|
Authored by: Anonymous on Saturday, March 30 2013 @ 12:50 PM EDT |
Of course they CAN be, but the common use of BCD is that 4
bits are used to represent the digits 0-9 and can either be
one digit per byte or (packed) 2 digits per byte. Of course,
there was and is hardware not based on 8-bit bytes which
might pack digits differently.
However, in the common BCD usage, there was no decimal point
mark. The code could declare how many digits would be before
or after an implied decimal point, but this was fixed point
(or scaled), and not floating point, which has a mantissa
and exponent (which btw might be base-2 (most common) or
base-16 (less common) or even some other base (not common))
as part of the number itself, as stated in the patent (did I
read that part right?).
As you mentioned, some mainframes had firmware (or possibly
hardware) support for BCD formatted numbers. These existed
well before 1995 and hence if they are infringing, then they
are also prior art.[ Reply to This | Parent | # ]
|
|
|
|
|