174 replies to this topic

[CrimsonCasio]

In this topic you may discuss new bugs, after discussion is ended the bug will be reduced to a report like the one below and the discusion will be placed in a new topic and linked to the report.

-------------------------------------------------------------
Bug Name: Factorial Bug
Credit: BiTwhise
Rom Versions: 1.00

Cause:

Doing a factorial of a number larger than 297

Effect:

doesn't give correct answer

Workaround:

It is in fact possible to "fool" the CP to do it, by doing it in 2 expressions (or even 1.. but some trickery is necesary to avoid the Cas system simplifying it and returning the same result as it normally would)
I created a new factorial function, fact(n):
piecewise(n>297,297!*(n!/297!+1)-297!,n!,0)

URL: http://www.casiocalc...t=ST&f=6&t=1203
-------------------------------------------------------------
Bug Name: StrToExp Bug
Credit: CrimsonCasio
Rom Versions: 1.00-1.51

Cause:

using strToExp() with a string that contains another string. ex: strToExp("{1,2,\"bob\"}")

Effect:

cuts off the string at the first " it finds, almost always causing an error or returning invalid data.

Workaround:

None

URL: http://www.casiocalc...t=ST&f=6&t=1205
-------------------------------------------------------------
Bug Name: dSolve Bug
Credit: Tonyp
Rom Versions: 1.00

Cause:

dSolve(y'+y=0,y,x,y)

Effect:

Crashes Classpad

Workaround:

None

URL: http://www.casiocalc...t=ST&f=6&t=1207
-------------------------------------------------------------
Bug Name: KeyB Bug
Credit: mkanter
Rom Versions: 1.00

Cause:

if you write a program wich prints text and also allows the user to input strings.

Effect:

Keybord is not redrawn correctly

Workaround:

None

URL: http://www.casiocalc...t=ST&f=6&t=1210
-------------------------------------------------------------
Bug Name: Geometry Bug
Credit: qwerty
Rom Versions: 1.00

Cause:

some times occures in the geometry mode, can vary but seems to be related to using animations.

Effect:

Classpad Crashes

Workaround:

Clear your geometry memory and the error will be fixed.

URL: http://www.casiocalc...t=ST&f=6&t=1211
-------------------------------------------------------------

[Pem]

Huge bug !
Try to type : solve(ln(3x+2)<ln(x-5),x)

It displays {x>5}, which is totally false ! It is impossible !

That's the first time I see my classpad displaying me a wrong solution

[Daruosh]

Try to type : solve(ln(3x+2)
It displays {x>5}, which is totally false ! It is impossible !

:/

I got {x=-1/3}
Tyr it agin

[CrimsonCasio]

same here

[Pem]

Huge bug !<br />Try to type : solve(ln(3x+2)<ln(x-5),x)<br /><br />It displays {x>5}, which is totally false ! It is impossible !<br /><br />That's the first time I see my classpad displaying me a wrong solution

Curious bug in this forum : I typed

solve(ln(3x+2)<ln(x-5),x)

and it displayed solve(ln(3x+2) in my post
(you can see it by quoting my first post)

[Daruosh]

In ClassPad i got {x>5} you're right, but in CP Manager I got No Solution !!!
My OS version is 1.2

[CrimsonCasio]

what ever the forum bug is its fixed, I'll look into this CP error...

it is still unfixed on my OS so i'll inform casio.

[Lovecasio]

No solution is right, because the equation can be written as:
3x+2>0
x-5>0
3x+2<x-5

=> No x satisfied.

[Kal Culator]

When I tried solve (ln(3x+2)<ln(x+5), I got {-2/3 < x <3/2).
I wonder if anybody else got the same result.

[Lovecasio]

Hi.
Yes, I got the same result.
About equation ln(3x+2)<ln(x-5), The ClassPad gave you x>5 in Complex mode. Change into Real mode, it couldn't solve. Lets graph y=ln(3x+2)-ln(x-5) in real mode, you will see that the graph doesn't cut horrizontal axes, and no part is under x-axes. Therefore the answer is "No Solution". I don't know why the manager can did the right thing, while ClassPad gave other answer ?

