R devel - May 2005 - Windows freezes plotting large line plots (PR#7856)

Full_Name: Timo Becker
Version: 2.1.0
OS: Windows XP
Submission from: (NULL) (193.170.87.36)


Hi!

The following command causes Windows XP to freeze so that I can only pull the
plug:
> plot(seq(1, 44100), rnorm(44100), type="l")
This does not happen with
> plot(seq(1, 44100), rnorm(44100))
and
> plot(seq(1,100), rnorm(100), type="l")
platform i386-pc-mingw32
arch     i386           
os       mingw32        
system   i386, mingw32  
status                  
major    2              
minor    1.0            
year     2005           
month    04             
day      18             
language R

This is already addressed in R-patched: we do ask you to check before 
filing a bug report so as not to waste our time.  (It was also discussed 
on R-help: see the archives.)

It is a bug *in Windows* and not one in R: nothing an application can do 
should be that fatal (and this is a perfectly legitimate Windows GDI 
call).  I trust that you will file a bug report in the correct place.
>From the CHANGES file:
   The support for mitred lines encountered a Windows problem with a
   dramatic slowdown if more than 1000 or so segments were plotted in a
   polyline.  We now plot in bunches of 1000, which is less accurate but
   works around the Windows problem.

On Thu, 12 May 2005 timo.becker@oeaw.ac.at wrote:
> Full_Name: Timo Becker
> Version: 2.1.0
> OS: Windows XP
> Submission from: (NULL) (193.170.87.36)
>
>
> Hi!
>
> The following command causes Windows XP to freeze so that I can only pull
the
> plug:
>> plot(seq(1, 44100), rnorm(44100), type="l")
>
> This does not happen with
>> plot(seq(1, 44100), rnorm(44100))
> and
>> plot(seq(1,100), rnorm(100), type="l")
>
> platform i386-pc-mingw32
> arch     i386
> os       mingw32
> system   i386, mingw32
> status
> major    2
> minor    1.0
> year     2005
> month    04
> day      18
> language R
>
> ______________________________________________
> R-devel@stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>
>
-- 
Brian D. Ripley,                  ripley@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

timo.becker@oeaw.ac.at writes:
> The following command causes Windows XP to freeze so that I can only pull
the
> plug:
> > plot(seq(1, 44100), rnorm(44100), type="l")
> 
> This does not happen with
> > plot(seq(1, 44100), rnorm(44100))
> and
> > plot(seq(1,100), rnorm(100), type="l")
This was fixed (worked around -- the bug is in Windows) once before
during "belated beta testing" of 2.1.0. Try 2.1.0 patched.

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907

On 12-May-05 ripley@stats.ox.ac.uk wrote:
> This is already addressed in R-patched: we do ask you to check
> before filing a bug report so as not to waste our time.
>  (It was also discussed on R-help: see the archives.)
> 
> It is a bug *in Windows* and not one in R: nothing an application
> can do should be that fatal (and this is a perfectly legitimate
> Windows GDI call).  I trust that you will file a bug report in the
> correct place.
> 
>>From the CHANGES file:
> 
>    The support for mitred lines encountered a Windows problem
>    with a dramatic slowdown if more than 1000 or so segments
>    were plotted in a polyline.  We now plot in bunches of 1000,
>    which is less accurate but  works around the Windows problem.
Clearly a neccessary change, even though it is less accurate.
But does this change apply only to Windows compilations (i.e.
conditional compilation), or will we all have to put up with it?

Ted.


--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding@nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 12-May-05                                       Time: 08:26:43
------------------------------ XFMail ------------------------------

On Thu, 12 May 2005 Ted.Harding@nessie.mcc.ac.uk wrote:
> On 12-May-05 ripley@stats.ox.ac.uk wrote:
>> This is already addressed in R-patched: we do ask you to check
>> before filing a bug report so as not to waste our time.
>>  (It was also discussed on R-help: see the archives.)
>>
>> It is a bug *in Windows* and not one in R: nothing an application
>> can do should be that fatal (and this is a perfectly legitimate
>> Windows GDI call).  I trust that you will file a bug report in the
>> correct place.
>>
>>> From the CHANGES file:
>>
>>    The support for mitred lines encountered a Windows problem
>>    with a dramatic slowdown if more than 1000 or so segments
>>    were plotted in a polyline.  We now plot in bunches of 1000,
>>    which is less accurate but  works around the Windows problem.
>
> Clearly a neccessary change, even though it is less accurate.
> But does this change apply only to Windows compilations (i.e.
> conditional compilation), or will we all have to put up with it?
It applies only to the windows() device (including win.metafile etc).

-- 
Brian D. Ripley,                  ripley@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