Thread Subject: done in by a dim Sym

Never bothered with symbolic math before, but now I want to use it for high
precision calculations.
The following didn't work as I expected:

digits = 50;
pi50 = sym('3.1415926535897932384626433832795028841971693993751');
e50 = sym('2.7182818284590452353602874713526624977572470936999');
piovere = vpa(pi50/e50, 50)

% expected output (reflecting the set value of digits = 50)
piovere =
1.1557273497909217179100931833126962991208510231644

% actual output (is equivalent to digits = 32, which is the default)
piovere =
1.1557273497909217179100931833127

Apparently I misunderstand how to use DIGITS with SYM.
Would someone please explain why the output doesn't reflect the DIGITS settings?

Thanks
Ross

ross:
<SNIP symbolic conundrum...

> digits = 50;
> pi50 = sym...
> e50 = sym...
> piovere = vpa(pi50/e50, 50)
> % expected output (reflecting the set value of digits =
50)
> piovere =
> 1.1557273497909217179100931833126962991208510231644
> % actual output (is equivalent to digits = 32, which is
the default)
> piovere =
> 1.1557273497909217179100931833127

...because:

     vpa(pi50/e50,50)
% uses a sym result (pi50/e50) with DIGITS-precision...
% and the subsequent request of 50 digits
% cannot be satisfied...

% hence
     digits(100); % -or- >=50
     r=vpa(sym(pi50/e50),50)
% r = 1.1557273497909217179100931833126962991208510231644

us

"us ":
<SNIP down to superfluous syntax...

> r=vpa(sym(pi50/e50),50)

should read

     digits(100); % -or- >=50
     r=vpa(pi50/e50,50)
% r = 1.1557273497909217179100931833126962991208510231644

us

"us " <us@neurol.unizh.ch> wrote:

> digits(100); % -or- >=50

I've just realised that the condition above is not necessary.
It works fine with digits(50):
that fault was that I had written digits = 50
instead of digits(50).

ross