77 replies to this topic

[kucalc]

Symbolix CAS for fx-9860G/fx-9860SD/fx-9860Slim/Graph85/Graph85SD
New Version!!! - v0.4

It doesn't slice and dice, rather... It factors and expands! A valuable tool for highschool mathematic students. Symbolix provides the fx-9860 user with algebraic and symbolic manipulation tools at their fingertips. It's still in the early stages of development, but more to come soon!

* Simplification
* Polynomial expansion
* Polynomial factorization (rational factors)
* Factor quadratics (using quadratic equation)
* Collect coefficient
* Substitution
* Common factoring
* Differentiation



Download it now! (User Manual included): http://www.casiocalc...amp;load_that=1

[kucalc]

SYMBOLIX v0.4 is coming soon! Heavily improved with much more features.

Here's what's going to be in v0.4:
* Simplification
* Differentiation
* Collect coefficient
* Polynomial division
* Cubic solver (exact value)
* Numerous trigonometric and utility functions
- Functions can be nested

[DarkDead]

Wow! Thank you kucalc, I was waiting for this since you released that first version of your CAS. A lot of people told me that a CAS would really help in my exams, lets see . Now I'll try to figure out how it works.
The download link is broken, can you fix it?
Keep the good work.

[kucalc]

I've released v0.4 Beta of SYMBOLIX CAS. Check out the file sharing to download it.

Remember, it's still Beta, so I know there are still some problems with it. The differentiator doesn't support inverse trigonometric functions. I'm considering removing cubic and replacing it maybe with a command for integration or limits. cubic is just too memory intensive. Well, see... The final release of v0.4 should contain the fixes. But the Beta, should give you a feel of how progress is coming along for SYMBOLIX.

You'll see that v0.4 is a big improvement over v0.3. For example, try to find the derivative of (x^3+5x^2+2)(3x-10) with SYMBOLIX and you'll see what I mean.

For v0.5, do you guys like 3D?

[kucalc]

Hello Tom, thanks for trying out SYMBOLIX.

If you read the manual, commands don't require a closing parentheses. This helps SYMBOLIX distinguish commands from functions.

So polydiv actually does work:

polydiv(x^2-y^2,x-y
>x+1

The factoring seems to be my error, not SYMBOLIX's. As a last minute change before releasing the beta, double negatives "--" were suppose to turn into positives. Instead, I had it programmed to trim away a negative sign, leaving one negative sign. That'll be easily fixed in the final release. As for prime, it's intended to prime factorize an integer, not an expression. The expression could be easily simplified into an integer before using prime just by typing 2^2*4^4 getting you 1024.

[barnaby]

Some more tests. I'm using an fx9860g-SLIM, therefore running 1.10:

cubic(x^3-x

System ERROR
REBOOT: [EXIT]
INITIALIZE: [EXE]
TLB ERROR!!
TARGET = 080FFF9E
PC = 88023448


expand((x-2)(x+2)

x^2-4


expand(x(x+1)(x-1)

x(x+1)*x-x(x+1)


factor(x^2+x-2

(x-1)(x+2)


factor(x^3-3x^2-6x-8

(x-1)(x--2)(x-4)


quad(x^2-4

(x+((16^1/2))/(2))(x+(-1(16^1/2))/(2))


polydiv(x^3-3x^2-6x+8,x+2

x^2-5x+4


diff(8x^3+36x^2+54x+27

24x^2+72x+54


diff((2x+3)^3

(2x+3)^3


diff((2x+3)(2x+3)(2x+3)

16x^2+48x+2(2x+3)^2+36


diff((2x+3)^2

(2x+3)^2


diff((2x+3)(2x+3)

8x+12


diff(sin(2x)

2cos(2x)


diff(cos(x)

-sin(x)


diff(tan(2x)

(2)/((cos(2x)^2))

Overall I am extremely impressed with SYMBOLIX - there are just a few issues that require 'ironing out' - which is completely to be expected. Perhaps integration would be a nice feature...

[kucalc]

Thanks barnaby for your tests. It's nice to see tests of Symbolix running on the fx-9860Slim.

I've released a new version contain bug fixes and new features. You'll find a new version in the FileShare.

