Actually, that was double precision. You asked for float precision. >>> struct.unpack('f', struct.pack('L', 1))[0] 1.4012984643248171e-45
which is the same as: >>> 2**-149 1.4012984643248171e-45 I think that's it. Sorry for posting to the list three times in a row. corrections welcome.
-- http://mail.python.org/mailman/listinfo/python-list