[mkanter]

3rd-pixel-bug.

Try to draw (with line or something else) at one pixel right of the left line sorrounding the graph window (it's the 3rd from left). This pixel is always blank (the program cannot draw on it)

Try this code

ViewWindow 1,155,1,-77,-1,1
'standard for small window
SetAxes Off
Cls
Line 1,-1,1,-11
Line 2,-1,2,-11

There should not be a blank space left of the 2 lines, but there is. and the left g-controler has 2 blank pixels (the others only 1) to the border.

(There is no workaround.)

----------------------------
My Webpage

[CrimsonCasio]

I dont know for sure but i always assumed that it was meant to be like that for some reason...

[Daniel (CZE)]

Another bug:

Try to find antiderivative of 1/x. What solution will you get?

I will get "ln(x)" but it isn't right. It should be "ln(|x|)".

[betoe]

Thats right, the CP must specify that:
If x>0 then the result is ln(x)
If x<0 then the result is ln(-x)

[Lovecasio]

Try it in Real mode, it will give you ln(abs(x)).

[Daniel (CZE)]

Of course! Thank you. Sometimes I'm quite

[Andy.Davies]

Sometimes I'm quite

arn't we all sometimes?

[Pem]

Hi.
Yes, I got the same result.
About equation ln(3x+2)<ln(x-5), The ClassPad gave you x>5 in Complex mode. Change into Real mode, it couldn't solve. Lets graph y=ln(3x+2)-ln(x-5) in real mode, you will see that the graph doesn't cut horrizontal axes, and no part is under x-axes. Therefore the answer is "No Solution". I don't know why the manager can did the right thing, while ClassPad gave other answer ?

Anyway, it shouldn't display x>5 in complex mode. Could anyone tell me whether it's already been reported to Casio ?

[timppa1988]

I've got a bug like that one too
solve(23.4*x^4=3.8,x) Solution (according to the CP) is 0 what is absolutly wrong...
I'm supposed to use the CP in my exams but if it gives me answers like this...

Timppa

[timppa1988]

I updated my CP OS to ver 1.24 and now it says that there is no solution which is not better than 0

[CrimsonCasio]

I get:
{x=0, x=((-1/3*i*sqrt(3))/(13^1/4))*19,x=((1/3*i*sqrt(3))/(13^1/4))*19}

[Overlord]

it's because you're in complex mode. but x=0 is false so there is a BIG problem...

[timppa1988]

Right. We orderd the CP for the whole class and on my opinion it was great but there where some who would like to it. Now I can almost understand them (even if they had other reasons, I dont think this will make them like it more )

[CrimsonCasio]

lol
its a work in progress unfortunately.

[mkanter]

try to get the value of
integral(sqrt(r^2-x^2),x,-r,r)

the formula is the one of an half-circle
and the solution is the area below it:
0.5*pi*r

in real mode the classpad says there is no solution. in complex there only are solutions with ln(...)*i

------------------------
www.mkanter.de.vu

[qwerty841]

i don't know if this is a bug

when i graph x|x>0 it works fine, but when i graph x|5>x>0 it says "syntax error"

[mkanter]

Huge bug !
Try to type : solve(ln(3x+2)<ln(x-5),x)

It displays {x>5}, which is totally false ! It is impossible !

That's the first time I see my classpad displaying me a wrong solution

ln(3x+2)<ln(x-5)

cannot be written as
3x+2<x-5


ln(3x+2)<ln(x-5) (1)

|-ln(x-5)

ln(3x-2)-ln(x-5)<0 (2)

|logarithm rules

ln((3x-2)/(x-5))<0 (3)

|substitute: ln ( x ) < 0 for 0 < x < 1

0 < (3x-2)/(x-5) < 1 (4)

the solution for that is -3/2 < x < 5

3x-2 is negative for x < 2/3
x-5 is negative for x < 5

for -3/2 < x < 2/3 you have to compare the absolute value of the complex solutions of the parts (for x = -3/2 you get an intersection)

abs(ln(3x+2))<abs(ln(x-5))

-------------------------------
www.mkanter.de.vu

[Lovecasio]

No, I think it is not right. Lets graph ln(3x+2)-ln(x-5) in real mode. No part of the graph is under x-axes, so it should be "No solution". The equation can be written 3x+2>0 and x-5>0 and 3x+2<x-5 <=> no x satisfied.
Can someone good at math verify it ?

