Post Reply 
The lack of handling root functions in hp prime
12-22-2018, 03:29 AM (This post was last modified: 12-28-2018 06:50 AM by yangyongkang.)
Post: #1
The lack of handling root functions in hp prime
As we all know, rational functions are better than transcendental functions, and polynomial is better than root in rational functions. Therefore, it is more difficult to deal with root simplification. Because of the large computational memory and fast speed on the simulator or XCAS, the problem of jamming or restarting rarely occurs, but the reason for the operation card being stuck or restarted often occurs on the hp prime physical machine. First come to the classic and simple representative example: simplify(sqrt(x+y+2*sqrt(x*y))|x>0, y>0), we want to get sqrt(x)+sqrt(y), But did not get it. Another example: simplify(λ*√((x-a)^2+y^2)+μ*√(x^2+(y-b)^2)|y=sqrt(r^2-x^2)), Simplify(∂(λ*sqrt(a^2-2*x*a+r^2)+μ*sqrt(b^2-2*sqrt(r^2-x^2)*b+r^2) ,x)),solve(∂(λ*sqrt(a^2-2*x*a+r^2)+μ*sqrt(b^2-2*sqrt(r^2-x^2)*b +r^2), x)=0, x),solve(x^2+(x-2)^2+2*y^2-sqrt(x^2+y^2)*sqrt((x-2)^2+y^2)=4,y),solve((2*√3*x^2-2*√3*x*x0+2*√3*y^2-√3*√(3*x^4-6*x0*x^3+6*x^2*y^2+(-2*√3*√(-3*x0^2+6*x0+1)+2*√3)*x^2*y+(6*x0+2-2*√(-3*x0^2+6*x0+1))*x^2-6*x0*x*y^2+3*y^4+(-2*√3*√(-3*x0^2+6*x0+1)+2*√3)*y^3+(6*x0+2-2*√(-3*x0^2+6*x0+1))*y^2)-2*y*√(-3*x0^2+6*x0+1)+2*y) = 0,x),simplify(((-(sqrt(3)))*sqrt(3*((-3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))+((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+3))/6)-((sqrt((3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))-((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+6+48/(sqrt(3)*sqrt(3*((-3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))+((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+3)))))/2)-(1/2),((-(sqrt(3)))*sqrt(3*((-3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))+((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+3))/6)+((sqrt((3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))-((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+6+48/(sqrt(3)*sqrt(3*((-3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))+((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+3)))))/2)-(1/2),(sqrt(3)*sqrt(3*((-3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))+((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+3))/6)-((sqrt((3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))-((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+6-48/(sqrt(3)*sqrt(3*((-3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))+((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+3)))))/2)-(1/2),(sqrt(3)*sqrt(3*((-3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))+((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+3))/6)+((sqrt((3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))-((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+6-48/(sqrt(3)*sqrt(3*((-3*2^(1/3)/(2*(2*sqrt(13)+5))^(1/3))+((2*(2*sqrt(13)+5))^(1/3)/2^(1/3))+3)))))/2)-(1/2)),tcollect((-sqrt(-12*cos(2*x)+26*cos(3*x)+40*cos(4*x)+30*cos(5*x)+12*cos(6*x)+2*cos(7*x)-58*cos(x)-42*sin(2*x)-26*sin(3*x)-2*sin(4*x)+14*sin(5*x)+14*sin(6*x)+6*sin(7*x)+sin(8*x)-34*sin(x)-40)*y*sin(x)-sqrt(-12*cos(2*x)+26*cos(3*x)+40*cos(4*x)+30*cos(5*x)+12*cos(6*x)+2*cos(7*x)-58*cos(x)-42*sin(2*x)-26*sin(3*x)-2*sin(4*x)+14*sin(5*x)+14*sin(6*x)+6*sin(7*x)+sin(8*x)-34*sin(x)-40)*y+8*y*cos(x)^5-8*y*cos(x)^4*sin(x)-8*y*cos(x)^4-32*y*cos(x)^3*sin(x)-40*y*cos(x)^3-8*y*cos(x)^2*sin(x)-8*y*cos(x)^2+16*y*cos(x)*sin(x)+16*y*cos(x))/(16*cos(x)^5+16*cos(x)^4-32*cos(x)^3*sin(x)-32*cos(x)^3-32*cos(x)^2*sin(x)-32*cos(x)^2)), these are slow or stuck or restarted on the hp prime physical machine. The above is the problem I found.Off-topic, it is a Christmas in the West. As a Chinese, I hope everyone is happy Christmas. We are also doing activities here to celebrate the preparation of Christmas, although the relationship between China and the West is very delicate.sorry my poor english


Attached File(s) Thumbnail(s)
   

study hard, improve every day
Visit this user's website Find all posts by this user
Quote this message in a reply
12-22-2018, 05:22 PM
Post: #2
RE: The lack of handling root functions in hp prime
Hello, try the online version of xcas and check if there are also problems, https://www-fourier.ujf-grenoble.fr/~par...casen.html
Find all posts by this user
Quote this message in a reply
12-22-2018, 06:01 PM
Post: #3
RE: The lack of handling root functions in hp prime
simplify(sqrt(x+y+2*sqrt(x*y))|x>0, y>0):
It looks simple, because *you* know how to do the simplification, but it is hard to have an algorithm that can solve that: you have to compute with algebraic extensions of Q[x,y]. The best I can do in Xcas (after fixing some code about assumptions) is return 0 for simplify(sqrt(x+y+2*sqrt(x*y))-sqrt(x)-sqrt(y)|x>0, y>0). It's slow on a desktop, therefore it won't work on the Prime.
In the second example, you probably meant
simplify(λ*√((xa)^2+y^2)+μ*√(x^2+(yb)^2)|yb=sqrt(r^2-x^2))
not y, and there the result seems OK.
solve expects polynomial-like equations. It's (again) very hard to handle sqrt or other fractional exponents.
Find all posts by this user
Quote this message in a reply
12-22-2018, 06:43 PM
Post: #4
RE: The lack of handling root functions in hp prime
(12-22-2018 06:01 PM)parisse Wrote:  simplify(sqrt(x+y+2*sqrt(x*y))|x>0, y>0):
It looks simple, because *you* know how to do the simplification, but it is hard to have an algorithm that can solve that: you have to compute with algebraic extensions of Q[x,y].

WolframAlpha returns for Simplify[Sqrt[x + y + 2 Sqrt[x y]]]:
Quote:piecewise | sqrt(x) + sqrt(y) | -π/2<arg(sqrt(x) + sqrt(y))<=π/2
-sqrt(x) - sqrt(y) | (otherwise)

Quote:It's slow on a desktop, therefore it won't work on the Prime.

It doesn't feel much slower than simpler examples.
However I don't know how much it takes to also render the 3D plot and the Contour plot.

Cheers
Thomas
Find all posts by this user
Quote this message in a reply
12-23-2018, 06:59 AM
Post: #5
RE: The lack of handling root functions in hp prime
It's certainly possible to implement simple algorithms to simplify expressions like that, but they will just be able to handle some kind of expressions, and that's precisely expressions that a human can easily detect and simplify mentally. What I have implemented is a normal form, a little bit like when you expand a polynomial expression, if it does not return 0 then you are sure the polynomial is not 0. An analogy could be integration by table lookup and integration using the Risch algorithm.
The algorithm is computationnally expensive, because you must handle several algebraic extensions of Q[x,y]. I agree that this is not really adapted to the Prime, because the normal form is expressed as a rootof, an object that HP found mathematically to complicated to be in a command result.
Find all posts by this user
Quote this message in a reply
12-23-2018, 09:01 AM
Post: #6
RE: The lack of handling root functions in hp prime
(12-23-2018 06:59 AM)parisse Wrote:  It's certainly possible to implement simple algorithms to simplify expressions like that, but they will just be able to handle some kind of expressions, and that's precisely expressions that a human can easily detect and simplify mentally. What I have implemented is a normal form, a little bit like when you expand a polynomial expression, if it does not return 0 then you are sure the polynomial is not 0. An analogy could be integration by table lookup and integration using the Risch algorithm.
The algorithm is computationnally expensive, because you must handle several algebraic extensions of Q[x,y]. I agree that this is not really adapted to the Prime, because the normal form is expressed as a rootof, an object that HP found mathematically to complicated to be in a command result.
Thanks to the professor, Mr. Bernard can see my other post, but also about CAS, but no one answered me. There are still some questions that I will come up with in succession.

