Thanks for your message. Different results can be produced at different precisions, basically because two numbers which are the same at low precision are different at a higher precision.
So I don't think these three tests show that the software is incorrect, but they show I need to do more work in defining the tests and particularly the starting set given by ssfn. It excludes 2 from the set, which was a mistake. Regards On Thu, Aug 18, 2016 at 10:24 AM, Erling Hellenäs <[email protected]> wrote: > Hi all ! > > I am in J8.0.4 in 64 bit Windows.* > * > > In Lab 36 I get: > > (P1_P2 OR TWO OR BIGPOS) = MINUSONE PLUS BIGPOS NB. check this > > 0 > > > HALF = ONE MINUS HALF > > 0 > > > ONE = HALF TIMES TWO NB. not in 3-bit cos TWO=BIGPOS > > 0 > > > These three tests are failing? > > > Cheers, > > > Erling > > > > > On 2016-08-17 15:32, Raul Miller wrote: > >> Yes. >> >> ({.~ i:&'/');(4!:4 <'is_unum') { 4!:3'' >> >> gives the directory where unum's utils.ijs was loaded from. (And this >> can even be done in that same script, after is_unum has been defined.) >> >> That said, for the long run, it would probably be better to stuff this >> in a variable and use it where needed, instead of changing the >> directory for the entire session. Changing the directory should be >> reserved for code that is supposed to take over the session (or, ok, >> labs which are intended to teach the user about changing directory). >> >> Thanks, >> >> > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
