Prism cannot compare survival curves when the sample size is huge.

Prism, up to Prism 6, cannot compare survival curves when the number of subjects at risk at any time point is greater than 46340. There is a math overflow. The survival curves are computed just fine with large data sets, but the comparison of curves with the logrank test cannot proceed. Prism simply leaves that table blank, without any error message. The limit will be lifted in Prism 7.