Hi I m trying to use GCRMA 1.0.1 under R1.8.1 with Affy 1.3.26 under win XP Pro with 18 U133A Affychips and I obtain this message: background correcting...Error: cannot allocate vector of size 96 Kb which means that I dont have enough RAM How can I increase the memory size... knowing that I already have 3.62 Go RAM !!! rhdf5 is not anymore available as a Win package...just as a source package: could it be of some help here ? If so can somebody send me the compiled version for Win ? Is there a GCRMexpress version such as the RMAexpress that does not need R ? Should I move to a 1.4.x version of Affy ? thanks for your precious help as usual Philippe

1.) Put --max-mem-size=2500M in the target field for the shortcut you use to start R (after the ...RGui.exe part). Theoretically a 32 bit machine can use 4 GB RAM, but Windoze takes a bunch, so the true limit is something less than 4 GB. I think the patched version uses a better memory allocation scheme than 1.8.0, and for sure 1.9.0 pre-release does, so you might want to compile that one up and try it. 2.) To use rhdf5, you have to get the hdf5.dll first and then compile the package. There was a conversation about this on R-help yesterday that you might want to look at. http://finzi.psych.upenn.edu/R/Rhelp02/archive/0816.html HTH, Jim James W. MacDonald Affymetrix and cDNA Microarray Core University of Michigan Cancer Center 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623 >>> <phguardiol@aol.com> 12/02/03 11:07AM >>> Hi I m trying to use GCRMA 1.0.1 under R1.8.1 with Affy 1.3.26 under win XP Pro with 18 U133A Affychips and I obtain this message: background correcting...Error: cannot allocate vector of size 96 Kb which means that I dont have enough RAM How can I increase the memory size... knowing that I already have 3.62 Go RAM !!! rhdf5 is not anymore available as a Win package...just as a source package: could it be of some help here ? If so can somebody send me the compiled version for Win ? Is there a GCRMexpress version such as the RMAexpress that does not need R ? Should I move to a 1.4.x version of Affy ? thanks for your precious help as usual Philippe _______________________________________________ Bioconductor mailing list Bioconductor@stat.math.ethz.ch https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor