Fatal software bug?

[Lakeboard]

Hello, everyone

If I copy a result from an earlier calculation on my HP Prime calculator, it will be copied incorrectly.

Is that my fault?

[KeithB]

Can you show us a screenshot?

[Lakeboard]

Hello KeithB

Many thanks for your response.
In my main post I have included a "gif" in which the problem can be seen. Since this might not be displayed correctly, I added a picture here:


https://i.ibb.co/16jbvvz/HP-Prime.png

[StephanP]

In what sense would this be fatal?
Okay, the parentheses got moved. But that did't break the value or the consecutive calculation, did it?
I don't see anything going fatally wrong here.

[Lakeboard]

Hello Stephan P.

Thank you, too, for your answer.
It really doesn't look too bad at first look. But the position of the parentheses is important.

Let me show you:

[ijabbott]

It seems the "Intelligent Math" option isn't very intelligent. :-)

The strange thing is that with "Entry: Algebraic", Textbook Display: [ ]", "Intelligent Math: [x]":

(10^(1/4))^3 => 10^(1/4)^3

which is incorrect, but with "Textbook Display: [x]" it gets displayed as:

\(\left(10^\frac{1}{4}\right)^3 => \left(10^\frac{1}{4}\right)^3\)

which doesn't match the incorrect non-textbook display, even though the incorrect non-textbook display is what gets copied when selected.

[StephanP]

(12-01-2021 08:44 PM)Lakeboard Wrote:  Let me show you:

Ah, my bad. I overlooked the exponent 1/64. Probably a bit fuzzy in the early morning, I was.

I cannot reproduce this on my Prime, though.
Which firmware version are you using and particularly what settings do you use in Settings page 1 and 2?

[StephanP]

Ah, got it.
It's the option Intelligent Math on Settings page 2.
(I've not used this before)

Well, that's definitely something the team wil have to look into.

[Lakeboard]

Many thanks for your help. Especially for checking @ ijabbott and Stephan.

Apparently, it is really a software bug.

So what can we do to report this issue and receive a new firmware update?
I tried to contact my local HP office. But they didn't respond.

[StephanP]

I've reported the issue with the product manager.
As to the release of a solution it's wait and see.

[Lakeboard]

Thank you, Stephan, for reporting. That’s really great!

Let's hope that there will be a new firmware soon and that the problem will be resolved.

Stephan, I may have discovered another little thing.

In CAS mode, if the imaginary number is a floating point, the angle is output as radians.

Can you also report this? Unless it is intended. But it doesn't seem that to me.

Here is a picture:

[ijabbott]

That arg() bug seems to be in the Giac/Xcas code, so one for Bernard Parisse to look at.

Here is a giac session running in a Linux terminal. The angle_radian variable seems to have changed from 0 to 1 by itself after
arg(5+6.001i)

Code:

// Using locale /usr/share/locale/
// en_GB.utf8
// /usr/share/locale/
// giac
// UTF-8
// Maximum number of parallel threads 4
Added 26 synonyms
Welcome to giac readline interface, version 1.7.0
(c) 2001,2021 B. Parisse & others
Homepage http://www-fourier.ujf-grenoble.fr/~parisse/giac.html
Released under the GPL license 3.0 or above
See http://www.gnu.org for license details
May contain BSD licensed software parts (lapack, atlas, tinymt)
-------------------------------------------------
Press CTRL and D simultaneously to finish session
Type ?commandname for help
0>> angle_radian:=0
0
// Time 0.01
>> arg(5+6.001i)
50.6594818402
// Time 0
>> angle_radian
1
// Time 0
>> arg(5+6.1i)
0.884174755465
// Time 0
>>

[KlaasKuperus]

(12-02-2021 02:21 PM)Lakeboard Wrote:  Many thanks for your help. Especially for checking @ ijabbott and Stephan.

Apparently, it is really a software bug.

So what can we do to report this issue and receive a new firmware update?
I tried to contact my local HP office. But they didn't respond.

Thank you for noticing. We will add it to the bug tracker and work on a solution asap!

[parisse]

There is indeed a bug, now fixed in giac source, thanks!

[Lakeboard]

Many thanks to KlaasKuperus and parisse.
It is incredibly nice that such bugs are fixed and that the product is actively developed and supported.
It's a great service to the community.
Thank you very much! =)

Together for a great product.

Also thanks to ijabbott for specifying the complex number problem.