study hard, improve every day
Visit this user's website Find all posts by this user
Quote this message in a reply
12-28-2018, 02:51 PM (This post was last modified: 12-28-2018 03:03 PM by yangyongkang.)
Post: #7
RE: The lack of handling root functions in hp prime
This equation group, XCAS calculations fall into an infinite loop.
Code:
 f(x,y):=4*sqrt(16*y^2*tan(x/2)^8*(1/(tan(x/2)^2+1))^8+16*y^2*tan(x/2)^6*(1/(tan(x/2)^2+1))^8+16*y^2*tan(x/2)^7*(1/(tan(x/2)^2+1))^7+48*y^2*tan(x/2)^5*(1/(tan(x/2)^2+1))^7+4*y^2*tan(x/2)^6*(1/(tan(x/2)^2+1))^6-32*y^2*tan(x/2)^7*(1/(tan(x/2)^2+1))^7/(tan(x/2)^2+1)+20*y^2*tan(x/2)^4*(1/(tan(x/2)^2+1))^6-32*y^2*tan(x/2)^6*(1/(tan(x/2)^2+1))^6/(tan(x/2)^2+1)-16*tan(x/2)^4*(1/(tan(x/2)^2+1))^6-8*y^2*tan(x/2)^3*(1/(tan(x/2)^2+1))^5+16*tan(x/2)^6*(1/(tan(x/2)^2+1))^6/(tan(x/2)^2+1)-8*y^2*tan(x/2)^5*(1/(tan(x/2)^2+1))^5/(tan(x/2)^2+1)-16*tan(x/2)^3*(1/(tan(x/2)^2+1))^5-4*y^2*tan(x/2)^2*(1/(tan(x/2)^2+1))^4+16*tan(x/2)^5*(1/(tan(x/2)^2+1))^5/(tan(x/2)^2+1)-4*tan(x/2)^3*(1/(tan(x/2)^2+1))^3/(tan(x/2)^2+1)+(1/(tan(x/2)^2+1))^2-tan(x/2)^2*(1/(tan(x/2)^2+1))^2/(tan(x/2)^2+1)+4*tan(x/2)*(1/(tan(x/2)^2+1))^2/(tan(x/2)^2+1))/(8*tan(x/2)/(tan(x/2)^4+2*tan(x/2)^2+1)-32*tan(x/2)^3*(1/(tan(x/2)^2+1))^3/(tan(x/2)^2+1)-16*tan(x/2)^2*(1/(tan(x/2)^2+1))^2/(tan(x/2)^2+1)+4/(tan(x/2)^2+1))-8*y*tan(x/2)/((tan(x/2)^4+2*tan(x/2)^2+1)*(8*tan(x/2)/(tan(x/2)^4+2*tan(x/2)^2+1)-32*tan(x/2)^3*(1/(tan(x/2)^2+1))^3/(tan(x/2)^2+1)-16*tan(x/2)^2*(1/(tan(x/2)^2+1))^2/(tan(x/2)^2+1)+4/(tan(x/2)^2+1)))-8*y*tan(x/2)^2*(1/(tan(x/2)^2+1))^2/((tan(x/2)^2+1)*(8*tan(x/2)/(tan(x/2)^4+2*tan(x/2)^2+1)-32*tan(x/2)^3*(1/(tan(x/2)^2+1))^3/(tan(x/2)^2+1)-16*tan(x/2)^2*(1/(tan(x/2)^2+1))^2/(tan(x/2)^2+1)+4/(tan(x/2)^2+1)))+16*y*tan(x/2)^3*(1/(tan(x/2)^2+1))^3/((tan(x/2)^2+1)*(8*tan(x/2)/(tan(x/2)^4+2*tan(x/2)^2+1)-32*tan(x/2)^3*(1/(tan(x/2)^2+1))^3/(tan(x/2)^2+1)-16*tan(x/2)^2*(1/(tan(x/2)^2+1))^2/(tan(x/2)^2+1)+4/(tan(x/2)^2+1)))+8*y*tan(x/2)^3*(1/(tan(x/2)^2+1))^3/(8*tan(x/2)/(tan(x/2)^4+2*tan(x/2)^2+1)-32*tan(x/2)^3*(1/(tan(x/2)^2+1))^3/(tan(x/2)^2+1)-16*tan(x/2)^2*(1/(tan(x/2)^2+1))^2/(tan(x/2)^2+1)+4/(tan(x/2)^2+1))+16*y*tan(x/2)^4*(1/(tan(x/2)^2+1))^4/(8*tan(x/2)/(tan(x/2)^4+2*tan(x/2)^2+1)-32*tan(x/2)^3*(1/(tan(x/2)^2+1))^3/(tan(x/2)^2+1)-16*tan(x/2)^2*(1/(tan(x/2)^2+1))^2/(tan(x/2)^2+1)+4/(tan(x/2)^2+1))+y;solve([diff(f(x,y),x)=0,diff(f(x,y),y)=0],[x,y]).
Wolfram Mathematica 11.3 is also not ideal.But in fact this equation system does have an exact solution. Wolfram Mathematica's code
Code:
Solve[D[((

