From Newsgroup: alt.lang.asm

Rosario,

if 64 number is a list of binary numbers, or 2 32 bits int
numbers in memory, there is nothing that one miss

The problem occurs when converting a (too big) 64-bit *float* into a 64-bit *integer*.

if 64 bit number is ask as input, in decimal form (or something
as long double 1.8968685675745464643;), yes some imprecision has
to be in the last binary digits

Thats also problem, but not as big as an the overflow described above.

And than there is the loss-of-precision that can occur when a 64-bit
*integer* is converted into a 64-bit *float*. If the integer has bits set above the 53-th bit than the float will not be able to exactly store that value (some of the integers lower bits will disappear).


Though as my previous message mentions, I've been able to find that all of
the above problems are caught caught and available in the FPU status word.

Thanks for the reply though.

Regards,
Rudy Wieser


--- Synchronet 3.21a-Linux NewsLink 1.2