Zoom Math 200

[hdhippodrome]

So, I got myself a nice new TI-84. And knew nothing about it, except I had to have it for school. So I did some research and found out this seems to be the home of the folks that know the most about TI calculators.
My question is, has anyone used the Zoom Math 200 app, and is it worth the cost? If I was wealthy, I would buy it and be happy- but, since buying it will mean eating ramen noodles and generic soda for a week, is it worth it? Or is there another app that is better or cheaper with the same functionality?
And, if this is the wrong place to ask, please tell me where to go, so to speak. HAHA.
Thanks :confused:

[FloppusMaximus]

[DrDnar]

No, it is not worth it. The application is only meant for algebra I students. Once you're through algebra I, it won't be any use to you. If you use to pass algebra I, you'll only be screwing yourself over.

If you want to know what it's capable of, download the manual from the website. It won't solve logarithmics, trig problems, or any other fun things you won't encounter in algebra I. And the grapher doesn't have asymptote detection, so it's actually less power than the TI-84's normal one.

Symbolic is pretty neat, but the danger of such applications is the same as in getting more advanced calculators: you won't learn to do the real math.

[jeffhatch]

Hello hdhippodrome,

I work for Zoom Math.

As far as I know, there is no other App other than Zoom200 that shows the steps of solving Algebra 1 problems. And no TI calculator that does that, either.

But did you buy a TI-84 Plus Silver Edition? If you did, then it probably has App4Math installed already. App4Math doesn't do algebra, but it has the same interface as Zoom200. Go ahead and try it; see if you like the Zoom Math interface.

We plan to release Zoom Math 300 in a month or so. It will have some Algebra 2 features, such as logarithmic equations. Once Zoom300 comes out, we'll knock down the price of Zoom200 a little bit. So if you just want Algebra 1, then I recommend you join our email list and wait a little while.

I'm pretty sure some of our customers use Zoom Math to cheat in Algebra 1, and they don't learn to do the real math. They're just "screwing themselves over," as Dr. D'nar said. But one college prof told us that his Elementary Algebra students who buy Zoom200 usually do better in his Stats course later on. Some high schools bought classroom sets of Zoom200, and they seem to be satisfied. So Zoom200 can help with learning, or it can prevent learning. Please calculate responsibly.

(Also, Dr. D'nar said, "The [Zoom Math] grapher doesn't have asymptote detection, so it's actually less [powerful] than the TI-84's normal one." But technically, the TI-84's normal grapher does not have asymptote detection either. When the TI-84's normal grapher draws a line that looks like an asymptote, the grapher is making a mistake drawing the graph, and the mistake happens to look like an asymptote. The Zoom Math grapher is slightly modified so these false asymptotes show up less often. Visit http://mathbits.com/mathbits/tisection/Trig/asymptotes.htm for a very good explanation of why the TI-84 appears to draw asymptotes.)

Sincerely,

Jeff Hatch
The Zoom Math Team

[FloppusMaximus]

Jeff, you're talking about the behavior of the old grapher, which doesn't know anything about discontinuities, and hence usually draws vertical lines at asymptotes. From OS 2.30 onwards, there is a slightly more clever graphing algorithm, which does detect most discontinuities and avoids drawing those lines. This is done by evaluating the function at one or more additional X values in between each pair of plotted points (which means, of course, that graphing is considerably slower when using the new algorithm.) The new algorithm is only used when the calculator is in function mode and Xres is set to 1.

[DrDnar]

I guess we should have asked what math you're in. I was thoroughly unimpressed because I'm a calculus student. The stuff Zoom does is child's play to me. That's why Flop and I seem so negative about it. I suppose from the perspective of an Algebra I student it might be a wonderful app.

When I mentioned asymptote detection, I meant how TIOS <2.30 would graph lines at discontinuities. (Try graphing y=int(x) on the 84 with xres = 1 and in Zoom Math.) We considered how TIOS would graph lines at discontinuities a bug. Zoom Math doesn't have discontinuity detection. If I wanted TIOS to graph asymptotes, I would want it done with a dotted line. For that matter, I would want horizontal asymptotes to be graphed too.

I suppose teachers that allow Zoom in their classroom had better also be aware of Push-to-Test, another feature Flop and I aren't fond of. Maybe I'm just old-fashioned.

I use Symbolic for checking my work sometimes. Symbolic sometimes simplifies expressions in a funny manner. I've been thrown off for a second a few times before realizing the output was mathematically equivalent to my answer.

[jeffhatch]

I didn't realize that newer versions of the TI OS detected discontinuities and avoided drawing those false asymptotes. (TI never updated the OS that comes with the SDK, and that's the OS I usually see.) That seems like a good idea--and really, I don't see any reason it should be limited to xres=1. I will make a note to fix our algorithm accordingly. Thanks for the info, Flop and D'nar.

Hey, do you happen to have any idea what algorithm the TI OS uses to detect a discontinuity? Do they just sample a few extra points between each pair of pixel columns? Hmm, I suppose we could try using symbolic manipulation to find all the discontinuities in advance. That might actually save time by reducing our sampling requirements. I'll have to think about this.