       Sec[x/2]^8 Sqrt[

        Cos[x/2]^12 (2 - y^2 + y^2 Cos[2 x] + 2 Sin[x]) (-1 + 

           Tan[x/2])^2 (1 + Tan[x/2])^4])/Sqrt[2] + 

       y (1 + Tan[x/2])^2 (1 - 2 Tan[x/2] + Tan[x/2]^4))/((-1 + 

         Tan[x/2])^2 (1 + Tan[x/2])^4), x] == 0 && 

  D[((Sec[x/2]^8 Sqrt[

        Cos[x/2]^12 (2 - y^2 + y^2 Cos[2 x] + 2 Sin[x]) (-1 + 

           Tan[x/2])^2 (1 + Tan[x/2])^4])/Sqrt[2] + 

       y (1 + Tan[x/2])^2 (1 - 2 Tan[x/2] + Tan[x/2]^4))/((-1 + 

         Tan[x/2])^2 (1 + Tan[x/2])^4), y] == 0, {x, y}, Reals]

study hard, improve every day
Visit this user's website Find all posts by this user
Quote this message in a reply
12-28-2018, 03:39 PM
Post: #8
RE: The lack of handling root functions in hp prime
Let's take a look at the equation of this nonlinear triangle plus root number, and XCAS is also in a loop.
Quote:solve([(cos(α)) = ((a+b)^2+(a+b+c)^2-(a+c)^2)/(2*(a+b)*(a+b+c)),(cos(ξ)) = ((a+b+c)^2+(a+c)^2-(a+b)^2)/(2*(a+b+c)*(a+c)),(b^2+a^2-2*a*b*cos(ξ)) = (x^2),(cos(β)) = (b^2+x^2-a^2)/(2*b*x),β = (2*α)],[α,β,ξ,x,c])

study hard, improve every day
Visit this user's website Find all posts by this user
Quote this message in a reply
12-28-2018, 06:52 PM
Post: #9
RE: The lack of handling root functions in hp prime
While I'm happy to see Xcas compared to Mathematica, it is obvious that I do not have the ressources of Mathematica. There are obviously problems that Mathematica will solve and Xcas will not (but there are also problems that Mathematica will not solve that Xcas solves...).
If you don't like Xcas, nobody force you to use it. Maybe you have noticed that Xcas is free, this is certainly not the case of Mathematica. And comparing Mathematica on a desktop with an Xcas port on a calculator like the Prime is not precisely what I would call a fair comparison.
Now I'm on holiday like many other, I won't comment anymore on your posts.
Find all posts by this user
Quote this message in a reply
12-29-2018, 01:07 AM (This post was last modified: 12-29-2018 01:08 AM by Luigi Vampa.)
Post: #10
RE: The lack of handling root functions in hp prime
yangyongkang ​您: as Prof. Parisse has already noted, trying to compare Wolfram Mathematica vs. XCAS/GIAC doesn't seem very fair. They are simply different species belonging to the same genus of CAS. The former plays in the commercial SW ground, while the latter stems from the academic world, and it has been successfully incarnated in different calculators. AFAIK, no Wolfram Mathematica version is available in a calculator, is it?
Please, note I have been a Wolfram Mathematica user, and I really admire such a piece of SW. Nevertheless, saying Prof. Parisse's brainchild is outstanding would be an evident understatement, specially when you take into account its development constraints and its main purpose, mostly academic. I wish I could have count on his work during those years I spent at the university, when I was younger.
Besides, I would really suggest you spend some words introducing your scenario, before you go to the point. There are plenty of gurus in this forum (don't get me wrong, myself not included at all). They are not only brilliant people, but also truly helpful, though you might need to accommodate your wording to the western style of asking for support, since 個人淺見 I think you went too straight to the point :O) ​谢谢大家配合。
​​最后,祝大家新年快乐!
​88

