17 replies to this topic

[vanhoa]

After almost a day to try for many examples to compare between CP and TI, I see that we cant say which calculator model is the best: Ti is good at numeric operation and Classpad is good at symbolic operation, it is a big different. For example (Ti 92plus ver 2.09, CP ver 2.20):

numeric:
to solve x+y=4 and x^y=2 ofr x and y (numeric):
TI: x=3.4388033655143 and y=0.56119663448568 or x=1.2916635660656 and y=2.7083364339344
CP:{x+y=4,x^(y)=2}

to solve x^80-1 =0:
TI: 80 roots (in complex, use csolve() , but I have to wait so long )
CP: ERROR :Insufficient Memory

To solve |x|+|x+1|=1 (that mean "abs(x)+abs(x+1)=1"):
TI: x<=0 and x>=-1
CP: {x=0} (Why? )

And: factor(10^40+1):
TI: the result appear after over 4 hours ( I dont know exactly because I cant wait it).
CP: So long...

symbolic:
to calculate : 1^2-2^2+3^2-4^2+...+(-1)^(n+1)*n^2:
TI: SUM((-1)^(n+1)*n^2,n,1,n)
CP: ((-(2n^3+3n^2+n))/6)

And some which I cant remember.

And much more...

(Anyway, I hope OS 3.0 can make CP become the best )

[Khỉ con]

Hic
symbolic:
to calculate : 1^2-2^2+3^2-4^2+...+(-1)^(n+1)*n^2:
TI: SUM((-1)^(n+1)*n^2,n,1,n)
CP: ((-(2n^3+3n^2+n))/6)
Mathematica is an another answer...

[vanhoa]

I am sorry, there is a mistake when I copy the result, and now I cant go home totake the calc. So tried it again cp (o a or an emualator) it must got a right answer.

Sorry, it must be:
Cp: (((-1)^(n+1)*n^(2)+(-1)^(n+1)*n)/(2))

Because I input the wrong expression (Sum((-1)*x^2,x,n) it must be sum((-1)^(x+1)*x^2,x,1,n).

[The_AFX_Master]

(1)
to solve x+y=4 and x^y=2 ofr x and y (numeric):
TI: x=3.4388033655143 and y=0.56119663448568 or x=1.2916635660656 and y=2.7083364339344
CP:{x+y=4,x^(y)=2}

(2)

And: factor(10^40+1):
TI: the result appear after over 4 hours ( I dont know exactly because I cant wait it).
CP: So long...

(3)
to solve x^80-1 =0:
TI: 80 roots (in complex, use csolve() , but I have to wait so long )
CP: ERROR :Insufficient Memory

1.- Classpad O.S 2.2 DOES NOT solve systems of non linear equations, this is a known fact. don't waste time trying to solve it, and wait for the O.S 3.00 (or use LNA broyden's method instead, is a lot faster than any calc)

2,3.- A calc isn't designed to perform these annoying (and unusable) calculus ,even the CP manager still slooow in (2) , use mathcad for this funny tests

If you wanto to do a benchmark within the calcs, do real day to day problems, not funny and unusable math. suggestions:

RREF with a Hilbert 20x20 matrix
matrix inversion with the same 20x20 hilbert matrix
matrix product, 50x50 random
some trigonometric integrals
some -infinity, infinity integrals
some improper (complex plane) integrals. such as 1/(2+cos(x)) from 0 to 2 pi (AFX2.0+ does it wrong)
Graphing speed test (a good measure of the numeric speed of the calc)

[vanhoa]

Only Example Master

to who can: please change the link of LNA like the other.

[vanhoa]

In some way I see that AFX 2.0+ is better than CP
For example, to solve 3sin(x)^4+cos(x)^4, Afx give a result that is a polynomial degree 4 so we can solve it by hand. In this way, TI-CP cant compare with afx...

[The_AFX_Master]

There is no a "exact definition" polynomial as answer to 3(sin(x))^4+(cos(x)^)=0

my AFX ROM 1.05 gives me (pretty fast):

tan(x/2)^8 - 4tan(x/2)^6...... a loong result. i think that you can't solve it by hand easy, and you cant put it in the polynomial solver of the calc. The only way to solve it is using numeric solver.. and depending of the initial root guess, you can receive each different root of this "poly". Or using LNA with a metod that find all the roots on the given interval. in summary there isn't a result of the given eqn, is only a simplyfied one.

Classpad 300 O.S 2.20 gives me.. no answer . Same as CP Manager v2.0

I can understand whyCP can't solve this. The cas would be programmed for solutions only (x=-/-++-%*^&), no simplifications, remember that CP is a didactic focused device.

