Threaded Mode | Linear Mode

dm319 · 01-08-2024, 12:13 AM

(please don't laugh at me as I never did complex numbers) but why does sqrt(-2) not give me an imaginary result? Do I need to be in a special mode?

bxparks · 01-08-2024, 12:43 AM

(01-08-2024 12:13 AM)dm319 Wrote: (please don't laugh at me as I never did complex numbers) but why does sqrt(-2) not give me an imaginary result? Do I need to be in a special mode?

You need to change from RRES to CRES mode:
https://github.com/bxparks/rpn83p/blob/d...tion-modes

This part is the same as the HP-42S.

jthole · 01-08-2024, 08:58 AM

Thanks again for a very useful update! (complex numbers)

The TI-84 already has a workable algebraic solver / root finder / integrator, so I personally would find simple keystroke programmability more useful than a solver / numerical integrator.

I would like a well defined API (e.g. through the LastX and register variables) between RPN83P and TI-Basic, but I can imagine that being low on the priority list.

Regarding unit conversions, it would be great if there were a few "free slots", so that users can define their own unit conversions (but that requires some kind of programmability).

My own opinion regarding matrices is the same as with the solver and numerical integrator; there already is a matrix editor, and matrix math, available in the native TI-xx interface. I would be happy switching to that level for matrix calculations.

HobbitJack · 01-23-2024, 02:18 AM

This reminds me of my old (TI-BASIC) RPN program for TI-82... I never ended up finishing it due to a few issues. There I had to custom-implement complex numbers... that worked 'brilliantly!' and has 'no issues!'
Then another time I tried updating a TI-86 z80 ASM RPN program, but that also never ended up working... me and RPN programs!

Keep it up.[/quote]

johnb · 03-02-2024, 05:51 PM

(01-07-2024 11:28 PM)dm319 Wrote: Well done! It installed fine on my TI-83+, and seems to work well. I don't understand complex numbers at all, I'm kinda a bit suspicious of them.

For practical applications, complex numbers are really only needed (AFAICR) for electrical engineering and physics. There, you will find problems that are solved quite nicely by using them, and there are things in the real world you use all the time that rely upon those solutions. (Mobile phones for example.) So, if complex numbers really aren't your thing, it's not a big deal unless you're an EE, physicist, or mathematician.

I believe complex numbers originally arose from the fact that it irks mathematicians to say that any given operation is undefined for some values. For some of those instances, there's no way around it, but for roots of negative numbers, there was, and complex numbers were the result.

There's not a single area of mathematics that I can think of that hasn't found applications somewhere once it became rigorous enough (though my knowledge is annoyingly paltry compare to some people's). Invent it, and they will come. LOL.

johnb · (This post was last modified: 03-02-2024 06:12 PM by johnb.)

(11-21-2023 08:50 PM)bxparks Wrote:
(11-21-2023 07:38 PM)spiff72 Wrote: I don't have an 83 or an 84, but I do have an 89 Titanium laying in a drawer. Is there any possibility that this could be ported to that platform? I think there was an RPN app of some kind for the non-Titanium version of the 89, but it doesn't work on the Titanium.

The TI-89, TI-89 Titanium, TI-92, TI-92+, and the TI Voyage 200 calculators use the Motorola 68000 processor, not the Z80. Basically nothing of the RPN83P code base would be reusable. The development tools of the 89/92/200 are completely different as far as I know, and I have not done any research on those tools. I don't know if TI ever published an SDK for those calculators, similar to their 83 Plus SDK. I know that those calculators can be programmed in C instead of assembly language, which would increase productivity.

Spiff, I succumbed to a post in a different thread and have my bidding engine set to try to grab a Ti-89 titanium for around $30-35. (I want one to as a non-HP comparison point for the accuracy of certain calculations in the HP-48s/g and 50g.) If I'm lucky and grab one cheaply, I might be able to be suckered, er, I mean convinced into collaborating on a port/rewrite attempt.

I've done a little research, and the following tools exist:

Z80-to-68000 assembler translators
Z80 assembler to C decompilers
C/68000 compilers/decompilers

I haven't gotten as far as looking for Ti-89 development environments, but if it can be programmed in either C or assembler, we might have maybe 70% of our effort already done for us. The "workhorse" code @bxparks has written should be completely translatable; only the code that interfaces with the OS or the hardware would have to change. i.e. some of the code might be "recoupable" if not strictly "resuable."

Which brings me to a vaguely related question: "From the viewpoint of a user, not a feature implementor, what's the difference between the Ti-89 series and the 83/84 series?" If I don't snag a Ti-89, what would I miss -- if anything -- by grabbing an 84, for example?

