Skip to content

OS 10.5.8 Compatibility

9 messages · David Winsemius, John Maindonald, Francis Smart +1 more

Original
Francis Smart
# 
Dear R people,

I am running OS 10.5.8 on my PowerBook G4 and unfortunately it will
not allow me to upgrade to Snow Leopard, because it does not run an
Intel Chip.

I have downloaded a version of R and it does not appear to be stable.
I have thought about about installing Linux to ensure a stable build
but unfortunately I bought the PowerBook G4 used and I did not come
with install disks.  The Leopard install disks I do have refuse to run
on an non-Intel machine.  So if I installed Linux, that would mean
that I would have to uninstall OS X and I probably would not be able
to reinstall it.

So I guess I am wondering if anybody has a build of R that they know
is stable on 10.5.8.  Also, I am sure there are other potential
solutions that I have not thought of.

Any recommendations?

Thank you for your time,
Francis
David Winsemius
#
On Dec 5, 2009, at 3:05 PM, Francis Smart wrote:

            

      
      
      
    

        
... a version?

            

      
      
      
    

        
That is a bit vague.

            

      
      
      
    

        
I currently have R 2.8.0 on a G4 PB running 10.5.7. I don't remember  
that there was a conscious decision not to upgrade R, I also cannot  
remember if I had a reason not to go to OSX 10.5.8. Almost all of my R  
work is on later Macs and the PB now is just for checking PowerPoint  
presentations for compatibility with OpenOffice.

            

      
      
      
    

        
Not sure I know much else to say other than you can get old disk-image  
versions of the universal 32 bit build here:

http://lib.stat.cmu.edu/R/CRAN/bin/macosx/universal/base/

            

      
      
      
    

        
David Winsemius, MD
Heritage Laboratories
West Hartford, CT

  
  


  


      

    

    

      
      

        


  

        

      John Maindonald
    

    

      
      #
    

  


  

      
I am using R from ESS, with X11 as the default graphics device.
In due course, when I run a function that tries to open a new
graphics window, I get the message:

            

      
      
      
    

        
Error in X11(d$display, d$width, d$height, d$pointsize, d$gamma, d$colortype,  : 
 unable to start device X11cairo
In addition: Warning message:
In device.call(...) : cairo error 'out of memory'

This was happening also under R-2.9.1; decided I had better
report/inquire.  I then have the choice of opening a quartz
window, or restarting both R and X11.

-----------------------------------------------------------------------------------

            

      
      
      
    

        
R version 2.10.0 (2009-10-26) 
i386-apple-darwin9.8.0 

locale:
[1] C

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

other attached packages:
[1] DAAG_1.01           randomForest_4.5-33 rpart_3.1-45       
[4] MASS_7.3-4          lme4_0.999375-32    Matrix_0.999375-32 
[7] lattice_0.17-26     reshape_0.8.3       plyr_0.1.9         
[10] proto_0.3-8        

loaded via a namespace (and not attached):
[1] coda_0.13-4   ggplot2_0.8.3 tools_2.10.0 


John Maindonald             email: john.maindonald at anu.edu.au
phone : +61 2 (6125)3473    fax  : +61 2(6125)5549
Centre for Mathematics & Its Applications, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.
http://www.maths.anu.edu.au/~johnm

  
  


  


      

    

    

      
      

        


  

        

      Simon Urbanek
    

    

      
      #
    

  


  

      
Francis,

        
On Dec 5, 2009, at 3:05 PM, Francis Smart wrote:

        

            

      
      
      
    

        
FWIW that makes no difference as we don't build anything specially for Snow Leopard. And, yes, ppc is no longer supported by Apple since 10.6.

            

      
      
      
    

        
Can you tell us which version, please? I know that you already answered that you're happy with 2.9.2 but unless you tell us there may be more users sharing your frustration ... I don't have any ppc machines anymore so I cannot easily test that ...

            

      
      
      
    

        
R is supposed to be stable but there was one big change in the R.app GUI for R 2.10 which may be the culprit (but then there were some GUI mismatch reports in the Tiger build so possibly the Leopard build is fine...). The real question is whether you installed the Tiger or the Leopard build of R (both work on your machine). Once we know we can try to trace that back ...

Cheers,
Simon

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Simon Urbanek
    

    

      
      #
    

  


  

      
On Dec 5, 2009, at 4:55 PM, John Maindonald wrote:

        

            

      
      
      
    

        
You may want to report that to R-devel (possibly with more details such as a reproducible example - I don't think g10.6() is in any recommended package...) since the X11 device is not Mac-specific.

Thanks,
Simon

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      John Maindonald
    

    

      
      #
    

  


  

      
As this occurs only after some combination of computations and
use of graphics commands, maybe extending over several days,
it is not possible to provide a reproducible example.   Once the
session gets into this state, any attempt to open an X11 window
triggers the output that I obtained.  It seems unimportant whether
this is a direct call to x11(), or to a plotting function that requires
the opening of an X11 window.  

So all that I can usefully report on may be the state of my session 
when the behaviour was triggered.  I wll try running a large number 
of graphics commands in a new session, that open and close the 
X11 graphics device, to look for a sequence of commands that 
consistently triggers the behaviour.

John Maindonald             email: john.maindonald at anu.edu.au
phone : +61 2 (6125)3473    fax  : +61 2(6125)5549
Centre for Mathematics & Its Applications, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.
http://www.maths.anu.edu.au/~johnm

        
On 06/12/2009, at 11:02 AM, Simon Urbanek wrote:

        

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Francis Smart
    

    

      
      #
    

  


  

      
Hi Simon,

I am not sure if you saw the email that I responded to that David
sent.  In that response I used the link he sent me.  In that response
I used the versions listed on the link.

And described exactly what I did that resulted in it freezing up.

I hope this is clear.  If it is not, I am happy to provide whatever
additional information would be useful.  I think the r-team does
amazing work and I would be happy to help in whatever way I can.

In case you missed that email, I have copied it below.

The referenced R installs are from
http://lib.stat.cmu.edu/R/CRAN/bin/macosx/universal/base/ and the two
are R-2.10.0.dmg and R-2.9.2.dmg

Cheers!
Francis

Hi David,

I know I was being vague when I posted but I had gotten frustrated
with R and deleted it hours before and I was not sure what the version
was at the time.  But now I am.

Version R-2.10.0 worked on startup and worked when I typed >1:10
however it froze when I typed >f() without f being defined.  There
were other functions that caused it to freeze, but I did not keep
track of all of them.

So I tried R-2.10.0 and it froze again under identical circumstances.

I went to the next oldest version R-2.9.2 and it froze on the last
stage of the installation process.  I forced it to quit and went to
the application folder and so far it is stable.

If it has problems, I will just keep going back to older releases.

Thank you for your help,
Francis

On Sat, Dec 5, 2009 at 7:00 PM, Simon Urbanek

        
<simon.urbanek at r-project.org> wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Simon Urbanek
    

    

      
      #
    

  


  

      
On Dec 6, 2009, at 2:27 AM, Francis Smart wrote:

        

            

      
      
      
    

        
Those are for Tiger (10.4), please try the R-2.10.0.pkg which is for Leopard (10.5) and let me know if the problem persists.

Thanks,
Simon

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Francis Smart
    

    

      
      #
    

  


  

      
Simon,

The latest version seems to be working perfectly now that I have the
right install.

Thanks for that.  Sorry about the fuss.  I am new to Macs.

Francis

On Sun, Dec 6, 2009 at 8:47 AM, Simon Urbanek

        
<simon.urbanek at r-project.org> wrote: