36 replies to this topic

[vanhoa]

Look at http://www.tamuk.edu/math/mathclub/. Thease function is not so difficult, I think we can translate them to CP language, I'm tring to translate them now.

[The_AFX_Master]

Some functions are embedded on the calcs CAS, and the engineering zone is ridiculous. I think that itsn't a big deal

[vanhoa]

Demo version: http://www.casiocalc...amp;load_that=1.

[vanhoa]

Ver 1.1 with some new changes.

Link: http://www.casiocalc...amp;load_that=1.

[vanhoa]

Some more new changes in ver 1.2.

Link: http://www.casiocalc...amp;load_that=1.

Try it, give me your suggestion. I dont think there is any bugs.

[vanhoa]

Ver 1.3.
Link: http://www.casiocalc...amp;load_that=1.
No one post here?

[Behnoud]

What is the difference?
What is added?
Why you don't creat a new FULL describtion for your new version?

[vanhoa]

Look at "readme.txt".

Right link: http://www.casiocalc...amp;load_that=1.

[vanhoa]

Download v 1.4!
Link: http://www.casiocalc...sic#noname-v1-0.
I need a name for it! I need to know if there is any bugs! I need suggestion!

[vanhoa]

A problem: const name was changed.

[vanhoa]

No body else here? (

[vanhoa]

This is unuseful???

[PAP]

This is unuseful???

I didn't wanted to make any remark, but, well, you asked for it several times . No, your "NoName" library is not unuseful. However, I'm afraid that it has very limited use:
(1) First of all, your library is written in CPBasic; need I to say more about that? it is slow, and has all CPBasic's serious limitations. For example, Fibonacci numbers can be easily computed by a recursive function, but you can't do that in CPBasic, since it has primitive function support; this means that your fib function is a rather poor implementation.
(2) There are only a few functions, which are not "complete", in the sense they don't completely solve each problem; for example, your linear interpolation function (interp) solves a very basic problem: only two points are used, so you can't create an interpolating function based on many points; furthermore, you can interpolate or extrapolate, but you can't compute the derivative. Another example: you have 4 functions for solving polynomial equations: quad, cubic, deg3, deg4, although one more general function would be enough. But, even then, your hypothetical "general" function will be able to solve polynomial equations (a rather trivial problem). What about non-polynomial equations?
(3) There are functions which are not needed at all, such as nroot: you don't need that function, since you can easily replace it by the CAS command fMin(abs(f)). Why a user will prefer to use your nroot function instead?

I know that I didn't said anything good about your library, but please don't think that I want to be offensive. As a new CP owner, you have done several things already; it seems that you are a talented programmer. I just want to say that you are in a hurry to create something very useful, but I think that it is too early for that. Nevertheless, it was a nice try. Afterall, you are practicing, and practice makes perfect. Keep up your work, and you will become a powerful programmer; just don't expect to create a great program so easily.

[Kilburn]

What else can I tell you?

Start learning C++ and build REAL nice applications with CPSDK.

[vanhoa]

CPSDK IS NOT SO DIFFICULT BUT WHY I DI NOT USE IT, ONLY BECAUSE IT CAN NOT EMBED INTO THE MAIN APPLICATION - THE MOST IMPORTANT THING OF CLASSPAD!

CLASSPAD WAS MAKE FOR MATH, GAME IS NOT ALL! CP SDK IS GOOD FOR GAMES, NOT MATH, I DONT THINK CASIO USE CPSDK TO MAKE THE OS! IF I KOW WHAT LANGUAGE THEY USE I CAN MAKE ST BETTER! THERE ARE SOME PUBLIc CAS THAT USEFUL FOR ME TO BUILD THE NEW CAS BUT I CAN MAKE IT WITH CPSDK, LUA, OR BASIC, WHEN CPSDK BETTER?

TISDK CAN MAKE FUNCTIONS THAT CAN USE IN THE MAIN APP!

[Kilburn]

Wow wow wow, calm down. I see you don't know CPSDK at all, there is class called CPExpression that allows you to use CAS functions, exactly like in the Main application!

[PAP]

IF I KOW WHAT LANGUAGE THEY USE I CAN MAKE ST BETTER! THERE ARE SOME PUBLIc CAS THAT USEFUL FOR ME TO BUILD THE NEW CAS BUT I CAN MAKE IT WITH CPSDK, LUA, OR BASIC, WHEN CPSDK BETTER?

It seems that you still want to "build a new CAS" - which means that you didn't checked the huge codes included in open-source CAS like Axiom or Maxima, which are developed by many mathematicians for several years. Anyway, it's not my business; you think you can make a CAS? ok, stop shouting and go on, do it. Just keep in mind that capital letters are considered as shouting in forums.