[I already know what I'd gain: your (@bxparks) wonderful RPN environment ready to use! :-) ]

spiff72 · 03-02-2024, 06:51 PM

(03-02-2024 06:10 PM)johnb Wrote:
(11-21-2023 08:50 PM)bxparks Wrote: The TI-89, TI-89 Titanium, TI-92, TI-92+, and the TI Voyage 200 calculators use the Motorola 68000 processor, not the Z80. Basically nothing of the RPN83P code base would be reusable. The development tools of the 89/92/200 are completely different as far as I know, and I have not done any research on those tools. I don't know if TI ever published an SDK for those calculators, similar to their 83 Plus SDK. I know that those calculators can be programmed in C instead of assembly language, which would increase productivity.

Spiff, I succumbed to a post in a different thread and have my bidding engine set to try to grab a Ti-89 titanium for around $30-35. (I want one to as a non-HP comparison point for the accuracy of certain calculations in the HP-48s/g and 50g.) If I'm lucky and grab one cheaply, I might be able to be suckered, er, I mean convinced into collaborating on a port/rewrite attempt.

I've done a little research, and the following tools exist:

Z80-to-68000 assembler translators

Z80 assembler to C decompilers

C/68000 compilers/decompilers

I haven't gotten as far as looking for Ti-89 development environments, but if it can be programmed in either C or assembler, we might have maybe 70% of our effort already done for us. The "workhorse" code @bxparks has written should be completely translatable; only the code that interfaces with the OS or the hardware would have to change. i.e. some of the code might be "recoupable" if not strictly "resuable."

Which brings me to a vaguely related question: "From the viewpoint of a user, not a feature implementor, what's the difference between the Ti-89 series and the 83/84 series?" If I don't snag a Ti-89, what would I miss -- if anything -- by grabbing an 84, for example?

[I already know what I'd gain: your (@bxparks) wonderful RPN environment ready to use! :-) ]

I will be of ZERO assistance in this endeavor, but I would be happy to test out anything you can come up with. My TI89 Titanium is languishing in a drawer somewhere waiting for a good use case!

bxparks · (This post was last modified: 03-03-2024 04:01 AM by bxparks.)

(03-02-2024 06:10 PM)johnb Wrote: I haven't gotten as far as looking for Ti-89 development environments, but if it can be programmed in either C or assembler, we might have maybe 70% of our effort already done for us. The "workhorse" code @bxparks has written should be completely translatable; only the code that interfaces with the OS or the hardware would have to change. i.e. some of the code might be "recoupable" if not strictly "resuable."

While your heart is in the right place, I think you are being overly optimistic. I would guess that almost nothing of RPN83P is transferable to the 89/92/200 calculators. The Z80 assembly code of RPN83P is tightly coupled to the TI83/84 OS, because the OS provides all of the low-level float routines, all of the keyboard scanning and button decoding, and all of the font rendering and printing to the LCD screen. I think it would be easier to rewrite the app in C, rather than trying to convert Z80 into 68000 assembly and emulating the 83/84 OS environment under the 89/92/200 OS environment.

I would be curious to be proven wrong, though, if someone tells me that it's trivial to set up dev environment, and create a "hello world" program for the TI-89/92/200.

(03-02-2024 06:10 PM)johnb Wrote: Which brings me to a vaguely related question: "From the viewpoint of a user, not a feature implementor, what's the difference between the Ti-89 series and the 83/84 series?" If I don't snag a Ti-89, what would I miss -- if anything -- by grabbing an 84, for example?

I'm not sure what you are asking... The 89/92/200 calculators are clearly more powerful machines, for example, they have CAS capabilities, while the 83/84 calculators don't. But for 3rd party apps, the 83/84 machines are better documented and supported by the community as far as I can tell.

(03-02-2024 06:10 PM)johnb Wrote: [I already know what I'd gain: your (@bxparks) wonderful RPN environment ready to use! :-) ]

I don't know if you are aware, someone on reddit told me that there is already an RPN program for the TI-89 that was created 20 years ago: https://www.reddit.com/r/calculators/com..._for_ti89/ . Apparently it runs only on the TI-89, not the 89-T, 92, 92+, or 200. I don't think the source code is available.

Anyway, my impression is that the developer community around the 89/92/200 collapsed around 10-15 years ago, most 3rd party compilers and tools are unmaintained and in disrepair, and much of the technical knowledge on creating 3rd party apps have been scattered and lost. Here's a good summary of the situation as of 2020: https://www.cemetech.net/forum/viewtopic...83&start=0

Personally, I'm more interested in the Casio 9750giii and cg50 series of calculators. These calculators are cheap, currently in production, and use relatively modern processors. The 3rd party development community seems active, and the compiler tools and libraries are readily available and maintained.

bxparks · 03-02-2024, 11:06 PM

(03-02-2024 05:51 PM)johnb Wrote:
(01-07-2024 11:28 PM)dm319 Wrote: Well done! It installed fine on my TI-83+, and seems to work well. I don't understand complex numbers at all, I'm kinda a bit suspicious of them.

