Jump to content



Photo
- - - - -

Restricted Frequency Lists In Stat Mode


  • Please log in to reply
5 replies to this topic

#1 noname11

noname11

    Casio Technician

  • Members
  • PipPipPipPipPipPip
  • 314 posts
  • Gender:Not Telling

Posted 27 April 2010 - 05:29 PM

If you enter a value larger than 2^16-1 into a list that is going to be used by STAT mode as a frequency list, you'll get a Ma Error if GRPH was used (MedBox) or no error but missing values for Q1, Med and Q3 if CALC was used (1-Variable). Is there any workaround?

#2 TovAre

TovAre

    Casio Addict

  • Members
  • PipPipPip
  • 75 posts
  • Gender:Male
  • Location:Oslo/Norway

  • Calculators:
    fx-991CM
    fx-9860GII SD
    fx-9700-GE

Posted 28 April 2010 - 06:27 AM

If you enter a value larger than 2^16-1 into a list that is going to be used by STAT mode as a frequency list, you'll get a Ma Error if GRPH was used (MedBox) or no error but missing values for Q1, Med and Q3 if CALC was used (1-Variable). Is there any workaround?


That's really strange. A bug ?

It appears to work fine when you divide it up into multiple entries so that each individual entry is below the frequency limitation. I.e. 120 000 items are entered in two rows with a freq of 60000 in each row.

In some cases I quess dividing the frequencies would be ok, like entering 120 instead of 120e3, but the calc doesn't swallow fractional frequencies for Q1, Med and Q3 calculations so you can't do things like 1.2 instead of 1.2e9.


Maybe Casio should put this on the TODO list for a future firmware update :-)

Cheers :)

Edited by TovAre, 28 April 2010 - 08:46 AM.


#3 noname11

noname11

    Casio Technician

  • Members
  • PipPipPipPipPipPip
  • 314 posts
  • Gender:Not Telling

Posted 28 April 2010 - 08:54 AM

Thanks, that helped. :)

PS: Simon could probably fix this by manipulating a firmware image. :D

Edited by cfxm, 28 April 2010 - 08:56 AM.


#4 TovAre

TovAre

    Casio Addict

  • Members
  • PipPipPip
  • 75 posts
  • Gender:Male
  • Location:Oslo/Norway

  • Calculators:
    fx-991CM
    fx-9860GII SD
    fx-9700-GE

Posted 12 May 2010 - 08:12 AM

I asked casio about it, and apparently it's a restriction in the sorting implementation they use when computing the Q1, Q3 and median, that's probably why the restriction corresponds to a 16-bit integer.

The fx-9660GII is really speedy, so I guess they really wanted it that way. It's not a huge issue for me at least :-)

#5 TovAre

TovAre

    Casio Addict

  • Members
  • PipPipPip
  • 75 posts
  • Gender:Male
  • Location:Oslo/Norway

  • Calculators:
    fx-991CM
    fx-9860GII SD
    fx-9700-GE

Posted 11 July 2011 - 08:44 PM

I asked casio about it, and apparently it's a restriction in the sorting implementation they use when computing the Q1, Q3 and median, that's probably why the restriction corresponds to a 16-bit integer.

The fx-9660GII is really speedy, so I guess they really wanted it that way. It's not a huge issue for me at least :-)


Appears to be fixed in firmware 2.01

#6 TeamFX

TeamFX

    Casio Freak

  • Members
  • PipPipPipPip
  • 131 posts
  • Gender:Male

Posted 02 December 2016 - 05:41 PM

This bug is completely fixed as of OS 2.05 (French 2015 exam models) or OS 2.09
OS 2.04 still contains this bug.






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users