On Jul 12, 2013, at 6:42 AM, Ben Bolker wrote: > Jonsson <amen.alyaari <at> Bordeaux.inra.fr> writes: > >> >> Hello All, >> I am having a problem with this: >> >> file<-array(dim=c(1440,720,700,3)) >> Error in array(dim = c(1440, 720, 700, 3)) : >> 'dim' specifies too large an array >> >> I have a memory of 20GB, But I do not know where is the problem!Any help >> >> When I replaced 700 by any number bellow like( 600,500),it worked without >> any problem. > > From > > http://stat.ethz.ch/R-manual/R-devel/library/base/html/LongVectors.html : > > Prior to R 3.0.0, all vectors in R were restricted to at most 2^31 - 1 > >> 1440*720*700*3 > [1] 2177280000 >> 2^31-1 > [1] 2147483647 > > So you need R>=3.0.0 (on a 64-bit system).
And (unfortunately) perhaps three times the currently installed RAM will be needed for productive use of such an R data-object. The memory requirements of a numeric array are roughly 10 times the product of the dimensions: 10*prod( c(1440,720,700,3)) #[1] 21772800000 # G M K So the hardware limitations are now a constraint. I am a bit surprised that object with those lower dimensions could be handled "without any problem." Generally an object of dim =c(1440,720,700,3) will not be able to be productively used for anything except read access. Copying it or even assigning new values to it, which of necessity creates a copy, would generally overflow installed RAM and push your session into "virtual memory" at which point my system starts to display molasses-like behavior. Occassionally waiting on the order of 5-20 minutes allows the process to complete, but in many instances terminating the session is needed. -- David Winsemius Alameda, CA, USA ______________________________________________ R-help@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.
