s-news
[Top] [All Lists]

S-Plus timings

To: S-News <s-news@lists.biostat.wustl.edu>
Subject: S-Plus timings
From: Thom Burnett <Thom.Burnett@cognigencorp.com>
Date: Mon, 20 Mar 2006 13:49:44 -0500
User-agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.6) Gecko/20040113
This the the data that I got from people who sent me their timing results.
It mostly shows that the graphs get faster as the CPU speed and memory increase - very reasonable.
It also reassures me that my Windows times are normal.
I'm wondering if my Unix times are longer than usual.

Can anyone try the code on a Unix machine and let me know what they get?

Command         Platform        OS      CPU Speed (GHz)         Memory (G)      
Elapsed Time
source  PC      XP pro  4.0     2.80    0.45
source  PC      XP pro  3.4     2.00    0.642
source  PC      XP pro  1.3     0.50    0.9
source  PC      XP      2.1     1.00    1.05
source  PC      XP pro  2.6     0.50    1.3
source  PC      2000 pro        1.7     0.50    1.4
source  PC      
        1.5     3.00    2.52
line    PC      XP pro  2.6     2.00    2.56
source  PC      
        1.6     0.50    2.63
batch   PC      Win2k   2.0     0.75    10.4



The code we're all running is:

For Unix:

remove(objects(), where=1)
print(objects())

now <- proc.time()

plotToFile <- T

x <- seq(from=0, to=5, by=.001)
y <- cos(x)
outputDir <- "/myPath/myDir/"
outputFilename <- paste(outputDir, 'simpleSin1.png', sep='')
resolutionWidth <- 1000
formatString <- 'PNG'
if(plotToFile) java.graph(file=paste(outputFilename, sep=""), format=formatString, width=resolutionWidth)
  plot(x,y)

  z <- sin(x)
  points(x,z, type='l')
if(plotToFile)dev.off()


print(proc.time() - now)


<Prev in Thread] Current Thread [Next in Thread>
  • S-Plus timings, Thom Burnett <=