For practical applications, complex numbers are really only needed (AFAICR) for electrical engineering and physics. There, you will find problems that are solved quite nicely by using them, and there are things in the real world you use all the time that rely upon those solutions. (Mobile phones for example.) So, if complex numbers really aren't your thing, it's not a big deal unless you're an EE, physicist, or mathematician.

Speaking of complex numbers that I added in v0.9, I received feedback from... let me count... exactly *one* person. Hmm, either my implementation was basically perfect (unlikely!), or no one is using the functionality (more likely). Anyway, I'm polishing towards the next release in a few weeks. If you have any feedback about complex numbers, this might be a good time to send them to me.

johnb · (This post was last modified: 03-03-2024 07:32 PM by johnb.)

(03-02-2024 10:38 PM)bxparks Wrote:
(03-02-2024 06:10 PM)johnb Wrote: ...we might have maybe 70% of our effort already done...

While your heart is in the right place, I think you are being overly optimistic.

LOL! As I presume you're either a professional software engineer like me, or else a professional-caliber amateur with the time occasionally on his hands to do things right (a luxury the professional craves and rarely has), I have to say you're not the first colleague who has pointed out this trend (in general) I seem to have! ;-D

(03-02-2024 10:38 PM)bxparks Wrote: I would guess that almost nothing of RPN83P is transferable to the 89/92/200 calculators. The Z80 assembly code of RPN83P is tightly coupled to the TI83/84 OS, because the OS provides all of the low-level float routines, all of the keyboard scanning and button decoding, and all of the font rendering and printing to the LCD screen.

I was fearing you might say exactly this. Sigh...

(03-02-2024 10:38 PM)bxparks Wrote: I think it would be easier to rewrite the app in C, rather than trying to convert Z80 into 68000 assembly and emulating the 83/84 OS environment under the 89/92/200 OS environment.

Likely, but given your statement below (taken out of order, sorry), I'm sadly losing the motivation to do so. I'm really looking for a project that I can gain traction on quickly, because when Real Life [tm] gets busy, it's altogether too easy to abandon a project that hasn't produced something concrete, whereas if you have something to show and that people are using, somehow the time gets "made" to continue development and maintenance!

(03-02-2024 10:38 PM)bxparks Wrote: Anyway, my impression is that the developer community around the 89/92/200 collapsed around 10-15 years ago, most 3rd party compilers and tools are unmaintained and in disrepair, and much of the technical knowledge on creating 3rd party apps have been scattered and lost.

Amplifying my previous statement above... "well, drat."

* * *

(03-02-2024 06:10 PM)johnb Wrote: Which brings me to a vaguely related question: "From the viewpoint of a user, not a feature implementor, what's the difference between the Ti-89 series and the 83/84 series?" If I don't snag a Ti-89, what would I miss -- if anything -- by grabbing an 84, for example?

(03-02-2024 10:38 PM)bxparks Wrote: I'm not sure what you are asking... The 89/92/200 calculators are clearly more powerful machines, for example, they have CAS capabilities, while the 83/84 calculators don't. But for 3rd party apps, the 83/84 machines are better documented and supported by the community as far as I can tell.

Actually, that pretty much answers exactly what I'm asking. It tells me (a) if I snag a Ti-89 titanium for my desired price range, I'll be happy I got it, and will enjoy playing with it and comparing results with my beloved HP's, (b) if I don't snag one I won't be heartbroken, and (c) regardless of the a/b outcome, I probably should grab me an 84+ and try out your application!

(03-02-2024 10:38 PM)bxparks Wrote: I don't know if you are aware, someone on reddit told me that there is already an RPN program for the TI-89 that was created 20 years ago: https://www.reddit.com/r/calculators/com..._for_ti89/ . Apparently it runs only on the TI-89, not the 89-T, 92, 92+, or 200. I don't think the source code is available.

Nope, was not aware. By "89-T" do you mean the Ti-89 titanium? Sounds like I might need to acquire Yet Another Calculator. LOL.

(03-02-2024 10:38 PM)bxparks Wrote: Personally, I'm more interested in the Casio 9750giii and cg50 series of calculators. These calculators are cheap, currently in production, and use relatively modern processors. The 3rd party development community seems active, and the compiler tools and libraries are readily available and maintained.

I will have to try to find one to look at and play with to see if I want to dive down this additional rabbit hole.

I can hear my wife now, saying "please don't encourage him!!" :-D

spiff72 · 03-03-2024, 07:42 PM

I actually own BOTH a Casio 9750giii and CG50 - so if someone were to port this over to one of those I would be happy to beta test there too!

bxparks · 03-03-2024, 10:21 PM

(03-03-2024 07:30 PM)johnb Wrote: LOL! As I presume you're either a professional software engineer like me, or else a professional-caliber amateur with the time occasionally on his hands to do things right

Retired software engineer, who used to be a physicist. Retirement means that I can do things more "right". But I also have more projects than I can finish in my lifetime, so I still have to force myself to be happy with "good enough".

