A "trick" I read in some library for calculating e1e1 is to calculate the series for x=1/2kx=1/2k where kk is the root of the amount of required bits, then you have to square the result k times to get your final number, it avoids that big/small numbers make the result too inacc...