>> Please try it again with samples/ch30h_ace.out and send me the new log >> file. > > Attached. (gziped) > > BTW, the fact that it takes some time to happen, might suggest that the > problem is caused by the exponential increase of some random error > caused by machine accuracy... is that possible? Hmmm ... Seems plausible. Although I thought the behavior of floating point operations was quite standardized these days. I believe that the in the early days the JVM was more specific about floating point behavior. Later, they relaxed the definition somewhat to give more freedom on different platforms.
Q: How long does it take to happen? It is possible that I'm not trying it long enough on my machine ... Miguel ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Jmol-developers mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jmol-developers
