On 4/13/07, Raymond Toy <toy.raymond at gmail.com> wrote:
>
> > it would be nicer if we could provide a way *within Maxima* to change
> > defaults, e.g. set_init(fpprintprec,6) )
> What do you mean by this? Does set_init modify an init file to set the
> given parameter to the given value?
I was giving a behavioral specification. How it works doesn't matter, but
yes, the usual implemention would be to save the information in a file,
either the regular init file, or more likely a special file just for this.
But I must confess that I don't really use maxima for anything other
> than a glorified calculator and grapher. I guess my preferences might
> differ wildly from someone who actually uses maxima to do interesting
> stuff.
I'm sure preferences vary all over the place. But I'm surprised that if you
use it this way, you actually care about the 16th decimal digit. You
certainly shouldn't be relying on it; if you really want 16 full digits, you
should probably be using bfloat with fpprec=18 or something... (depending on
the calculation).
In a previous email, you said:
> I think maxima should display everything as well as possible by default.
The strongest version of this, where the binary fraction would be expressed
exactly in decimal, is surely not a good idea. The next-strongest version,
where enough digits would be output to allow exact re-input, is a good idea
for some applications (notably when intermediate results are dumped to a
file then re-read), but gives a misleading number of digits. So in the end,
it comes down to a matter of good judgement and utility.
-s