Jump to content



Photo
- - - - -

Error In Complex Number Handling Os V3.02


  • Please log in to reply
4 replies to this topic

#1 Camarogeddon

Camarogeddon

    Newbie

  • Members
  • Pip
  • 2 posts

  • Calculators:
    Casio Classpad 300PLUS

Posted 25 October 2007 - 11:02 AM

Ok, I do know this topic has been brought up before, but after reading through the posts, i discovered that some of the implications of this error was not noticed.

The issue is regarding the handling of polar complex numbers, when angles are entered in degrees.

In electrical engineering this is widely used. angles in power systems are given in degrees, not radians. One can argue wether this is right or not, but that is beside the point.

The issue is that when entering a complex number into the classpad with it set to radians, it does this operation correctly. In other words, 5e^(90?i) exe will yield 5i , Which is correct.

Tho, with the calculator set to degrees, entering 5e^(90?i) will yield -2.24+4.47i , which is not correct. What happens is that the calculator ignores the degree sign and reads the number as radians. One would think that considering i am in degree mode 5e^(90i) would yield the correct answer, but unfortunately this also gives -2.24+4.47i .

It has been argued that one should do all calculations in radians to avoid this issue. This is true, but why did this work perfectly in OS V2.2?

But the more serious part of this is illustrated in the following example.

This is done with the calculator set to degrees, complex and decimal calculation is on.

CompToPol(4+3i)

This returns: (which is correct)
5*e^36.86989765?i

If i now execute
ans

It will return
3.377277101-3.68700412i

3.377277101-3.68700412i is NOT equal to 4+3i !In other words, one thing is that it fails to read the user input correctly, but it also fails to correctly interpret it's own answers. This is something i would expect this type of product to handle correctly. This is basic math!

I have mailed casio about this issue, first recieving the "standard" reply that has been previously posted (the solution using a variable to automatically convert the angle to radians). But after a second email, this was apparently forwarded to Japan. I am still awaiting a reply from there.

#2 vanhoa

vanhoa

    Casio Overlord

  • Members
  • PipPipPipPipPipPipPip
  • 854 posts
  • Gender:Male
  • Location:Vietnam

  • Calculators:
    AFX 2.0, CP 300, CP 330, nSpire, TI 89, FX 5800

Posted 25 October 2007 - 05:35 PM

What's wrong?

Posted Image

OS 3.0

#3 Camarogeddon

Camarogeddon

    Newbie

  • Members
  • Pip
  • 2 posts

  • Calculators:
    Casio Classpad 300PLUS

Posted 25 October 2007 - 08:39 PM

Hmm.. If you left it on degrees all the time, that IS odd. Tho i do have V3.02. Might be something in the update.

So far we have tested the issue on 5 CP300+ calculators that are upgraded to 3.02 and on one CP330, all returning the same wrong answer (we are an entire school class and 70% of us have these calculators).

Yes, we have tried resetting the calculator to factory. no help.

The ones that still run the old OS 2.2 does not exhibit this behavior.

#4 afshin_electronic

afshin_electronic

    Casio Addict

  • Members
  • PipPipPip
  • 50 posts

  • Calculators:
    casio classpad 300plus
    casia fx-991ms

Posted 26 December 2007 - 02:23 PM

me and many other are having the same problem ...
unfortunately CA$$IO has nothing to say about it ...

it's just like many other calculation mistakes and bugs in CLA$$PAS......
I have made myself a list of such stupid bugs and before any serious calculations I usually take a look at that to be more careful .....

take a look :

http://www.casiocalc...wtopic=3184&hl=

its really troublesome

#5 The_AFX_Master

The_AFX_Master

    Casio Overlord

  • [Legends]
  • PipPipPipPipPipPipPip
  • 519 posts
  • Gender:Male
  • Location:Black Mesa Research Facility (sector C)
  • Interests:BASIC +FORTRAN 90+ C++.....and HALF LIFE

  • Calculators:
    Casio Algebra FX 2.0 Plus, Casio fx 570 ms, Classpad 300, And a crowbar

Posted 26 December 2007 - 07:19 PM

can you post the list?




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users