2 messages · nicolas_simonis at dfci.harvard.edu, Duncan Murdoch
Full_Name: Nicolas Simonis Version: 2.6.1 OS: windows Submission from: (NULL) (155.52.10.223) Very funny bug :
0.6<0.6
[1] FALSE
0.6<0.4+0.2
[1] TRUE control :
0.7<0.7
[1] FALSE
0.7<0.4+0.3
[1] FALSE With R 2.6.1 windows. it's not doing it with R 2.3.1 linux...
Full_Name: Nicolas Simonis Version: 2.6.1 OS: windows Submission from: (NULL) (155.52.10.223) Very funny bug :
0.6<0.6
[1] FALSE
0.6<0.4+0.2
[1] TRUE
I don't see a bug here, I see failure to read FAQ 7.31. The value 0.6 is less than 0.4+0.2, when the operation is done in floating point approximations. None of those three numbers is exactly representable. > (0.4+0.2) - 0.6 [1] 1.110223e-16 Duncan Murdoch
control :
0.7<0.7
[1] FALSE
0.7<0.4+0.3
[1] FALSE With R 2.6.1 windows. it's not doing it with R 2.3.1 linux...
______________________________________________ R-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel