Extract an exact integer representation of a floating point number. The largest number I've been able to enter is 1.7976e308 This makes me think Scratch uses 12 bits for exponent and 52 for the mantissa? (assuming 64 bit data) yeah - that's consistent with IEEE representation: https://en.wikipedia.org/wiki/Double-precision_floating-point_format - so we should be able to reverse engineer the 64 bit representation used internally and display it in binary...
from a discussion about how to accurately checksum an array of floating point numbers without worrying about rounding errors (see https://scratch.mit.edu/projects/43755392/#comments-42137004 )
