in reply to Re^4: Unpacking extended-double precision floats
in thread Unpacking extended-double precision floats

Howdy!

If I recall correctly (and I'll be checking the references before I do anything), "double precision" and "extended double precision" have different numbers of bits in the exponent. It's not quite that simple to smush the precision down. I think I go from 64 bits of mantissa to 53 bits of mantissa (but don't hold me to precise numbers here).

In the end, I should be able to pick the three components apart and Just Do The Math.

yours,
Michael
  • Comment on Re^5: Unpacking extended-double precision floats

Replies are listed 'Best First'.
Re^6: Unpacking extended-double precision floats
by Fengor (Pilgrim) on Nov 30, 2006 at 13:08 UTC
    According to the wikipedia article on double precision floats it has 1 sign bit 11 exponent bits and 52 mantissa bits

    --
    "WHAT CAN THE HARVEST HOPE FOR IF NOT THE CARE OF THE REAPER MAN"
    -- Terry Pratchett, "Reaper Man"

[reply]

      52-actual bits, but there is also an impled bit in for an effective mantissa of 53-bits.

      The tricky bit[sic] is that in the 64-bit mantissa of the 80-bit animal, there is not an implied bit.

      Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
      Lingua non convalesco, consenesco et abolesco. -- Rule 1 has a caveat! -- Who broke the cabal?
      "Science is about questioning the status quo. Questioning authority".
      In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]
[d/l]
      Howdy!

      Yep. That's the 64-bit flavor. The 80-bit flavor is 1 sign bit, 15 or more exponent bits, and 64 or more mantissa bits.

      yours,
      Michael
[reply]

In Section Seekers of Perl Wisdom