[vanhoa]

Killburn: CPExpression canot be update, we cant update it to use it, I need a new parse, but if you are very good at SDK, why dont you make a new main? Now I am boring with upgrading the CAS , I'm making a real 3d game, I know how to make gray scale, so wait .

PAP:"many mathematicians for several years". That's true, but they developed, and I copy and translate! Sorry, I nearly finished it, but I have some more few problem .

[The_AFX_Master]

PAP:"many mathematicians for several years". That's true, but they developed, and I copy and translate! Sorry, I nearly finished it, but I have some more few problem

i think that you're underestimating the problem... "I copy and translate", come on.. you think that's so easy?. see the Orwell's LUA.. Lua already exists, Orwell is porting it to the CP, easy task?. hard task.. ask Orwell. . Think before shout

[vanhoa]

Difficult? Not at all! Orwell's job and my job are different! We didnt make the same thing!

[Orwell]

Now you will definitely calm down. You're really noisy don't you know?

The fact is, that you have absolutely no idea of what your are talking about.
You seem to have some knowledge as programmer already, but hey, aren't you only 14?
There are many things you just don't realize, a lot of concepts that you never approched, a lot of stuff you'll learn later in your studies (like PAP said e.g., there are some other things in math than polynomials...).

PAP already said it too: several experimented programmers and mathematicians need to work during several years to create a decent CAS system. And you, with your own experience, and "copy-paste" skills, you want to do a better CAS than Casio's engineers? Stop dreaming, learn what "programming" really is, and think about it again later.

One of your main concern just now is that we cannot call an Add-In from the Main Application. That's right, they just not allowed this in the CP OS. It's true that it's a pity, but they certainly had some reasons for that. This is not a reason for not using the SDK btw. And do no think this SDK is only good for games : BitEdit, SpreadSheet, CPLua and all other existing Add-In were all written with it. These have nothing to do with games.

And about this: "I DONT THINK CASIO USE CPSDK TO MAKE THE OS! IF I KOW WHAT LANGUAGE THEY USE I CAN MAKE ST BETTER!"
... No, you can't. You don't have the skills at all.
Or, if you think you are able to do it though, I can tell you right now that the CAS is written in C. You can write applications in C with the SDK. Go ahead and try.

Now about AFX_Master's remark: he didn't say we are doing the same thing; he said that CPLua also consists of porting an existing application to the CP (instead of starting it from scratch). It's been more than one year now that I started the project, and it's not finished yet. And I think I can say that I have a little more experience than you in this job

You have some good skills for your age, okay. But stop thinking that you can do everything. If it was as easy as you think, Casio would hire students from college and not engineers

[vanhoa]

Ok, Im 14 but Im really good at Math I think (my Math test is allways 10 mark). I know Im make noise here. But, the new CAS's public day is not far from now. SO WAIT, I AM WORKING (thanks PAP for the link).

Sorry to everyone, but I know what I am doing. Why no one believe me?!?

One more time I am sorry, the new CAS is short of:

1./. I cant make a nature display.
2./. I cant link to any where.

Althought I have many other work to do but I believe the new will come soon .

Btw I need some great web skins, help me for it!

The last time, I want to say I am sorry.

[Orwell]

Ok, Im 14 but Im really good at Math I think (my Math test is allways 10 mark).

Sure, perhaps you are really good at school with the things you learned in mathematics this far. And you probably have a very poor (or totally inexistant) knowledge of the topics in math than mathematicians and engineers use everyday, which are differenciation, integration, matrix manipulations, eigen values, complex numbers, vectorial spaces, Taylor series, Fourier & Laplace transforms, (partial) differential equations, etc etc. I'm just giving here the most basic chapters, PAP can complete the list if he wants to. These are the kind of things a good CAS should provide.

Sorry to everyone, but I know what I am doing. Why no one believe me?!?

Because of you saying "it's not difficult", "it should be easy", "someone just has to ...", "juste copy-paste and adapt"... Those are words coming from someone who really does not know what he says - and from a kid, also.

the new CAS is short of:
1./. I cant make a nature display.
2./. I cant link to any where.

These things concern a "Main Application" which could be use for interaction with the CAS, not the CAS itself...
And I'm really curious to see what kind of "link" you would be able to set up and how you will do it

[vanhoa]

"And I'm really curious to see what kind of "link" you would be able to set up and how you will do it ".
Link here is call to other app.
But why no one like basic, the functions writed in basic can be used in "Main" app. I am still upgrade Noname, do you think the "cons" and "conv" good?

