Re: Parsing timestamps?

Am 11.07.2025 um 03:40 schrieb Paul Rubin:

minforth <minforth@gmx.net> writes:

Kahan was also overly critical of dynamic Unum/Posit formats.
Time has shown that he was partially wrong:
https://spectrum.ieee.org/floating-point-numbers-posits-processor

 I don't feel qualified to draw a conclusion from this.  I wonder what
the numerics community thinks, if there is any consensus.  I remember
being dubious of posits when I first heard of them, though Kahan
probably influenced that.  I do know that IEEE 754 took a lot of trouble
to avoid undesirable behaviours that never would have occurred to most
of us.  No idea how well posits do at that.  I guess though, given the
continued attention they get, they must be more interesting than I had
thought.
 I saw one of the posit articles criticizing IEEE 754 because IEEE 754
addition is not always associative.  But that is inherent in how
floating point arithmetic works, and I don't see how posit addition can
avoid it.  Let a = 1e100, b = -1e100, and c=1.  So mathematically,
a+b+c=1.  You should get that from (a+b)+c in your favorite floating
point format.  But a+(b+c) will almost certainly be 0, without very high
precision (300+ bits).

AFAIK Cuda does not support posits (yet). BFLOAT16 etc. still win the
game, until the AI industry pours big money into the chip foundries
for posit math GPUs.
Even then, it is questionable, whether or when it would seep into the
general-purpose CPU market.
For Forthers to play with, of course. ;o)