Richard Yeh
2008-Mar-31 05:52 UTC
[R] WinXP exhibits sluggish graphics window movement & mouse tracking (repainting?) over windows graphics devices
I just noticed when using 2.6.1, 2.6.2 (2008-02-08), 2.6.2pat (2008-02-21 r44582), and 2.6.2pat (2008-03-24 r44975) on my poor old Celeron D330 (2.6 GHz; 3.5 years old) running Windows XP, that the mouse cursor appears to be redrawn more sluggishly when the pointer is over R windows graphics devices than over the R console window or other applications' windows. The slowdown seems to start only after I plotting something in the window (running "windows()" to open the device does not cause any slowdown), but the behavior starts after a "plot(rnorm(100))". The slowdown seems to depend on the area of the R graphics window that is visible. The slowdown also occurs when moving the plot window. For example, if the graphics window is frontmost, and I want to move it until it is mostly offscreen, then the movement is jerky. However, once the window is mostly offscreen, dragging it back onto the screen is smooth and fast. (When I drag windows, I only see the frame, not the contents.) This seems to affect SDI and MDI modes. The Windows task manager confirms that when the plot window is frontmost, Rgui.exe takes most of the CPU time. My graphics card is based on an ATI Radeon 7000, with 32 MB of RAM. I am surprised that I never noticed this before. Nobody else seems to have reported it to the r-help list. Is this problem restricted to my system? (My work machine is much newer, and I do not notice the problem there.) -- 607-351-4838 / Richard.C.Yeh at gmail.com
Prof Brian Ripley
2008-Mar-31 08:03 UTC
[R] WinXP exhibits sluggish graphics window movement & mouse tracking (repainting?) over windows graphics devices
I don't see this on a much older and slower machine, so suspect a problem with your Windows. It looks like something is set up to ask R to repaint after the mouse pointer, whereas Windows ought to be doing that. Even then, R uses double buffering, so the repaint should be fast (provided graphics acceleration is turned on). If you want to try updated versions of R, we suggest you use 2.7.0 alpha and not R-patched. But R updates will not solve Windows problems. On Mon, 31 Mar 2008, Richard Yeh wrote:> I just noticed when using 2.6.1, 2.6.2 (2008-02-08), 2.6.2pat > (2008-02-21 r44582), and 2.6.2pat (2008-03-24 r44975) on my poor old > Celeron D330 (2.6 GHz; 3.5 years old) running Windows XP, that the > mouse cursor appears to be redrawn more sluggishly when the pointer is > over R windows graphics devices than over the R console window or > other applications' windows. The slowdown seems to start only after I > plotting something in the window (running "windows()" to open the > device does not cause any slowdown), but the behavior starts after a > "plot(rnorm(100))". The slowdown seems to depend on the area of the R > graphics window that is visible. The slowdown also occurs when moving > the plot window. For example, if the graphics window is frontmost, > and I want to move it until it is mostly offscreen, then the movement > is jerky. However, once the window is mostly offscreen, dragging it > back onto the screen is smooth and fast. (When I drag windows, I only > see the frame, not the contents.) This seems to affect SDI and MDI > modes. The Windows task manager confirms that when the plot window is > frontmost, Rgui.exe takes most of the CPU time. > > My graphics card is based on an ATI Radeon 7000, with 32 MB of RAM. > > I am surprised that I never noticed this before. Nobody else seems to > have reported it to the r-help list. Is this problem restricted to my > system? (My work machine is much newer, and I do not notice the > problem there.) > > -- > 607-351-4838 / Richard.C.Yeh at gmail.com > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >-- Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595