Good find Vanhoa, go ahead man!. Comparing AFX vs CP is a hard task because there is no absolute superiority on a side. The CP is stronger than AFX in things such as integrals and some limits. On numerical matrix algebra , at least O.S 2.2, AFX beats classpad, (that is if you don't use Lua,CPLua beats any calc on the world ) . But AFX CAS runs slower things as Eigenvalues and Eigenvectors analytically.

Summarized:
DiffEqns ---> AFX
Numeric Solve ----> AFX-CP . needs some serious benchmarks. CP advantages, Natural Disp and large variable names.
Poly Solve ---> I think that AFX culd be faster, but i'm not sure.
Matrix numeric: AFX
Matrix algebraic: CP of course

if you can do some benchmarking Vanhoa on all the subjects, we're very pleased to you

[vanhoa]

tan(x/2)^8 - 4tan(x/2)^6...... It is a true result, if we say a=tan(x/2) we have:
a^8-4a^6+... it is a poly degree 4 that we can solve by hand easily.

[Xtreme]

Anyway, I hope OS 3.0 can make CP become the best

It can't because the ClassPad doesn't come with OS 3.0. You have to buy a ClassPad and then you have to pay extra for OS 3.0.

Here is a good comparison where TI beats Casio hands down. TI updates their software and doesn't charge you for it. Casio fixes problems and then makes you pay for it. If you are comparing the cost of a TI and Casio add $50 to the Casio because you have to pay for updates. I think TI will be selling a lot of calculators and you will see a lot of ClassPads on eBay.

Does anyone know when the TI NSpire will be available? As soon as I have to open up my wallet and pay money for OS 3.0 there is no reason to keep my ClassPad. I might as well put the money towards an NSpire. At least I know TI will give free updates. Casio has made it obvious that they will NEVER make improvements unless I pay for them. To me that makes the ClassPad a dead end.

[2072]

when something is unfair, there is always a way to solve it using the internet and P2P networks...

[Orwell]

I'd like to insist on the fact that we don't really know if the OS update will be free or not. The new CP Manager isn't free, but there is no information about the simple OS update

[PAP]

when something is unfair, there is always a way to solve it using the internet and P2P networks...

Indeed .

I'd like to insist on the fact that we don't really know if the OS update will be free or not. The new CP Manager isn't free, but there is no information about the simple OS update

You are right, more confirmation is needed, but I don't think that they released OS 3.0 and Manager commercially, and they are planning to release OS 3.0 for free later, unless they read our posts.

[SoftCalc]

You are right, more confirmation is needed, but I don't think that they released OS 3.0 and Manager commercially, and they are planning to release OS 3.0 for free later, unless they read our posts.

Here's your confirmation.

[PAP]

Here's your confirmation.

Your previous post was a confirmation; now there is another one.
I have to say that I'm really upset. According to forum's rules shouting should be avoided, but this is an exceptional case, so I will shout again, and I'm sure that I'm not the only one who feels like this.
Every ClassPad owner should be able to download OS upgrades for free. Any upgrades, anytime. I'm upset because current OS needs a lot of things to be upgraded (every ClassPad owner knows what I mean very well).
There are several people here who are trying to make a better ClassPad, like Orwell, omegavirus, AFX_Master, unique33, Kilburn, like me, and others. We spend a lot of time to develop useful programs for ClassPad, and this is how we are awarded: We have to pay for an OS upgrade! I DON'T LIKE TO BE TREATED LIKE THIS!

[omegavirus]

Here's your confirmation.

I DON'T LIKE TO BE TREATED LIKE THIS!

That really makes CA$IO a bunch of... I don't even want to say anything to them because the are so meanless even for a bad word...

[vanhoa]

I mailed casio these topic line. Maybe they are reading.

[Xtreme]

I mailed casio these topic line. Maybe they are reading.

I don't think they read the forum. If they did they wouldn't make such a big mistake and charge for OS 3.0. Everyone's reaction is anger. Even a feeble minded person would see that this will kill the ClassPad. Shame on you Casio.

Maybe a TI spy is making these decisions in Casio. I am sure that TI is doing the happy dance right now. BTW, has anyone seen the TI NSpire? Where can I buy one?

I will probably keep my ClassPad for now because CPLua and LNA look very powerful. My guess is CPLua and LNA is probably better then all the improvements in OS 3.0 anyway.

[PAP]

I don't think they read the forum. If they did they wouldn't make such a big mistake and charge for OS 3.0. Everyone's reaction is anger. Even a feeble minded person would see that this will kill the ClassPad. Shame on you Casio. Maybe a TI spy is making these decisions in Casio. I am sure that TI is doing the happy dance right now.

What can I say, I totally agree.

My guess is CPLua and LNA is probably better then all the improvements in OS 3.0 anyway.

Concerning numerical calculations, I think that this is correct, judging from previous OS releases. However, CPLua and LNA will never improve the CAS itself. Only a new OS will improve the weak CAS; OS 3.0 is supposed to improve the CAS somehow (I don't know how much), but they charge for it.