Bug with x11(), Lion and CRAN R 2.13.1

I had a typo in my original post, fixed below.

This is not peculiar to Lion: it does seem something about Apple's X11 notifier and I've seen it quite often over the years on Leopard ...

On Tue, 26 Jul 2011, Konis Kjell wrote:

Hello,

I have just installed R 2.13.1 on a clean install of Lion on a 3 year old iMac with a 2GHz Core 2 Duo. When I run R in the terminal and use the x11 device I get some strange behaviors when I call the dev.off() function. If I plot several times then usually I get the error:

  Error in dev.off() : parameter "i" in "mfg" is out of range

and if I plot only once I get the warning:

  Warning message:
  In dev.off() : Display list redraw incomplete

In both cases, a new x11 window opens (not exactly the expected result of dev.off). Steps to reproduce both cases are given below.

Also, it seems that the x11 device is not 3-4 times faster than the quartz device. Has anyone else noticed this?
Why would you expect it to be?  X11(type='Xlib') is, but that is no longer the default on OS X.
Oops!  That was supposed to say "the x11 device is now 3-4 times faster than the quartz device."  Anyway, here are the timings I get (I expected quartz to be the fastest but not for any particular reason):
x11(type = "Xlib")
system.time(demo("graphics", echo = FALSE))
user  system elapsed 
  0.191   0.025   0.496
x11(type = "cairo")
system.time(demo("graphics", echo = FALSE))
user  system elapsed 
 19.223   9.757  42.559
system.time(demo("graphics", echo = FALSE))
user  system elapsed 
  0.716   1.290   5.626
quartz()
system.time(demo("graphics", echo = FALSE))
user  system elapsed 
  0.778   0.045   0.861
Cheers,
Kjell

## (re)start R ##

x11()
system.time(demo("graphics", echo = FALSE))
 user  system elapsed
0.213   0.025   0.517
dev.off()
Error in dev.off() : parameter "i" in "mfg" is out of range
dev.cur()
X11
3
dev.off()
null device
        1
dev.cur()
null device
        1

sessionInfo()
R version 2.13.1 (2011-07-08)
Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)

locale:
[1] en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

loaded via a namespace (and not attached):
[1] tools_2.13.1

## (re)start R ##

x11()
plot(1:10)
dev.off()
X11
4
Warning message:
In dev.off() : Display list redraw incomplete
sessionInfo()
R version 2.13.1 (2011-07-08)
Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)

locale:
[1] en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

_______________________________________________
R-SIG-Mac mailing list
R-SIG-Mac at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mac

-- 
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

Bug with x11(), Lion and CRAN R 2.13.1

Thread (3 messages)