[seqfan] Re: Numbers NOT of the form 2^x-y^2

[Richard J. Mathar]

In response to http://list.seqfan.eu/pipermail/seqfan/2014-October/013760.html :

Numbers can be proven not to be of the form 2^x-y^2 if one can
find a modulus m such that one can show that all x must be even.
In these cases one can construct a finite upper limit to x,
so that range of x can be checked numerically. (That kind of easier
bundle of questions concerning 4^x rather than 2^x is targeted in A200524.)
I added an example to http://oeis.org/A051213 and more general comments in a PDF file
in http://oeis.org/A247763 . (Note that A247763 actually deals
with n that have the opposite sign.)