26 Nov
2018
26 Nov
'18
8:26 a.m.
On Mon, Nov 26, 2018 at 4:28 AM Peter Corlett via cctalk <
cctalk at classiccmp.org> wrote:
On Sun, Nov 25, 2018 at 07:59:13PM -0800, Fred Cisin
via cctalk wrote:
[...]
As I understand it, 36 bits was used as it could represent a signed 10
digit decimal number in binary; the Frieden 10 digit calculator was the
"gold standard" of banking and financial institutions, so to compete in
that market, you computer had to be able to match the arithmetic standards.
-- Charles
--
X-Clacks-Overhead: GNU Terry Pratchett
Alas, "current" computers use 8, 16,
32. They totally fail to understand
the
intrinsic benefits of 9, 12, 18, 24, and 36 bits.
Oh go on then, I'm curious. What are the benefits? Is it just that there
are
useful prime factors for bit-packing hacks? And if so, why not 30?