[vanhoa]

Ok, every one, I dont now anything about daily-use math, I cant do evarything,... just because im a kid.
So ok, im a kid, a kid cant do any thing, a kid must learn a lot, a kid have no skill... ok, the cas wont be public anyway, but I made the source myself, not copy-pase, with some good function that take me for over a week: calulate the result to any digits, upgrade the Taylor, upgrade the solve, upgrade the factor and upgrade matrix. I will send the source to Casio to have the OS 3.0, may be they will use it in future ver of Classpad.
Why did I try to creat every thing that I dont use at the moment? Because I want to say CP is better than TI, althought I own a TI-92+, I never use it for math, only for games. I make all for a more useful and powerful classpad,and I dont want to make noise to any one.

[PAP]

Ok, that's more than enough. I want to say something about it.

The fact is, that you have absolutely no idea of what your are talking about
....
And about this: "I DONT THINK CASIO USE CPSDK TO MAKE THE OS! IF I KOW WHAT LANGUAGE THEY USE I CAN MAKE ST BETTER!"
... No, you can't. You don't have the skills at all.

I agree.

And you probably have a very poor (or totally inexistant) knowledge of the topics in math than mathematicians and engineers use everyday, which are deritatives, integration, matrix manipulations, eigen values, complex numbers, vectorial spaces, Taylor series, Fourier & Laplace transforms, (partial) differential equations, etc etc. I'm just giving here the most basic chapters, PAP can complete the list if he wants to. These are the kind of things a good CAS should provide.

Several topics can be added to Orwell's list, but i don't think there is any reason to do it. Mathematics is full of things which a 14-year kid cannot even imagine.

But why no one like basic, the functions writed in basic can be used in "Main" app.

Because CPBasic is slow, slow, slow, and poor, poor, poor. For your skills, however, it's a good start. You will realize why everbody here doesn't like CPBasic later, when (and if) you will upgrade your programming skills.

I am still upgrade Noname, do you think the "cons" and "conv" good?

No.

ok, the cas wont be public anyway,

Oh, no! Please, don't do that to us! Make your "CAS" available to the public... everybody needs it!

but I made the source myself, not copy-pase, with some good function that take me for over a week: calulate the result to any digits, upgrade the Taylor, upgrade the solve, upgrade the factor and upgrade matrix. I will send the source to Casio to have the OS 3.0, may be they will use it in future ver of Classpad.

I have a feeling that we have just discovered Filip No 2.

[Orwell]

I will send the source to Casio to have the OS 3.0, may be they will use it in future ver of Classpad.

Hey, I want OS 3.0 too!

[Kilburn]

"So ok, im a kid, a kid cant do any thing, a kid must learn a lot, a kid have no skill"

I'm 14 too Of course, you have skill but not enough to re-build a whole CAS.Try to imagine how difficult it is to solve equations, differentiate, etc... You should learn very advanced mathematics, numerical analysis (like PAP)...

[The_AFX_Master]

I have a feeling that we have just discovered Filip No 2.

M E G A L O L!!!!! .. sure PAP

@ Vanhoa:

hey man, calm down.. your'e going so fast. Linking the main isn't a big deal. I explain: basic maths would be covered easy on the CP CAS. recoding the CAS isn't easy, isn't a game. An idea. forget the CAS and start to learn more advanced maths. then you will reach a point in that surely you can't solve the way of how to code a determinated CAS procedure. The problem is that the basic maths that you know is "linear". You input the data, perform the operations, and you get the result, without the need of discovering "easter eggs" on the problem.

an example, how do you integrate a polynomial, good is a formula int(x^n)=x^(n+1)/(n+1) int(x)=x^2/2. now how you integrate sin(x)cos(x)(e^2x).. problem! your algorithm doesnt work, why? because there are a lot of ways to solve the same problem, and a computer program needs VERRRY LONG AND HARD task trees to perform these "thinking".. how do you say to the machine that 3x+4x is 7x?.. you're an ++ kid, use these abilities and learn more. then make your first attempts


BTW, we have some weeks flaming casio for the O.S 3.00.. but, where are the O.S?

[vanhoa]

They sell it, but our contry didnt buy it to sell us.

[vanhoa]

an example, how do you integrate a polynomial, good is a formula int(x^n)=x^(n+1)/(n+1) int(x)=x^2/2. now how you integrate sin(x)cos(x)(e^2x).. problem! your algorithm doesnt work, why? because there are a lot of ways to solve the same problem, and a computer program needs VERRRY LONG AND HARD task trees to perform these "thinking".. how do you say to the machine that 3x+4x is 7x?...

