On Thu, 5 Mar 2009, Pierre Ossman <oss...@cendio.se> wrote: > On Thu, 05 Mar 2009 10:13:40 -0500 > > It's not clear to me how the choice would affect network bandwidth > > usage because the vnc client normally selects the bpp and depth, right? > > Silly me. You're of course correct.
BTW, I think your desire to change the default depth of Xvnc to be 24 is a smart choice. > So this furthers my cause. vncviewer picks the color depth of the local > screen (commonly 24), so it is just a lot of extra CPU overhead for us > to have anything else on the server. Yes, it likely adds to the amount of local work done by Xvnc. The only potential performance downside in going from 16bpp to 32bpp for Xvnc that I can think of is the 2X slower local memory I/O of Xvnc to and from its virtual framebuffer in RAM. This is likely quite small compared to the transformations that need to be done in the CPU (but I am just guessing that since I don't haven't done any benchmarks, or if I did, I have long forgotten the results.) On a real h/w framebuffer (not Xvnc) this can still be an important effect. I wonder why the Xvnc default was set to 16bpp? Do you know if Realvnc does this or is it a Tightvnc change? Karl ------------------------------------------------------------------------------ Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise -Strategies to boost innovation and cut costs with open source participation -Receive a $600 discount off the registration fee with the source code: SFAD http://p.sf.net/sfu/XcvMzF8H _______________________________________________ Tigervnc-devel mailing list Tigervnc-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/tigervnc-devel