Added:
+ Common factoring
+ Substitution
+ Support for inverse trigonometric functions in the differentiator
+ Simplify radicals (360^0.5 -> 6*10^0.5)

Removed:
- Cubic factorization

New features being planned for v0.5:
* 3D graphing
* Integration
* GCD and LCM


PS: I've renamed SYMBOLIX to just Symbolix. I've been getting emails about what the letters in SYMBOLIX stand for. They don't stand for anything. I just thought it looked cool with the letters uppercase, but to prevent confusion it's just now Symbolix.

[Fez]

*kneel and bow*

Thank you XD

[uvoK]

Just a question:
Is it me or why do I write on the first line when I have done sth.`?

I have to delete the whole 1st line to do sth else.

[kucalc]

Umm yeah, the CAS on the Algebra FX works like that.

The first line is where you enter your expression. The line can easily be cleared completely by pressing AC/ON.

[uvoK]

I just read the Readme and wonder what the difference between ccoef and comfct is?

[kucalc]

I just read the Readme and wonder what the difference between ccoef and comfct is?

I don't know how else to explain it, the User's Manual does a good job in explaining it.

@macdenaif: Obviously, it is a bug. BTW, please don't post in such a large font. It makes people less inclined to help you.

[kucalc]

Hello rakiem, it seems that that addin got corrupted during the file transfer. What I suggest you do is clean out the STORAGE memory and Optimize it. Then download Symbolix into the calculator again.

Another thing I ask to you is about programming CAS with lisp. I'd like to make some program with symbolic elements in C, but i don't find a good program to translate from lisp to c code.

I don't think I understand what you are completely saying. You want to translate some LISP code into C, right?

[kucalc]

Ok, so you got the addin to run successfully now?

Yeah, you can't use Visual C++ libraries with the CASIO fx-9860G SDK. Some matrix operations in Symbolix isn't such a bad idea...

[kucalc]

You just have to write them yourself, without the use of using other libraries. Everything needed to set up a symbolic system you'll have to write yourself.

However, if you want a program that symbolically find the determinant of a matrix, I can add it as a new feature for Symbolix.

[tuxianer]

Great program! But i have detected some errors:

diff((2x+3)^5)
BhAH

diff(1/(10x - 3)^2)
System Error

expand((a+ B) ^2
(a+ B) ^2


when i close the term by expand with a ")" i get:

Error; Unexpected ) or ,

[ben_i]

Hey kucalc , fantastic work Your program was the first I put on my new 9860g . Jus wondering if you will do a new release soon , maybe add partial differentiation and fix some bugs like dy/dx(x^1/3) which crashes . Cheers

[kucalc]

Thanks for using Symbolix guys!

Unfortunately, Symbolix is no longer being developed for the fx-9860G platform. The fx-9860G is just too crappy for Symbolix to run in. Yeah, I know it could be more worse (imagine Symbolix on a TI-84!!! ), but... Symbolix is memory hungry and the fx-9860G just can't provide that. When the heap memory gets full, the fx-9860 just fails and crashes.

You may Symbolix again in the future, but it won't run on the fx-9860G, instead on a better platform.

[polo961]

did this mean that you will not fix the bugs and develop it again?
i hope the answer is no

[kucalc]

The bugs will be fixed and Symbolix will still be developed, but it won't run on the fx-9860G.

[polo961]

ok but for us the users of fx 9860G SD
we have no problem in memory we can add as much as we can(the sd card)
so can you please please please even fix the bugs of the version of 9860 please
it's a great utility and one of the reasons why i bought the 9860G SD was that program so please develop it or even fix the bugs please

[kucalc]

I might release one version (with 3D graphing, matrix operations and integration and some bug fixes), but that will be the last for the fx-9860 series.

[LordNPS]

symbolix is being abondoned for the 9860 yeah?

Bloody hell, how's it that CASIO's CAS manages to run in an FX1 then? Magic is it?
I don't know (and probably won't) exactly which algorithms for parsers, lexical Math analysis and actual CAS are you using, but without further nicety I can say that they are plain inefficient, perhaps you ought to change them.

I find it sad to see you so frustrated with 9860 kucalc in a short period, I've seen you more and more irritated with the machine, what's going on?

[kucalc]

I'm not frustrated with the fx-9860G. I just don't find that the fx-9860 satisfies the requirements. I don't know if it's me, but the fx-9860 memory management system sucks. In order to prevent Symbolix from overflooding the heap, I tell it to free the memory no longer in use, but instead it sometimes frees the memory currently being used causing the system to crash faster than if I was to let it overflood the heap.

[The_AFX_Master]

@ LordNps

I think AFX's had 144 Kb on user ram (and the RAM chip is a 256K i think).
years ago when i was a "9860 is a crap" supporter i saw that many people said that the 9860 system RAM is almost 512Kb based on the chip specs. Is that true??.
AFX was programmed at the core on the good ol' days using weird things as assembler, for sure those guys did make a memory management system that fit their needs, as when you do grayscales and such.

I think that the 9860 was a step backwards on the casio line, isn't it supposed to be better than the AFX series? (in all aspects, not only far greater speed).
If they wanted a calculator for the school, an upgraded 9850 series is far enough to do the job. in fact, i think that more than a 570 ES on school is too much horsepower.
So they put a SD slot ("pro" feature) on a CAS-less and RAM-less calc without doing a newly revamped OS (as Classpad OS is)

Casio must release a "keyed" version of the classpad with a high res screen, Classpad's O.S and more RAM. I see the CP as the most powerful calc to date. And it will be at least when Nspires get serious programming abilities (i don't know if that's already done).

"offtopicking" end.
PS: sorry for the new nonexistant words

[polo961]

I might release one version (with 3D graphing, matrix operations and integration and some bug fixes), but that will be the last for the fx-9860 series.

yes please please i and all this casio model owners will be very very happy and thankfull to you
kucalc you are the best
but on which plateform are you gona develop it if it's not on the 9860 gsd?

[kucalc]

years ago when i was a "9860 is a crap" supporter i saw that many people said that the 9860 system RAM is almost 512Kb based on the chip specs. Is that true??.

It is true, but the RAM is used for program execution/addins, etc. Not for storing BASIC progs and that stuff.

I think that the 9860 was a step backwards on the casio line, isn't it supposed to be better than the AFX series? (in all aspects, not only far greater speed).

Each TI/HP/CASIO/SHARP calculator has their pros and cons. fx-9860G is better than the AFX at many things like USB port (a really big feature that the AFX doesn't have, you think I'm going to pay $99 for a cable...), speadsheets, more flash mem, etc. TI-Nspire has limited programming capabilities (however, their latest OS update has a program editor with syntax highlighting and new powerful programming functions). HP-49G+ just sucked with it's low quality hardware (keys get stuck or broken). And as for the ClassPad 300, I know lot's of people a stylus based calculator is a major turn off. Stylus based calculators are not allowed on some major exams just in the USA alone.

but on which plateform are you gona develop it if it's not on the 9860 gsd?

Well first of all, I'm not developing it anymore. It's been taken over and it's out of my control now. They've made a lot of modifications to it and I didn't make a backup of the code before I gave it to them, so most of the Symbolix code has been modified and expanded to accommodate a different system and will be hard for me to "port" them back to the fx-9860G platform. Also, I can't tell you what platform they're developing on. If I do I have to erase your memory. Seriously. You won't like that and since I can't do that, so I just can't tell you.

[ben_i]

Are you saying someone hired you from Casio to develop and extend the Symbolix under their name ? Or maybe they want it stopped and they are paying you money to cease development . I am confused....

[Kaos]

Thank you for this effort kucalc.

Well, now I am at the point where a CAS just doesn't help me that much (except for making some tedious work just a little bit shorter) but it was for the best.

But still, I appreciate the effort of making the fx-9860 that more useful to people around the world.

Because as you said....

Symbolix provides you the tools to accomplish your tasks. Don?t always trust Symbolix to give you the right answers .Always check them

[kucalc]

Me and my friends are in no way connected to CASIO. CASIO might however consider paying us to stop what we are doing, once they learned what were we doing. That's why we are trying to keep this project a secret.

@Kaos: Yeah, thanks. I don't know of any other non-CAS calculator that has a CAS as powerful as Symbolix. I guess that's a good thing.

[polo961]

Me and my friends are in no way connected to CASIO. CASIO might however consider paying us to stop what we are doing, once they learned what were we doing. That's why we are trying to keep this project a secret.

@Kaos: Yeah, thanks. I don't know of any other non-CAS calculator that has a CAS as powerful as Symbolix. I guess that's a good thing.

are you really gona release a new version of symbolix with bug fixes for the fx 9860 gsd
please please say yes please

[peenee]

Wait, what exactly is happening with Symbolix? Sorry for the stupid question...I've been living under a rock for the last couple of months-ish