Integrate is a little more difficult then diff because sometime there is no solution. But it's simplify, we need to creat some power obj., mutipli object and some addition object, then expand the expression and rewrite the exp. in some object, so we can work easily in these obj. for example: int(2(x^3+x^4),x) it will simply the expresstion to 4 part: 2,x,3,2,x,4 then it solve, int(sin(x+2)^2,x) first it will store x+2 in a obj., do simplify by finding match bracket example:

-input exp=int(sin(x+2)^2,x)
-parse the exp:
+first rewrite exp=(int(sin(x+2)^2,x))
+repeat:
*read the nth char from the left, continue if it is a var-char.
+now it have "int(", we must find the match close bracket:
*b=1,ch=""
*repeat until b=0
/p=p+1
/if mid(exp,p,1)="(" then b=b+1
/ch=ch &mid(exp,p,1)
/if mid(exp,p,1)=")" then b=b-1
*now we have x+2, store it in an object and define it with 'sin'
*continue parse, we have ^2, store 2 in a power object, define with sin(x+2)
*do int (sin(x+2))^2 : ((-(cos(2x+4)-1))/(2)) or methol:cos(a)*cos(b )=-(cos(a+b )-cos(a-b ))/2 with a=b, for example, we expand sin(x)^9 we must use that methol 8 times.
*expand ((-(cos(2x+4)-1))/(2)) = ((-cos(2x+4))/(2))+((1)/(2)) by division each sum obiect to the denominator (use finding match bracket methol after "/" symbol)
*now we have 2 sum obj to solve, we must int them in order.
+...

So long for a short exp, but Casio use that methol, TI and HP doesnt use it, a methol sometimes is good and sometime is bad.

[Orwell]

Integrate is a little more difficult then diff because sometime there is no solution.

Not at all.
Like AFX_Master said, Integrate is a lot more difficult, because you have several possible paths to lead to the solution (if it exists). With differenciation you have only one path, and you keep following it. Integration consists of re-finding this path in reverse order, and you must find the good way

This is a very basic method indeed (and I'd like to know how you could say that Casio does use it and not TI and HP). Things should be a little more complicated though

But actually we are more interested in the way you will execute the parsed expression, and choose the correct formula to have a good result.

Take the previous example: how will your program know that (3x+4x) = 7x ?

Another example: you said

do int (sin(x+2))^2 : ((-(cos(2x+4)-1))/(2)) or methol:cos(a)*cos(b )=-(cos(a+b )-cos(a-b ))/2 with a=b, for example, we expand sin(x)^9 we must use that methol 8 times.

What would you do in this case: int( sin(x)^9 * cos(x) , x) ?

What happens with int( sin(x)^9 * cos(x)^2 , x) now?

[vanhoa]

3x+4x=7x is easy. We must creat some product-object and a sum object.

int( sin(x)^9 * cos(x)^2 , x) is a little difficult more now, but it's still not difficult at all. I dont want to write the code here because i dont think you cant understand the CAS source. But everything I want to say now is build the CAS not difficult at the parse, the most difficult is methol, not as before I think build the CAS is easy. So Now I know, I must learn a lot . Althought I finished my project and sent it to CASIO, they said it is not good at all but it is very useful to improve the function, so the we can make function that can use in the main app so I want to say let's us wait for the next OS version but Casio wont give me the OS, they said it cant be compare with their OS ! Why some ones can own the OS without any thing but I cant own it althought I have done every thing. I need the OS, they dont sell it in my contry...

Everybody here, please stop talking about my ability to make CAS. I dont want to discuss any more, ok, you won now!




//Edit

I think someone should make a library to calulate the result of an inputed expression to any decimal number. For example 1/3=0.333333333333333333333333333333333333 with n letter "3", that would be useful. I can make it with only 4 basic function: +,-,*,/ no more and no less. PAP, The_AFX_Master, and maybe orwell or any one interested, START A LIBRARY TO BEAT TI.

[vanhoa]

@Kilburn: good news for me, you are 14.

[Orwell]

int( sin(x)^9 * cos(x)^2 , x) is a little difficult more now, but it's still not difficult at all.

Are you even able to solve it by hand?
And did you notice the difference with my second example? Did you try to calculate these examples on your CP? Do you know why the third example is a lot more difficult than the second one, altough there is just one "exp-object" more?

I dont want to write the code here because i dont think you cant understand the CAS source.

Huh... You don't want to write the code because you don't think we can't understand. Sorry I don't really get what you mean here

Casio wont give me the OS, they said it cant be compare with their OS !

Oh, really?

I think someone should make a library to calulate the result of an inputed expression to any decimal number.

For the last time, forget CPBasic. It's only advantage is indeed that you can call functions from the Main Application. For everything else, it just sucks. You won't beat anyone nor anything with it.

[vanhoa]

Casio said they make a big change in the OS 3.0 Cas and if I want to own it i have to buy it

[vanhoa]

Hey, I found this with screenshots