Please don't feel too bad about it, you do absolutely great work.

If anyone is to blame, it's Casio, and I have a feeling I was either getting "repaired" versions of my old calculator, or calculators from a bad batch because I was using different versions of Utilities each time, and I find it pretty strange that they all broke within 5 hours, as opposed to my previous 2 prizm's which each broke after 8 months to 2 years.

You replied to the wrong thread?...

wow, that was strange... I quoted the utilities thread o.o

Is session saving and restoring (i.e., saving the full Eigenmath state including variables, console and command history when exiting, and restoring it all seamlessly when reopening) something worthy of an extra five or six seconds* of waiting time when closing Eigenmath, plus an additional third of a second when opening it?

* at most, one second at minimum; average time: three seconds.

Why would it take so long? I hope this does not mean writing to storage memory every time I exit. I'd imagine that this would be possible with main memory.

I don't want to use Main Memory because there's no alternative to it for most built-in applications (Run-Mat, Stat, Graphs, Spreadsheets, Programs, etc.) so I like to leave as much Main Memory available. It's quite a few bytes of data - never less than 6KB, usually 11 KB or more - and if I used the main memory for it, then it wouldn't work on quite a few calculators (many people I know have less than 2 KB of main memory available, due to having to store many programs, for example).

That would be great, but I imagine some people won't love it; perheaps a kind of option, or even better (in my opinion) a command set like : save state / restore last saved state would be great, but this is perheaps a lot of work, I don't know... In any case, I like the idea.

I agree with Nemhardy it should be optional especially if storage memory is used. I do not think making it optional would add much work.

I need to add the settings framework, in that case. I was already planning it, anyway.

By the way:
Do these roots look nice?


PS: before anyone asks, yes, I know my blog, tny.im and the Prizm software downloads are down (18+ hours already). It's an issue with my server provider, BlueVM. Looks like their support doesn't quite work on Sundays, at least for KVM instances in their Buffalo, NY location...

It does look nice. It reminds of me the Run Matrix. I presume that is what you are going for.

And this, does it look nice?

It's about the best one can do by hacking a 32x10 console into displaying graphics...

That looks very nice !

It looks very nice !!! I love the pretty print !!
It would be a good idea if we could directly enter the expression in the pretty print form !!!! So that when we press the root sign we don't get a sqrt() but a root, exactly like in the RUN MATH application.
What you did is just beautiful !!

Hello Everybody

I'm Ronald G V A I'm new here. First of all, thank you so much gbl08ma, I love your work. Since 2013 I've been following your add-ins for Our Lovely Casio PRIZM!
Please let me share some useful ideas to improve Utilities & Eigenmath ... I really love Eigenmath, one crazy idea would be to implement it in the Run-Matrix, I don't know if it's impossible

Well, the ideal wishlist would be:
*Implement Solving Differential Equations with Laplace or whatever
*Implement Fourier & Laplace Transformations
*Fix the bug that You can't Integrate sin(x)^5 or cos(x)^5
*Add Equation System Solving with 3 variables
*Add Dynamic Matrix like Run-Matrix
*Continue adding the "save history" & "Load history" but it would be better that make it like an option

If other ideas come to my mind, I will write them here jaja Thanks a lot

Thanks for the support!

@jpto: pretty-printing the command input is a lot of work, it's not supported by the original Eigenmath anywhere, it would need to be coded from scratch and probably be a bit slow. Probably there's a way to use the existing OS routines for math input, but these routines operate on special tokens for operators, not simple ASCII strings like Eigenmath - one would need to code a translator, which would not be exactly easy since some syntax is different. Also, I'm not sure the OS syscalls cope well with arbitrary strings for the Eigenmath commands and custom functions.


Yes it's crazy, and it's impossible. The OS doesn't offer hooks like Z80-powered TI calculators do, everything done in this regard would be very hackish and probably kill lots of calculators in the development and testing process, without ever becoming stable enough for daily use. Even if it can eventually be done, I'm not the one to do it, unless a safe hooks system is found (I doubt it exists) with hooks to Run-Mat.


Not supported by Eigenmath;


Not supported by Eigenmath;


Apparently Eigenmath can't integrate sin(x)^n or cos(x)^n, when n is equal to or above 5. In other words, not my problem;


Already available in the OS's "Equation" menu;


By this I understand you mean a matrix editor, well, while not impossible it would take a lot of work which I'm not sure is worth it.


Yes, this is planned.

Hello !
Ronald G V A (MD-1032) you have some very good ideas !!
Its okay for the pretty print, I understand and it will be too much work for you.
Is there a way to implement new fonctionnalities in eigenmath like Fourier and Laplace transformations ?
Can you add a screenshot of the limit solver ?
Thanks a lot

There is a way, but I don't understand most of the Eigenmath code nor enough math to do it.


The feature will probably be removed because it doesn't work right, but here it goes anyway:
http://tny.im/brokenLimitSolver
The first limit it couldn't even solve - possibly because a two-sided limit doesn't exist, but if you try to solve abs(x/(1-x)) for x=0 which is a two-sided limit, it gives the same error. The two notable limits that follow were solved correctly, but then it "solved" the fourth limit returning a value that is not correct for any of the sides...

What type of codes do you need to add in orther to implement new fonctionnalities ?? Do you need to write it yourself (fair too complicated) or is it possible to find some codes parts on the net ?
It is sad that you remove the limit solver (I think it is one of the most important thing to have in a CAS) but you're right, it doesn't work correctly ...

Some code can be added from SMIB (a Eigenmath fork), but not all, because SMIB requirements are much higher than those of Eigenmath. If I were to add everything SMIB has the Eigenmath binary would be giant and the default symbols alone (function names, etc.) would not fit in the RAM. Also SMIB was forked from an older code tree and many things need work to be compatible. The limit solving function doesn't work well on SMIB either, so it's not a problem in my job of porting it to Eigenmath.

Ok gbl08ma Thanks for your answer, don't worry.
Anyway I will give you some ideas that come to my mind

Jpto ! You're welcome, yes I wish Laplace & Fourier would be in the Eigenmath That would be awesome !

Oh I understand that the SMIB code requires a lot of RAM, that it sad because CASIO did not creat the PRIZM with Strong RAM nor Huge Storage But you can partially add the functions that eigenmath does not have There's no need to add everything

gbl08ma, can you add ImageViewer to the Utilities add-in? That would be perfect !