The Museum of HP Calculators

HP Forum Archive 21

[ Return to Index | Top of Index ]

HP Prime Bug
Message #1 Posted by bluesun08 on 12 Oct 2013, 3:08 p.m.

1. Go in CAS-Settings and deactivate "Exact"
2. Then In CAS-View type "solve(x³+x²=x,x)"
3. Look what happen.
Whats the matter ? (I have firmware 2013 8 13)

      
Re: HP Prime Bug
Message #2 Posted by Gilles Carpentier on 12 Oct 2013, 4:15 p.m.,
in response to message #1 by bluesun08

What do you get?

I get the 3 approx results wich seems OK

-1.6...., 0, 0.618...

            
Re: HP Prime Bug
Message #3 Posted by bluesun08 on 12 Oct 2013, 5:27 p.m.,
in response to message #2 by Gilles Carpentier

Look at my video on 0:38. The calculator crash.
http://www.youtube.com/watch?v=YE1LiY6wOZo&feature=youtu.be
Sorry for the bad quality.

                  
Re: HP Prime Bug
Message #4 Posted by bluesun08 on 13 Oct 2013, 9:41 a.m.,
in response to message #3 by bluesun08

Is there anybody who can explain?

                        
Re: HP Prime Bug
Message #5 Posted by Tim Wessman on 13 Oct 2013, 10:16 a.m.,
in response to message #4 by bluesun08

I cannot. I have never seen that before. Have you made any changes in the CAS settings screen? If so, please post them.

TW

Edited: 13 Oct 2013, 10:23 a.m.

                              
Re: HP Prime Bug
Message #6 Posted by bluesun08 on 13 Oct 2013, 11:21 a.m.,
in response to message #5 by Tim Wessman

Hi Tim,

please look also to the video of Edwin. He had the same problem.
I make no other changes in the CAS or general settings. I unpacked the calculator, go in CAS mode, change the exact-mode setting an try to solve the equation.
Tim, can you reproduce the bug of a real HP Prime with firmware 213 8 13 (5106)?
By the way: I have another HP Prime with firmware 2013 7 22. This calculator doesn't reboot but show only the answer {0.} in non-exact mode.

                                    
Re: HP Prime Bug
Message #7 Posted by CompSystems on 14 Oct 2013, 11:34 a.m.,
in response to message #6 by bluesun08

Quote:
By the way: I have another HP Prime with firmware 2013 7 22. This calculator doesn't reboot but show only the answer {0.} in non-exact mode.


press enter again and get a second solution :)

eq0:= x^3+x=x; [ENTER]

solve(eq0, x); [ENTER] => returns a different answer every time you press enter

                                          
Re: HP Prime Bug
Message #8 Posted by Han on 14 Oct 2013, 12:43 p.m.,
in response to message #7 by CompSystems

Quote:
press enter again and get a second solution :)

eq0:= x^3+x=x; [ENTER]

solve(eq0, x); [ENTER] => returns a different answer every time you press enter


The answers are somewhat close to 0. The solutions are on the order of 1E-9 or 1E-18. This behavior is at least consistent with the fact that the CAS is in non-exact mode, but certainly the solver could be more accurate.

That said, there is a setting in the CAS for epsilon. So users will likely expect that the solver be at least within epsilon. The default is 1E-12, so values on the order of 1E-9 are not sufficient.

                        
Re: HP Prime Bug
Message #9 Posted by Edwin on 13 Oct 2013, 10:19 a.m.,
in response to message #4 by bluesun08

yes, it is a bug. I made another video: http://youtu.be/I1CtxqS5W4g

                              
Re: HP Prime Bug
Message #10 Posted by Marcus von Cube, Germany on 14 Oct 2013, 12:42 p.m.,
in response to message #9 by Edwin

+1 here. Reboot.

                                    
Re: HP Prime Bug
Message #11 Posted by bluesun08 on 14 Oct 2013, 1:25 p.m.,
in response to message #10 by Marcus von Cube, Germany

Do you know the reason of the mistake?

                                          
Re: HP Prime Bug
Message #12 Posted by Marcus von Cube, Germany on 14 Oct 2013, 3:27 p.m.,
in response to message #11 by bluesun08

No, I've just reproduced the steps and the calculator did a reboot.

                                                
Re: HP Prime Bug
Message #13 Posted by bluesun08 on 14 Oct 2013, 4:32 p.m.,
in response to message #12 by Marcus von Cube, Germany

really phat!

            
Re: HP Prime Bug
Message #14 Posted by bluesun08 on 13 Oct 2013, 11:22 a.m.,
in response to message #2 by Gilles Carpentier

What is your firmware version?

