includes the error in the two's complement operation and the
roundoff error in the multiplication. As in the NR method, the final error
edr[m]
is positive as long as the generated error in the last iteration is not
negative. Also generated errors are reduced quadratically.
Error in
R[j]"
If the multiplications
R[i]d
are performed in full precision,
then the error is the same as that ofd[j] (including there the errors in
roundoff and two's complement). To that error it is necessary to add the
generated errors for all ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month, and much more.