBench, to ensure that you are reading only the frames that get delivered to the client. Using TCBench may not make any difference on a fast network, because generally TigerVNC won't drop frames in that scenario, but it definitely can drop frames in a low-bandwidth environment, so the frame rate reported by GLXspheres wouldn't be accurate in that case. Also, if you want to reproduce the exact scenario under which I am running the benchmarks, set your TigerVNC geometry to 1240x900 and use a 1280x1024 resolution on the client display. DRC On 2/12/11 1:13 PM, Martin Koegler wrote: > On Fri, Feb 11, 2011 at 10:53:48AM -0600, DRC wrote: >> On 2/11/11 2:17 AM, Martin Koegler wrote: >>>> What do you mean "still get 20 Mbit"? You mean Megapixels/second? Or >>>> do you mean Megabits/second of throughput? Because without VeNCrypt, my >>> >>> I mean the value showen in the viewer info dialog. What tools are you using? >> >> A thorough description of my methodology is here: >> >> http://www.virtualgl.org/pmwiki/uploads/About/vglperf21.pdf > > I'm trying to setup something to reproduce your measurements. > > So I have a 64 Bit Debian Linux box [a cheap AMD Athlon(tm) II X2 250 > with enough memory] with my Tigervnc shapshot packages > (http://e9925248.users.sourceforge.net/tigervnc/). > > The snapshot is some month old, but there have not been any changes > affecting TLS encryption and it is available to the public for > reproducing. As your statement is, that TLS is slower than non-TLS, I > should see the 2x slowdown. > > I also downloaded VirtualGL-2.2.tar.gz and built glxspheres. > > I started my Xvnc Server with TLSNone and None enabled - the only > other option was a display number. > > Running it without any viewer connected: > ./glxspheres -fs > Polygons in scene: 62464 > Visual ID of window: 0xc9 > OpenGL Renderer: Software Rasterizer > 11.388422 frames/sec - 8.956220 Mpixels/sec > 11.515387 frames/sec - 9.056069 Mpixels/sec > 11.411444 frames/sec - 8.974325 Mpixels/sec > 11.604320 frames/sec - 9.126008 Mpixels/sec > 11.401914 frames/sec - 8.966830 Mpixels/sec > 11.632486 frames/sec - 9.148159 Mpixels/sec > 11.597135 frames/sec - 9.120358 Mpixels/sec > 11.477439 frames/sec - 9.026226 Mpixels/sec > 11.577937 frames/sec - 9.105260 Mpixels/sec > 11.305388 frames/sec - 8.890919 Mpixels/sec > > Running it with viewer with security type None: > ./glxspheres -fs > Polygons in scene: 62464 > Visual ID of window: 0xc9 > OpenGL Renderer: Software Rasterizer > 9.525048 frames/sec - 7.490803 Mpixels/sec > 9.588003 frames/sec - 7.540312 Mpixels/sec > 9.698757 frames/sec - 7.627413 Mpixels/sec > 9.559365 frames/sec - 7.517791 Mpixels/sec > 9.640171 frames/sec - 7.581339 Mpixels/sec > 9.434236 frames/sec - 7.419385 Mpixels/sec > 9.605030 frames/sec - 7.553703 Mpixels/sec > 9.790638 frames/sec - 7.699671 Mpixels/sec > 9.605973 frames/sec - 7.554445 Mpixels/sec > 9.637021 frames/sec - 7.578862 Mpixels/sec > 9.738392 frames/sec - 7.658583 Mpixels/sec > 9.651538 frames/sec - 7.590278 Mpixels/sec > > Running it with viewer with security type TLSNone: > ./glxspheres -fs > Polygons in scene: 62464 > Visual ID of window: 0xc9 > OpenGL Renderer: Software Rasterizer > 8.142692 frames/sec - 6.403674 Mpixels/sec > 7.506465 frames/sec - 5.903324 Mpixels/sec > 7.979971 frames/sec - 6.275705 Mpixels/sec > 8.150443 frames/sec - 6.409770 Mpixels/sec > 8.064119 frames/sec - 6.341881 Mpixels/sec > 8.110632 frames/sec - 6.378460 Mpixels/sec > 7.685281 frames/sec - 6.043951 Mpixels/sec > 8.366945 frames/sec - 6.580033 Mpixels/sec > 8.602331 frames/sec - 6.765148 Mpixels/sec > 7.869705 frames/sec - 6.188988 Mpixels/sec > 7.773481 frames/sec - 6.113314 Mpixels/sec > 8.114384 frames/sec - 6.381411 Mpixels/sec > 7.788822 frames/sec - 6.125379 Mpixels/sec > 7.489821 frames/sec - 5.890235 Mpixels/sec > > The numbers don't show such a fundamental slowdown (6.0/7.8=0.78 - so > only about 22 %). > > If I can't reproduce the big slowdown, I can't analyse it. > > Regards, > Martin Kögler ------------------------------------------------------------------------------ The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE: Pinpoint memory and threading errors before they happen. Find and fix more than 250 security defects in the development cycle. Locate bottlenecks in serial and parallel code that limit performance. http://p.sf.net/sfu/intel-dev2devfeb _______________________________________________ Tigervnc-devel mailing list Tigervnc-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/tigervnc-devel