And yeah, sooner or later I would love to give Zoom the ability to draw dotted-line asymptotes of common functions--including vertical, horizontal, and diagonal asymptotes.

Jeff Hatch
The Zoom Math Team

[thornahawk]

"Hey, do you happen to have any idea what algorithm the TI OS uses to detect a discontinuity? Do they just sample a few extra points between each pair of pixel columns? Hmm, I suppose we could try using symbolic manipulation to find all the discontinuities in advance. That might actually save time by reducing our sampling requirements. I'll have to think about this."

- I don't know either, but I'd hazard that barring symbolic manipulation, you can crudely detect poles (at least for functions y=f(x) ) by monitoring the slope or distance between two consecutive sample points. If the slope is too steep or the distance too far (depending probably on a user-settable tolerance), the line joining two points in a plot should probably be excluded.

That strategem of course is useless for functions like the floor and unit step functions, which have step discontinuities rather than poles. The slope monitoring strategy may work, but ii is my experience that results are less than satisfactory especially if such functions are buried in complicated expressions.

As for symbolic manipulation for rational functions, you now bump into the problem of solving a polynomial equation if the degree of the denominator is greater than two or three...

thornahawk

[jeffhatch]

Good ideas, thanks!

We might end up combining several techniques. For example, we could use a symbolic manipulator to differentiate a function, and where the first derivative is positive but the slope of the connecting line turns out to be negative, we can assume there is a discontinuity and not draw the line. Or maybe we can sample more points in suspicious cases, to get a better idea.

An interesting issue.

Sincerely,

Jeff Hatch
The Zoom Math Team

[thornahawk]

[DrDnar]

TIOS oversamples. Compare a TI-84+ with OS >2.30 with and TI-84+ or TI-83+SE with OS <2.30. The CPU clock speed is the same, but the graphing on OS >2.30 is much slower when xres = 1. I don't know the details of how the OS does it. You might try talking with BrandonW. He'll most likely point you to some part of the OS that you should disassemble. For emulation, Wabbitemu is compatible with the TI-84's operating system. You can use ROM8X to dump a ROM for it. It has some nice debugging features you can use to look at what the OS is doing.

[Mapar007]

I've got a full disassembly lying around (coming from Brandon's server, by the way, but as he's reorganizing...), but I don't have it at hand right now, so just drop me a PM or something if you want it. You should look at the _GRPHPRS bcall, I think, and dig further from there. A school teacher asked me if I could find a way to fix that problem on the <2.30 OSs, so I'll be looking into it myself.

[jeffhatch]

At first glance, Wabbitemu looks good!

Hey, while I'm here, does anyone know much about the calculator's communication protocol? Some of our users are not very computer-savvy, and the process for downloading an App via TI-Connect is a bit complicated. I'd love to write a little program that lets you just double-click an .8xk file to download it to the calculator. Or has someone done that already?

Actually, TI's development team has been quite helpful so far. I might be able to get some answers from them.

Jeff Hatch
The Zoom Math Team

[Mapar007]

Here: http://www.ticalc.org/archives/files/fileinfo/247/24750.html

For serial link, and partially for USB too, but that documentation is incomplete. (AFAIK)

[jeffhatch]

Mapar007, I'm not going to take a look at the TI 2.30 graphing algorithm for several weeks, or possibly several months. I've got some other urgent projects to finish first, and so does my graphing guy. But in the meantime, if you find out anything, I would love to know.

Jeff Hatch
The Zoom Math Team

[DrDnar]

If you want to write your own software to send an application, check out TILP, a third-part program for linking to TI calculators from a PC. The author has stopped working on it, but I believe it already supports 64-bit Windows and Vista. I also believe it is open source, so you can borrow from it as long as your own application is open source.

Also, what in the world did you do to get TI actually to provide you with useful information? They haven't done so much as update the errors in the existing documentation (e.g. SFont_Len returns the width in B, not A) in the SDK since about 2002. Much less provide documentation on USB, or features like the dialogue BCALLs, which provide TIOS-like dialogues, but are never used by the OS itself.

[jeffhatch]

Some of our customers have a lot of trouble getting TI-Connect to work, especially with Vista. TILP might really help. Thanks for pointing me in the right direction!

Ever since TI asked us to create a math notation App for the TI-84 Plus Silver Edition, we've been on friendly terms. The last time I asked a technical question, I had to wait a month or two, but I did get an answer eventually.

My favorite SDK error is in the ZiLOG Z80 manual. They claim that POP qq doesn't affect any condition bits. Back in the day, it took me quite a long time to figure out why POP AF was mucking with the Z condition bit!

Jeff Hatch
The Zoom Math Team

[FloppusMaximus]

I don't know all the details of the graphing algorithm - and it's been a couple years, I think, since I looked at it - but I believe it does something similar to what you guys were describing.

[calcdude84se]

[FloppusMaximus]

No, POP AF does what it says on the tin - retrieves two bytes from the stack and stores them in A and F. It doesn't affect the flags beyond that. So if (SP) = 40h, the zero flag will be set (and all other flags will be reset) regardless of the value of (SP + 1).