How Are Floats Rerpresented In Memory?

Community Forums/General Help/How Are Floats Rerpresented In Memory?

_PJ_ (Posted 2010) [#1]

Hopefuilly, the (oops misspelled!) title should say it all. I know the basics, that it's to do with bits somehow tracking the remainders from division, but I am interested in finding out how the bytes of memory are 'translated' into the decimals we understand and vice versa. Essentially, I suppose it's the core of the "Read/Write Float" commands in Blitz.

Robert Cummings (Posted 2010) [#2]

4 bytes.

Who was John Galt? (Posted 2010) [#3]

http://www.psc.edu/general/software/packages/ieee/ieee.php

_PJ_ (Posted 2010) [#4]

Thanks, John - that really was helpful :)seems a strange way of doing it though, but I guess it's to maximise the range of values to encompass the large and the small.

Charrua (Posted 2010) [#5]

btw blitz uses the 32 bits scheme : 6,5 decimal significant digits with a range of aproximately +/-10 exp +/-38 Juan