Are HP Prime CAS and TI NSpire CAS the same?

[hugh steers]

[Tim Wessman]

Anyone else seeing the same message for these two posts?

To answer the question in the subject - no, most definitely not.

The CAS in prime is from Bernard Parisse xcas/giac project and is supposed to be more of a maple/matlab/mathematica style CAS. It is much more powerful in many ways (then the TI) and can do many things the nspire CAS (based on derive) can't or has not expressed any desire to do. The functions and capabilities are more tuned for higher level math and users. The assumptions it makes about the users are different.

However, it does not have the same level of polish as the Nspire CAS which has basically been around unchanged since the 92/89. It appears that all of TI's efforts have basically gone into making it as user friendly as possible to spit out "school" answers. Bernard has focused on what he is interested in which tends to be more advanced things.

[hugh steers]

Thanks for the reply. Somehow these posts have come out the same. sorry for the mixup. Yes, basically my question was in the title anyhow.

Good to hear the the Prime CAS is the latest evolution. That's what i was hoping.

[CompSystems]

Quote:

---

However, it does not have the same level of polish as the Nspire CAS which has basically been around unchanged since the 92/89. It appears that all of TI's efforts have basically gone into making it as user friendly as possible to spit out "school" answers. Bernard has focused on what he is interested in which tends to be more advanced things.

---

sorry for my bad English

as is currently the HP-CAS does not exceed to TI-CAS, for example solve command is not as developed as it seems, factor fails and many other things ...

[bluesun08]

Tim, my suggestions are:

1. Make the HP Prime CAS so user friendly just as the TI.

2. Make a consistent using of variables (capital letters / small letters) in all Apps/Home-View/CAS.

3. RPN must be available in all views.

4. The basement of all APPs, especially the Function-App must be CAS!!!!!


ANYTHING ELSE IS CONFUSING AND DO NOT CONVINCE USER/TEACHER/STUDENTS!!!

[Ron Ross]

Thank you for your clear demonstration of the capabilities of each system. You have shown some excellent examples and it is very much appreciated by me.

[Tim Wessman]

There is a large difference between "the results are not formatted in the way I want", "the CAS does not make assumptions such as switching from exact into approximate automatically", or "I don't understand how to tell the CAS to give me the results I want" (which is the root of nearly all your comments) and what I am saying which is "the CAS in prime is much more powerful and capable".

All your examples are simply further statements of what I have already said - "the CAS in Nspire has been polished". However, what I am saying is that the CAS in prime has more capability, has more algorithms implemented and hence can solve a much wider range of problems. The Nspire is very good at the types of things you will encounter in a classroom because that is basically exclusively where they have focused.

Are you saying that HP should exclusively focus on classroom only things and ignore engineering or professional users? I would think the majority here would disagree with that. While Prime right now may not work for a group of 50g users, I think there is a big group of others that have been pleasantly surprised!

int(sin(x)*ln(x),x) -> -ln(x)*cos(x)+Ci(x) on the nspire??? (does here on my unit) Does the Nspire have Si, Ci, Zeta, Ei, or other special functions you will never hit in in a classroom? Didn't think so.

Please let me know how long it takes to symbolically solve int(sqrt(tan(x)),x) on your nspire... (hint, it will run for 5 minutes or so and run out of memory)

We have also tested extensively in the AP calculus exams the problems that have *completely* been unsolvable on any TI calculator using the CAS. In nearly *every* case, Prime's initial release could solve those symbolically whereas the TI CAS could not solve them either numerically or symbolically (which resulted in a LOT of very angry students and teachers for those exams that year).

True, there will be plenty of specific examples of types of problems that work in 1 CAS system, but not in another. There always will be due to the nature of math and software. However, just a casual perusal through the calculus menu, the matrix menu, or nearly any menu on Prime will find a huge range of things that the Nspire does not have, and is *very doubtful* they will ever add. The Nspire is supposed to be a pure teaching tool. Anything that detracts from that (based on 4 years of watching) will never be added.

While I appreciate your enthusiasm and time you spend making these gigantic images, they are about the most unhelpful format to communicate to HP any sort of change. If you'd like us to look at the behavior of a specific thing, please provide plain text input, output, and expected results. I just don't have time to dig through gigantic Photoshopped pictures to try and glean items. As you've pointed out, there is still a ton of polish work that needs to go into the calculator so I will be spending my time primarily on that task.

If you could please provide all your examples in plain text, I can then make every effor to ensure they either work directly, or there is effort made to clarify the correct way to solve that in Prime.

Thanks!

TW

Edited: 30 Oct 2013, 1:09 p.m.

[Tim Wessman]

>1. Make the HP Prime CAS so user friendly just as the TI.

What is the "user friendliness" enhances bad behavior (black box thinking) or mismatches the behavior of all other CAS systems the student might encounter in professional use? (for example, nsolve using default ranges for non-polynomial inputs which can very often spit out several solutions, but gives the misguided idea that any equation can have this magic box return my answer. All other CAS systems will return "a" root, and more if you specify a range in which to look - which is what Prime does)

>2. Make a consistent using of variables (capital letters / small letters) in all Apps/Home-View/CAS.

Consistency needs to be improved, but there are many issues with every single solution you can possibly propose. This is an item that needs work - 100% in agreement.

>3. RPN must be available in all views.

Agreed (with the exception of programming). Right now the only place it is not functional is the CAS screen (yes, there are some stupid bugs in places like spreadsheet :-( )

>4. The basement of all APPs, especially the Function-App must be CAS!!!!!

I completely 100% disagree here. You should be able to use CAS stuff anywhere, but a CAS is by nature a symbolic engine and not a numerical one. A calculator is first and foremost about numbers and calculations. There are a lot of advantages with having the CAS separate. The interactions between the CAS and the numerical side of the system is the true problem here.

TW

Edited: 30 Oct 2013, 2:01 p.m. after one or more responses were posted

[Michael de Estrada]

Don't forget that it should be possible to use embedded non-CAS math functions in CAS commands in programs, e.g.

solve(DET(M9-x*M8)=0) works fine inside CAS, but CAS.solve(DET(M9-'x'*M8)=0) crashes the calculator in a program.

[Tim Wessman]

Of course. Can you please point me to that discussion with your code so I can file it if needed still? I think I missed it.

TW

[Michael de Estrada]

Sure thing. Here's the entire thread:

Using CAS and non-CAS in programs

[Patrice]

5. Make the calc custom tailored to my likes. Otherwise I will say it is a piece of junk. :)

You are not alone !

[Eric Smith]

I think the legend for your first text sample should read "no antialiasing".

[cyrille de Brébisson]

try ifactor(2^128+1) on Nspire and prime... do not forget the extra battery for Nspire... Prime will be ok.

cyrille

[hugh steers]

> should read "no antialiasing"

quite correct, also it's totally the wrong post message. sorry about that.