      
Re: HP Prime Bug
Message #15 Posted by CompSystems on 12 Oct 2013, 4:29 p.m.,
in response to message #1 by bluesun08

/!\ TESTED on HP-PRIME VIRTUAL CALCULATOR (Is Not an Emulator)

[CAS MODE & EXACT MODE ON & HP-PRIME "EMULATOR"]

solve( x^3+x=x,x); [ENTER] =>

{((-(v¬(5))-1)/2),0,((v¬(5)-1)/2)} // OK

[CAS & EXACT MODE OFF] (APPROXIMATE)

solve( x^3+x²=x,x); [ENTER] => { -1.61803398875, 0, 0.61803398875} // OK

decoding the previous output

x=-1.61803398875 OR x=0 OR x= 0.61803398875

x^3+x²=x|x=-1.61803398875; [ENTER] =>

-1.61803398875 = -1.61803398875

EVAL(Ans); [ENTER] => 1 (TRUE)

x^3+x²=x|x=0; [ENTER] =>

0 = 0

EVAL(Ans); [ENTER] => 1 (TRUE)

but eq0:= x^3+x=x;

solve(eq0, x); [ENTER] => returns a different answer every time you press enter

[enter] => ...

Another BUG with equations

solve( x²-10*x+y² = -8 AND x*y²+x-10*y = -8 , { x ,y } ); [ENTER]

=> { [ 1, 1 ], [ 2.19343941541,3.02046646812 ] }

decoding the previous output

x = 1 and y = 1 or x = 2.19343941541 and y = 3.02046646812

[CAS MODE & EXACT MODE ON]

purge(x,y); x²-10*x+y² = -8 AND x*y²+x-10*y = -8; => 0 ????? BUG

x²-10*x+y² = -8 AND x*y²+x-10*y = -8 | x = 1,y = 1; ==> 1 (true)

Left part (step to step):

x²-10*x+y² = -8|x= 1; => y²-9 = -8

y²-9 = -8|y = 1; =>

-8 = -8 => true

Right part:

x*y²+x-10*y = -8 |x= 1; => y²-10*y+1 = -8

y²-10*y+1 = -8|y = 1; =>

-8 = -8 => true

true and true => true

but

x²-10*x+y² = -8 AND x*y²+x-10*y = -8 | x = 1; => 0 ???? BUG

=> OK => y²-9 = -8 AND y²-10*y+1 => y²-9 = -8 AND y²-10*y+1

Ans| y=1 => true

[CAS & EXACT MODE OFF] (APPROXIMATE)

purge(x,y); x²-10*x+y² = -8 AND x*y²+x-10*y = -8; => x²-10*x+y² = -8 AND x*y²+x-10*y = -8; // OK

x²-10*x+y² = -8 AND x*y²+x-10*y = -8 | x = 1,y = 1 ==> 1 (true)

x:=1; y:=1;

x²-10*x+y² = -8 AND x*y²+x-10*y = -8 => 1 (true)

Edited: 13 Oct 2013, 5:10 p.m.

      
Re: HP Prime Bug
Message #16 Posted by bluesun08 on 14 Oct 2013, 10:46 a.m.,
in response to message #1 by bluesun08

Who is responsible for bugs? Is there an email-address for bugs? Or leaves hp ourselves?

            
Re: HP Prime Bug
Message #17 Posted by CompSystems on 14 Oct 2013, 11:40 a.m.,
in response to message #16 by bluesun08

Who is responsible for bugs?

HP-PRIME TEAM

Is there an email-address for bugs? Or leaves hp ourselves?

required bug reporting page

            
Re: HP Prime Bug
Message #18 Posted by Han on 14 Oct 2013, 12:49 p.m.,
in response to message #16 by bluesun08

Quote:
Who is responsible for bugs? Is there an email-address for bugs? Or leaves hp ourselves?

I am fairly positive that the folks at HP read the forums here, as well as other forums. I know that Tim, Cyrille, and Bernard in particular are active with the community, as well as many of the beta testers. You may be surprised that they frequent this board, adictoshp, cemetech, omnimaga, tiplanet, comp.sys.hp48 and many other forums.

                  
Re: HP Prime Bug
Message #19 Posted by bluesun08 on 14 Oct 2013, 1:24 p.m.,
in response to message #18 by Han

One reasonable forum would be sufficient.

                        
Re: HP Prime Bug
Message #20 Posted by Han on 14 Oct 2013, 10:48 p.m.,
in response to message #19 by bluesun08

Quote:
One reasonable forum would be sufficient.

My point was that users can choose to stick to any one forum they prefer. The folks working on the calculator do check all of them so that regardless of where you post, they will be read.

Edited: 14 Oct 2013, 10:48 p.m.


[ Return to Index | Top of Index ]

Go back to the main exhibit hall