Ilya Zakharevich on Tue, 9 May 2000 12:52:38 -0400

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: precision peculiarity

On Tue, May 09, 2000 at 03:58:49PM +0200, Karim BELABAS wrote:
> The bnfinit/quadclassunit computation don't guarantee anything about the
> actual precision of the output. They use an internal (much higher) precision
> using heuristics with the goal to get an exact class group, not an exact
> regulator.

Precision policies of functions should be documented (with function
docs, but preferably also in a separate place).  Is the answer's
precision based on argument precision?  Is it based on default
precision?  Are the internal calculations done with default precision,
and the answer has whatever precisioon it would have (:-()?  Or is it
something from blue sky, as you describe above?