PS: I have no affiliation with XCAS/GIAC. In fact, the only CAS I seldom use now is Maxima​​​ ​​哈哈哈

Saludos Saluti Cordialement Cumprimentos MfG BR + + + + +
Luigi Vampa +
Free42 '<3' I + +
Find all posts by this user
Quote this message in a reply
12-29-2018, 01:12 AM (This post was last modified: 12-29-2018 12:58 PM by yangyongkang.)
Post: #11
RE: The lack of handling root functions in hp prime
(12-29-2018 01:07 AM)Luigi Vampa Wrote:  yangyongkang ​您: as Prof. Parisse has already noted, trying to compare Wolfram Mathematica vs. XCAS/GIAC doesn't seem very fair. They are simply different species belonging to the same genus of CAS. The former plays in the commercial SW ground, while the latter stems from the academic world, and it has been successfully incarnated in different calculators. AFAIK, no Wolfram Mathematica version is available in a calculator, is it?
Please, note I have been a Wolfram Mathematica user, and I really admire such a piece of SW. Nevertheless, saying Prof. Parisse's brainchild is outstanding would be an evident understatement, specially when you take into account its development constraints and its main purpose, mostly academic. I wish I could have count on his work during those years I spent at the university, when I was younger.
Besides, I would really suggest you spend some words introducing your scenario, before you go to the point. There are plenty of gurus in this forum (don't get me wrong, myself not included at all). They are not only brilliant people, but also truly helpful, though you might need to accommodate your wording to the western style of asking for support, since 個人淺見 I think you went too straight to the point :O) ​谢谢大家配合。
​​最后,祝大家新年快乐!
​88

PS: I have no affiliation with XCAS/GIAC. In fact, the only CAS I seldom use now is Maxima​​​ ​​哈哈哈
I did not compare the meaning of XCAS and Wolfram Mathematica. I think the reason why it is misunderstood is that my English is too bad. I sometimes have to use Google translation, which leads to a misunderstanding of semantics. China attaches great importance to the discipline of English, but the society lacks the atmosphere to learn English. Moreover, the differences between Chinese and Western cultures are relatively large, which leads to some misunderstandings in language communication. This is inevitable and requires everyone to be tolerant. I am not malicious, I really like hp prime and XCAS, I just raised some questions. Soory my poor english 祝大家新年快乐。希望世界和平

study hard, improve every day
Visit this user's website Find all posts by this user
Quote this message in a reply
12-29-2018, 10:47 AM
Post: #12
RE: The lack of handling root functions in hp prime
(12-29-2018 01:07 AM)Luigi Vampa Wrote:  yangyongkang ​您: as Prof. Parisse has already noted, trying to compare Wolfram Mathematica vs. XCAS/GIAC doesn't seem very fair.

Yangyongkang​ was NOT making any "comparison" between the Wolfram product and XCAS. Understandably, Parisse is very protective of his work, his time spent, and the difficulty of making change, as it relates to unintended consequences in other parts of his code work. This makes it difficult, for users such as myself, when the use of his product leads to different results than expected.

This is the problem. So to mitigate differences, (like Yangyongkang), ​I often use the quick, on-line version, of Wolfram Alpha, to see if, (and, possibly, where), differences might be.

So it ISN'T "product comparison," it's "result confirmation." There are at least three ways this might be adding value to the Prime CAS, and XCAS, in general:

1. There may be an undiscovered bug in the CAS code.
2. There may need to be changes to accommodate unforeseen cases.
3. Results may, in fact, be wrong.

Speaking only for myself, when I encounter, and try to discuss "frustrating" results like this, I don't mean any disrespect towards Parisse. My goal is understanding, and to try to lead to improvements in the product. Language barriers, and personality challenges, impede progress. Sometimes, I sense that the personality side, overrides the goal objective. I don't mean for that to happen. I'm not a so-called "groupie." I purchased the product, not the author. Other resources are available, and suggesting that no one is forcing anyone to use one or the other, obscures the purpose of the topic under discussion.

