Re: trouble using fPart on TI-82 (fwd)


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

Re: trouble using fPart on TI-82 (fwd)



Thanks Dave, I understood that much beforehand.  What I was really looking
for is some clever person's solution to the inconsistency I mentioned (see
my first note below).  The sequence of steps below occurs in a program and
I am quite willing to add computational overhead to the program to obtain
consistency.  I have a couple of ideas (e.g. using round()), but wanted to
benefit from other's brainstorming too.  Anybody have ideas?  -Paul


On Sat, 21 Sep 1996, Dave Wollenberg wrote:
> I have an 83 (rom 1.0200) and I had the same problem. Yuo have to remember
> that the 82/83 internally deals with numbers 14 decimal digits, but only
> displays 10. So if it had 1.999999999999999 (or something real close like
> that) it would display 2, but it would know internallly that its really
> 1.whatever.
------> but then it shouldn't give the 8.5 result.
>
> On Sat, 21 Sep 1996, Paul E Kehle wrote:
> > Hello,  Can anyone help me with this problem?  Grab your TI-82 and
> > confirm the following:
> > 19/17  ENTER
> > 1.117647059
> > fPart Ans  ENTER
> > .117647059
> > Ans ^-1  ENTER
> > 8.5
> > fPart Ans  ENTER
> > .5
> > Ans ^-1  ENTER
> > 2
> > fPart Ans  ENTER
> > 1
> > !?!?  Any ideas how to get around this problem?  I assume some internal
> > flag is getting in the way of the fPart routine.  Technically I guess it's
> > correct since 19/17 = 1.11764705882352941176... and not 1.117647059 but
> > then it's inconsistent in displaying 8.5 as the reciprocal of .117647059.
> > Please either tell me this is a bug on old 82's or help me find a way
> > around this problem--meaning a way to get consistent results between
> > what the calculator is keeping track of internally and what it displays.
> >
> > Thank you in advance,
> > Paul


References: