login

Year-end appeal: Please make a donation to the OEIS Foundation to support ongoing development and maintenance of the OEIS. We are now in our 61st year, we have over 378,000 sequences, and we’ve reached 11,000 citations (which often say “discovered thanks to the OEIS”).

A336778
a(n) is the least number of repetitions such that the result of the repeated execution of the multiplication f <- f*n started at f=1 differs from the exact value n^a(n), when the multiplication is performed using 64-bit double precision floats according to the IEEE 754 standard.
8
1024, 34, 512, 23, 34, 19, 342, 17, 23, 16, 34, 15, 19, 14, 256, 13, 17, 13, 23, 13, 16, 12, 34, 12, 15, 12, 19, 11, 14, 11, 205, 11, 13, 11, 17, 11, 13, 11, 23, 10, 13, 10, 16, 10, 12, 10, 34, 10, 12, 10, 15, 10, 12, 10, 19, 10, 11, 10, 14, 9, 11, 9, 171, 9, 11
OFFSET
2,1
COMMENTS
See A336774 for more information and links.
For n a power of two the first deviation is caused by exceeding the maximum representable number of the binary64 double precision floating point format with result +infinity in standard conforming implementations.
LINKS
Hugo Pfoertner, Comparison with exact results, range n=2..128, computed on x86-64 architecture.
CROSSREFS
KEYWORD
nonn,fini
AUTHOR
Hugo Pfoertner, Aug 04 2020
STATUS
approved