The best outcome is when Parisse explains how his code result is correct, (or not). The fact that the Bernard Parisse legacy is very well established, immortalized in the history of hp calculators, etc., isn't particularly relevant to the issue of the moment. No offense should be taken over that, when questioning problems of the day.
Find all posts by this user
Quote this message in a reply
12-29-2018, 12:53 PM
Post: #13
RE: The lack of handling root functions in hp prime
(12-29-2018 10:47 AM)DrD Wrote:  
(12-29-2018 01:07 AM)Luigi Vampa Wrote:  yangyongkang ​您: as Prof. Parisse has already noted, trying to compare Wolfram Mathematica vs. XCAS/GIAC doesn't seem very fair.

Yangyongkang​ was NOT making any "comparison" between the Wolfram product and XCAS. Understandably, Parisse is very protective of his work, his time spent, and the difficulty of making change, as it relates to unintended consequences in other parts of his code work. This makes it difficult, for users such as myself, when the use of his product leads to different results than expected.

This is the problem. So to mitigate differences, (like Yangyongkang), ​I often use the quick, on-line version, of Wolfram Alpha, to see if, (and, possibly, where), differences might be.

So it ISN'T "product comparison," it's "result confirmation." There are at least three ways this might be adding value to the Prime CAS, and XCAS, in general:

1. There may be an undiscovered bug in the CAS code.
2. There may need to be changes to accommodate unforeseen cases.
3. Results may, in fact, be wrong.

Speaking only for myself, when I encounter, and try to discuss "frustrating" results like this, I don't mean any disrespect towards Parisse. My goal is understanding, and to try to lead to improvements in the product. Language barriers, and personality challenges, impede progress. Sometimes, I sense that the personality side, overrides the goal objective. I don't mean for that to happen. I'm not a so-called "groupie." I purchased the product, not the author. Other resources are available, and suggesting that no one is forcing anyone to use one or the other, obscures the purpose of the topic under discussion.

The best outcome is when Parisse explains how his code result is correct, (or not). The fact that the Bernard Parisse legacy is very well established, immortalized in the history of hp calculators, etc., isn't particularly relevant to the issue of the moment. No offense should be taken over that, when questioning problems of the day.
Thank you very much, this brother, you really understand me. I did not compare the meaning of XCAS and Wolfram Mathematica. I think the reason why it is misunderstood is that my English is too bad. I sometimes have to use Google translation, which leads to a misunderstanding of semantics. China attaches great importance to the discipline of English, but the society lacks the atmosphere to learn English. Moreover, the differences between Chinese and Western cultures are relatively large, which leads to some misunderstandings in language communication. This is inevitable and requires everyone to be tolerant. I am not malicious, I really like hp prime and XCAS, I just raised some questions. Soory my poor english

study hard, improve every day
Visit this user's website Find all posts by this user
Quote this message in a reply
12-29-2018, 12:54 PM
Post: #14
RE: The lack of handling root functions in hp prime
(12-29-2018 10:47 AM)DrD Wrote:  
(12-29-2018 01:07 AM)Luigi Vampa Wrote:  yangyongkang ​您: as Prof. Parisse has already noted, trying to compare Wolfram Mathematica vs. XCAS/GIAC doesn't seem very fair.

Yangyongkang​ was NOT making any "comparison" between the Wolfram product and XCAS. Understandably, Parisse is very protective of his work, his time spent, and the difficulty of making change, as it relates to unintended consequences in other parts of his code work. This makes it difficult, for users such as myself, when the use of his product leads to different results than expected.

This is the problem. So to mitigate differences, (like Yangyongkang), ​I often use the quick, on-line version, of Wolfram Alpha, to see if, (and, possibly, where), differences might be.

So it ISN'T "product comparison," it's "result confirmation." There are at least three ways this might be adding value to the Prime CAS, and XCAS, in general:

1. There may be an undiscovered bug in the CAS code.
2. There may need to be changes to accommodate unforeseen cases.
3. Results may, in fact, be wrong.

Speaking only for myself, when I encounter, and try to discuss "frustrating" results like this, I don't mean any disrespect towards Parisse. My goal is understanding, and to try to lead to improvements in the product. Language barriers, and personality challenges, impede progress. Sometimes, I sense that the personality side, overrides the goal objective. I don't mean for that to happen. I'm not a so-called "groupie." I purchased the product, not the author. Other resources are available, and suggesting that no one is forcing anyone to use one or the other, obscures the purpose of the topic under discussion.

