Re: Parsing timestamps?

minforth <minforth@gmx.net> writes:

I don't do parallelization, but I was still surprised by the good
results using FMA. In other words, increasing floating-point number
size is not always the way to go.

Kahan was an expert in clever numerical algorithms that avoid roundoff
errors, Kahan summation being one such algorithm.  But he realized that
most programmers don't have the numerics expertise to come up with
schemes like that.  A simpler and usually effective way to avoid
roundoff error swamping the result is simply to use double or extended
precision.  So that is what he often suggested.

Here's an example of a FEM calculation that works well with 80 bit but
poorly with 64 bit FP:

https://people.eecs.berkeley.edu/~wkahan/Cantilever.pdf

Anyhow, first step is to select the best fp rounding method ....

Kahan advised compiling the program three times, once for each IEEE
rounding mode.  Run all three programs and see if the outputs differ by
enough to care about.  If they do, you have some precision loss to deal
with somehow, possibly by use of wider floats.

https://people.eecs.berkeley.edu/~wkahan/Mindless.pdf