Review:
http://edspi31415.blogspot.com/2015/11/c...eview.html

but the #1 question is: can we shoehorn new firmware into it?!
cheers,
rob :-)

History is gone when powered off...variables stored in memories remain. Wish I could define f(X) in table, and use it in calculate mode to solve for roots or solve d f(X)/dx to find max/mins without having to retype the function in. Wish I could do the reverse process as well. That said, solve is great. I can do solve d/dx, then delete the d/dx and hit = to get the f(X) value for where d/dx = 0. I can even solve integral from point of interest to X to find what X value results in the integral producing some assigned value.
Solve and integral returns values accurate often to full display where as TI 36 Pro often is accurate to about 5 digits.
Polynomials return all zeroes including imaginaries, degree 2 returns zeroes, and the X and y values of the polynomial's max or min.
In regression, one can store a,b,c constants that the regression calculation generates, and use in rewriting the equation in calculation mode, and find predicted y value for given X or predicted X value for given y.
A nice pocket companion.

11-09-2015, 02:12 AM (This post was last modified: 11-09-2015 02:13 AM by Eddie W. Shore.)

Here is a speed test of the fx-991EX versus Canon's F-792SGA:
https://www.youtube.com/watch?v=39tcSQBakMo

(11-09-2015 02:12 AM)Eddie W. Shore Wrote: Here is a speed test of the fx-991EX versus Canon's F-792SGA:

What astonishes me is that, apart from the screen resolution, both calculators not only show the same results, they do it in a very similar way. Do they share a common software base?

The Casio fx-991DE X is a remarkable machine; the integration function is about as fast as the programme
http://www.hpmuseum.org/forum/thread-424...ht=lobatto
on the HP 50G.

The U.S. Model (fx-991EX) can handle d/dx and integrals in the solve environment (allowing one to quickly find a max/min of a function by asking d/dx to equal zero in the solver).
Can the European models accomplish this? I ordered the Japanese version with it's many more functions, and was surprised that d/dx and integrals are not allowed in the solver!

(12-21-2017 02:59 AM)lrdheat Wrote: ... I ordered the Japanese version with it's many more functions, and was surprised that d/dx and integrals are not allowed in the solver!

Hello,
Can you show an example of the formula with d/dx or integral you used in the 991 EX solver ?
Thanks

(11-09-2015 02:12 AM)Eddie W. Shore Wrote: Here is a speed test of the fx-991EX versus Canon's F-792SGA:
https://www.youtube.com/watch?v=39tcSQBakMo

Interesting the summation. It could be used, with a proper formula (maybe not only using the square and inverse functions), as general benchmark for calculators.
I always thought that the scientific calculators have the logic directly done in circuits instead of an operating system and a generic purpose CPU. Surely the Canon uses very similar circuits.

On the 991EX, I can type in an equation such as d/dx(sin(x))=0 ,enter the solver, put in a guess such as 2, and it will report a number equivalent to pi/2. It even works in the solver if I simply enter d/dx(sin(x)) It assumes in this case that I want to find a root.

On the JP900, this procedure produces a syntax error upon providing a guess. The actual entry is d/dx(sin(x)) x=x =0 Again, the =0 can be left out if you are only interested in finding a root. Works on the 991EX.

(12-21-2017 03:09 PM)lrdheat Wrote: On the 991EX, I can type in an equation such as d/dx(sin(x))=0 ,enter the solver, put in a guess such as 2, and it will report a number equivalent to pi/2. It even works in the solver if I simply enter d/dx(sin(x)) It assumes in this case that I want to find a root.

Most solvers, if given an equation, will look for a root. If you just give them a function then they'll look for a zero.

(12-21-2017 03:14 PM)lrdheat Wrote: On the JP900, this procedure produces a syntax error upon providing a guess...

It's surprising because the manual doesn't say anything about that while Casio usually does mention that d/dx and integral are prohibited within solver, as in the manuals for the fx 5800p, fx 991 ms for example.
Was Casio desperately hunting for bytes for it's jp900 ? It would have been better to leave table of periodic elements ... ?
In the 5800p I ended up implementing a simple secant solver in those cases.

I'm wondering if the U.S. 991EX model is unique in allowing d/dx, integrals in the solver, or if some of the European models have this capability.
The JP900 is a nice unit, but I would trade the periodic table for the d/dx in the solver!

(12-22-2017 03:39 PM)lrdheat Wrote: I'm wondering if the U.S. 991EX model is unique in allowing d/dx, integrals in the solver, or if some of the European models have this capability.
The JP900 is a nice unit, but I would trade the periodic table for the d/dx in the solver!

German fx-991 DE X can solve such integrals & differentials.

The fx-991CE X allows d/dx, integrals in the solver also.
I tried to find the upper limit for equation f(x)=0.45 with lower limit equals to 0 and the result is 1.645.
The f(x) means normal probability distribution

(12-23-2017 05:45 AM)Gerald H Wrote:

(12-22-2017 03:39 PM)lrdheat Wrote: I'm wondering if the U.S. 991EX model is unique in allowing d/dx, integrals in the solver, or if some of the European models have this capability.
The JP900 is a nice unit, but I would trade the periodic table for the d/dx in the solver!

German fx-991 DE X can solve such integrals & differentials.

Unfortunately the menu language German only.