The best outcome is when Parisse explains how his code result is correct, (or not). The fact that the Bernard Parisse legacy is very well established, immortalized in the history of hp calculators, etc., isn't particularly relevant to the issue of the moment. No offense should be taken over that, when questioning problems of the day.
Thank you very much, this brother, you really understand me. I did not compare the meaning of XCAS and Wolfram Mathematica. I think the reason why it is misunderstood is that my English is too bad. I sometimes have to use Google translation, which leads to a misunderstanding of semantics. China attaches great importance to the discipline of English, but the society lacks the atmosphere to learn English. Moreover, the differences between Chinese and Western cultures are relatively large, which leads to some misunderstandings in language communication. This is inevitable and requires everyone to be tolerant. I am not malicious, I really like hp prime and XCAS, I just raised some questions. Soory my poor english

study hard, improve every day
Visit this user's website Find all posts by this user
Quote this message in a reply
12-29-2018, 01:27 PM
Post: #15
RE: The lack of handling root functions in hp prime
[OFF TOPIC]

(12-28-2018 06:52 PM)parisse Wrote:  While I'm happy to see Xcas compared to Mathematica, it is obvious that [...]

@DrD, please note I just was following the mention from XCAS/GIAC's author about his feeling regarding the 'comparison'. I don't mind being this thread's scapegoat anyway >D Facts and feelings play in very different grounds, so everybody is got the reason.

Saludos Saluti Cordialement Cumprimentos MfG BR + + + + +
Luigi Vampa +
Free42 '<3' I + +
Find all posts by this user
Quote this message in a reply
12-29-2018, 02:04 PM
Post: #16
RE: The lack of handling root functions in hp prime
(12-28-2018 02:51 PM)yangyongkang Wrote:  ... Wolfram Mathematica 11.3 is also not ideal.But in fact this equation system does have an exact solution...

Please note, it was Parisse asserting "comparison," not yangyongkang. The "confirmation" attempted, (by using Wolfram), was "also not ideal," which is the whole point. It's only a means of confirming results, not a competition event.

All of that is beside the point though, and no scapegoating is being presumed here. When we can discuss results, focused on facts, not personalities, so much the better. For example, I don't always use the Wolfram product when trying to reconcile results. It just happens to be under my fingertips at the keyboard, instead of loading another product, (like wxmaxima), and so forth.

The undertaking of XCAS, and Parisse' own background, and his relationship with hp over the years are sound credentials. It needs no defending. However, bugs, have been found, code has changed, and sometimes results were wrong.

These are to be expected, and there is no shame in confronting them, or defending the outcomes. Personal challenges don't serve the cause, and are pretty worthless, overall.
Find all posts by this user
Quote this message in a reply
12-30-2018, 07:59 AM
Post: #17
RE: The lack of handling root functions in hp prime
I fully understand that yangyongkang can have problems to write in English (since I have too). However he choosed to report about advanced and sometimes complicated inputs (some of them were probably not typed on the calculator) and he did not mention about the software warnings when he wrote the topic of title "This time XCAS and hp prime are really wrong! ! !". Look at this title : it's like he is happy to eventually find a defect in Xcas (after several trials where Xcas was just running forever). And everywhere he reports about the same computations with Wolfram Alpha or Mathematica, that really looks like a comparison. DrD did also compare with Wolfram Alpha in the topic "Eigenvectors", assuming de facto that Wolfram Alpha answer was the best possible answer, I had to post twice to explain that it is not.
So yes, I react, and I believe that everyone would react too.