[Overlord]

Lovecasio > was the ClassPad in real or complex mode ?

[Lovecasio]

Hi.
I tried it in real mode.

when i graph x|x>0 it works fine, but when i graph x|5>x>0 it says "syntax error"

I think you can write as y=x/x>0/x<5. It will work. (/ stand for "with symbol")

[mkanter]

i've asked an math professor at the university were i'm studying.
he said this to the problem:

"the ln(x) for negative x has periodical solutions. the comparation should (! - not have to(!)) only be done in space of the real numbers"

this means: ln(x) for x < 0 is: ln(-x) + (2n + 1) * pi * i with n is ..., -3, -2, -1, ..., 3, 4, 5,...
and there is no solution for that with real numbers so that ln(3x-2): R -> R and ln(x-5): R -> R.

comparing ln(3x-2) with ln(x-5) for x < 2/3 will give more problems,
because you have to take for the right and the left part different formulas.

TO MAKE IT SHORT THE SLUTION THE CALC GET IS WRONG. POINT.
END.

-----------------------
www.mkanter.de.vu

[mkanter]

another thing we got is an fatal exception error with the inputstr-function.

sometimes, when you don't input something (let the textfield completely empty)
it will bring the error.

then - and that's the funny thing about that - all changes (programs copied from another classpad are deleted (or not accessable))

and it looses the display calibration information

after testing a bit i found out that's the strToExp()-Function the exception produces.

The workaround for the problem is the check the string passed to strToExp:

StrCmp s,"",t
If t=0
Then
'Error
IfEnd
strToExp(s)=>var

-------------------
www.mkanter.de.vu

[Nanard]

sometimes, when i run this prog, I have an memory error but i have about 3.5 Mo free

Local tmp
StoPict tmp

I really don't undrestand what it means

[Lovecasio]

Hi.
I think it is because the variable is Local variable. So, they cannot be lagre. A picture often takes more than 3000 Bytes.

[CrimsonCasio]

no, thats not the case... i use pictures as local variables all the time, and the call function is done by creating Local programs (can be much bigger) and ive never had a problem...

has anyone else?

[John]

There is a bug in two Classpad's statistic functions.

Example:

If we have a list {1,2,3,4,5,6,7,8,9,10}
- It's first quartile (Q1) should be 2.75
- It's third quartile (Q3) should be 8.25

In the ClassPad and ClassPad Manager alike Q1 gives 3 and Q3 gives 8, which are both completely wrong .

This bug also appears in the casio spreadsheet add-in

There is also a bug in the integration function. When in complex mode we write

i
/
\ x dx we get -0,5, which has no sense.
/
0

The bug is probably caused by the fact, that the calculator first finds the antiderivative, and afterwards substitutes numbers, not thinking about the logical meaning of the whole integration expression.

[PJay]

You input {1,2,3,4,5,6,7,8,9,10} => It means : 1≤x≤10 l xєN It isn't correct !
You should input : 1≤x≤10 l xєR
Please don't ask me about input of the new code ,Because I don't know ! "It's only an opinion from me!"

[Daruosh]

When eActivity application is runing and you press the On/Off key, classpad turns off with default ending sceern. Even you have set the ENDING SCREEN.
Oh, it's really connivance BUG

[Lovecasio]

Hi.
I totally agree. The inequality can be written as system of inequality: 3x-2>0 and
x-5>0 and 3x+2<x+5 <=> no x satisfied => No solution. The ClassPad Manager did the right thing: it displayed "No solution" , while the real one ...