this is awesome. how'd you stumble onto it?
Initially, I thought it's too good to be true, later found references in Wikipedia and Mathworld, and finally started to believe in it.
Sure we do, for flexible values of "normal":
from decimal import getcontext, Decimal
getcontext().prec = 600
n = Decimal(960939379918958884971672962127852754715004339660129306651505519271702802395266424689642842174350718121267153782770623355993237280874144307891325963941337723487857735749823926629715517173716995165232890538221612403238855866184013235585136048828693337902491454229288667081096184496091705183454067827731551705405381627380967602565625016981482083418783163849115590225610003652351370343874461848378737238198224849863465033159410054974700593138339226497249461751545728366702369745461014655997933798537483143786841806593422227898388722980000748404719)
# ignore floors, just using integer arguments
f = lambda x,y: (y/17 * Decimal(2)**(-17*x - (y%17))) % 2
results = {}
dhalf = Decimal('0.5')
for x in xrange(106):
for y in xrange(17):
# this takes a while
results[x,y] = '*' if f(Decimal(x), Decimal(y+n))>dhalf else ' '
# this prints reversed for some reason unless we reverse the x iteration
for y in xrange(17):
print(''.join(results[x,y]+' ' for x in xrange(105,-1,-1)))(And if you don't have the means to check it, it's probably better to remain agnostic, instead of starting to believe random things.)