Now let's make things clear: I'm open to bug reports and improvement suggestions and I do my best to take them in account. But I would appreciate a more friendly tone. Remember that Giac/Xcas is a free software, not a commercial one (not to say that commercial softwares should be criticized harder than free ones, but they usually do not participate in open forums like here or they give less informative answers because the developers do not participate themselves). One can of course argue that the HP Prime port is commercial, but in reality ask yourself how CAS improvements and bug fixes are done. And keep in mind that my ressources and priorities in Xcas are not the same as Mathematica.
Find all posts by this user
Quote this message in a reply
12-30-2018, 10:08 AM
Post: #18
RE: The lack of handling root functions in hp prime
Hello,
My two cents. I think that the era of calculators is over. What is the point of using them in this time and age? Yes, HP Prime is very capable calculating machine and nobody can deny that. It is also true that HP Prime has a lot of bugs and/or inconsistencies and this does not mean that it is the fault of prof. B.Parisse. In general it is amazing how much you can do with this calculator, but it is also true that it takes a lot of patience and perseverance to learn how to get the most out of it. What are graphing calculators for? The obvious and I think false answer is: to learn mathematics. I doubt that you can learn mathematics using for example calculator such as HP Prime. The HP Prime can only speed up your work and hopefully catch some mistakes in your work. The not so generally acknowledged fact is that mostly students are using graphing calcs. to 'cheet' if possible on exams. No graphing calc. will think for you,ever because if it could you become redundant and the calc. will not need you. Since, in many schools that care about real teaching you are not allowed to use the graphing calculators on exams so, the only role left to that calc is to help doing your studies and/or homework. My opinion is: why bother with that little and powerful 'toy' if you can get yourself a lot of real viewing estate with 13.3 in. convertible and install GIAC, XCAS, Mathematica, etc. or whatever your heart desires. This is what I will do in 2019, I am waiting for good and reasonably priced 13.3 in. convertible with the specs I want and I will buy it. What I write is in no way to disrespect and/or diminish the work prof. B. Parisse put in his project, but now it is the time to move on and why to doodle on that little screen if you can have 13.3 in. laptop. True, it takes more space and it does not fit in the pocket, but hey, who cares. The time will soon come that somebody will make real computer to put it in your pocket and the screen of your choosing you will be able to carry in your bag. Happy New Year 2019 for everybody.
John P.
Find all posts by this user
Quote this message in a reply
12-30-2018, 11:25 AM (This post was last modified: 12-30-2018 11:30 AM by Luigi Vampa.)
Post: #19
RE: The lack of handling root functions in hp prime
[OFF TOPIC]


Forget about calcs, PC's and tablets, the next generation simply prefers not to think at all >D
https://twitter.com/twitter/statuses/107...5751131136

Saludos Saluti Cordialement Cumprimentos MfG BR + + + + +
Luigi Vampa +
Free42 '<3' I + +
Find all posts by this user
Quote this message in a reply
12-30-2018, 11:31 AM
Post: #20
RE: The lack of handling root functions in hp prime
(12-30-2018 10:08 AM)John P Wrote:  Hello,
My two cents. I think that the era of calculators is over. What is the point of using them in this time and age? Yes, HP Prime is very capable calculating machine and nobody can deny that. It is also true that HP Prime has a lot of bugs and/or inconsistencies and this does not mean that it is the fault of prof. B.Parisse. In general it is amazing how much you can do with this calculator, but it is also true that it takes a lot of patience and perseverance to learn how to get the most out of it. What are graphing calculators for? The obvious and I think false answer is: to learn mathematics. I doubt that you can learn mathematics using for example calculator such as HP Prime. The HP Prime can only speed up your work and hopefully catch some mistakes in your work. The not so generally acknowledged fact is that mostly students are using graphing calcs. to 'cheet' if possible on exams. No graphing calc. will think for you,ever because if it could you become redundant and the calc. will not need you. Since, in many schools that care about real teaching you are not allowed to use the graphing calculators on exams so, the only role left to that calc is to help doing your studies and/or homework. My opinion is: why bother with that little and powerful 'toy' if you can get yourself a lot of real viewing estate with 13.3 in. convertible and install GIAC, XCAS, Mathematica, etc. or whatever your heart desires. This is what I will do in 2019, I am waiting for good and reasonably priced 13.3 in. convertible with the specs I want and I will buy it. What I write is in no way to disrespect and/or diminish the work prof. B. Parisse put in his project, but now it is the time to move on and why to doodle on that little screen if you can have 13.3 in. laptop. True, it takes more space and it does not fit in the pocket, but hey, who cares. The time will soon come that somebody will make real computer to put it in your pocket and the screen of your choosing you will be able to carry in your bag. Happy New Year 2019 for everybody.
John P.


I do have an ipad, a laptop and a phone of course, but I prefer to use a real calculator, my prime.
A calculator has hardware keys wich is more comfortable than typing on touchscreen or computer keyboard.
It starts faster, you can have your calculations immediately. You want to study something, you take your paper notebook, your books, your pencil and your calculator and here you go.
Also, your brain works better with sketches, handwriting, instead of typing everything on a computer.

Guy R. KOMAN, hp 50G, hp Prime Rev. C
Find all posts by this user
Quote this message in a reply
Post Reply 




User(s) browsing this thread: 1 Guest(s)