(03-03-2024 07:30 PM)johnb Wrote: I'm really looking for a project that I can gain traction on quickly

Are you looking for a *calculator* project, or a general programming project? Calculator programming seems to have a LOT of friction. The manufacturers don't want to support it, or are actively hostile against it. They consider calculators to be educational devices, instead of professional devices, so the anti-cheat requirements of school boards seem to be far more important than supporting a 3rd party development ecosystem. I don't know if programming the Casio calculators will be easier than TIs, but at least their hardware don't seem to be locked down like the TI calculators, at least not yet.

(03-03-2024 07:30 PM)johnb Wrote: Actually, that pretty much answers exactly what I'm asking. It tells me (a) if I snag a Ti-89 titanium for my desired price range, I'll be happy I got it, and will enjoy playing with it and comparing results with my beloved HP's, (b) if I don't snag one I won't be heartbroken, and (c) regardless of the a/b outcome, I probably should grab me an 84+ and try out your application!

I recommend the 83+ Silver Edition as your best bang for the money. It's basically identical to the 84+SE (same CPU, same clock speed, same flash and RAM capacity, same features), and can be picked up for very cheap on the used market.

(03-03-2024 07:30 PM)johnb Wrote: Nope, was not aware. By "89-T" do you mean the Ti-89 titanium? Sounds like I might need to acquire Yet Another Calculator.

Yes the 89 Titanium. I bought one on eBay. The screen is terrible. The font is too small, contrast is terrible, the screen surface is shiny instead of matte, so you are constantly angling it away from a light source. Some people say that the screen on the regular TI-89 is better, but there are multiple hardware versions of that model, so your mileage may vary.

johnb · 03-07-2024, 01:06 AM

(03-03-2024 10:21 PM)bxparks Wrote: Retired software engineer, who used to be a physicist.

I think this means we can be friends. :-D
Software engineer for the past 36 years.
Before that, I was a nuclear operator aboard a fast attack sub.
I am -- perhaps weirdly -- intensely envious of the skills held by professional physicists and mathematicians.

(03-03-2024 10:21 PM)bxparks Wrote:
(03-03-2024 07:30 PM)johnb Wrote: I'm really looking for a project that I can gain traction on quickly
Are you looking for a *calculator* project, or a general programming project? Calculator programming seems to have a LOT of friction.

Good question! I've already got one calculator project in limbo. I took a combined professional/avocational interest in interval arithmetic due to the fact that I've never really been satisfied with the methods I was taught for handling computational uncertainty. I've implemented a suite for the HP-48g/gx, but unfortunately it's right in that evil "sweet spot" where it does mostly what I want it to, so it's usable, but there's plenty that's still not quite correct, plus it needs some documentation, so I'd be ashamed to release it publicly as an example of my work.

What I need is a project -- PC or calculator -- that so captures my imagination it enables me to obsess over it for a while. It's great for motivating one to carve out free minutes here and there whenever one can!

(03-03-2024 10:21 PM)bxparks Wrote: The manufacturers don't want to support it [3rd party development], or are actively hostile against it. They consider calculators to be educational devices, instead of professional devices, so the anti-cheat requirements of school boards seem to be far more important than supporting a 3rd party development ecosystem.

Unfortunately, you seem to be correct -- with deadly accuracy -- in this regard. Honestly, from a purely business perspective, to the manufacturers they are educational devices, not professional. By concentrating on the desires/needs of educators, testing organizations, and parents, they're plugged into an ecosystem that guarantees a certain volume of sales... if they don't screw it up. However, only a certain fraction of those math students will move on to STEM careers. And only a fraction of those who do will seek a portable, professional computational platform. Most will go with PC-based or web-based tools like Wolfram Alpha. That means supporting a development ecosystem is likely more expensive than it is worth. Only a company run "by engineers, for engineers" (like the old HP) understands how goodwill percolates throughout a professional userbase.

(03-03-2024 10:21 PM)bxparks Wrote:
(03-03-2024 07:30 PM)johnb Wrote: Actually, that pretty much answers exactly what I'm asking. It tells me (a) if I snag a Ti-89 titanium for my desired price range, I'll be happy I got it, and will enjoy playing with it and comparing results with my beloved HP's, (b) if I don't snag one I won't be heartbroken, and (c) regardless of the a/b outcome, I probably should grab me an 84+ and try out your application!

I recommend the 83+ Silver Edition as your best bang for the money. It's basically identical to the 84+SE (same CPU, same clock speed, same flash and RAM capacity, same features), and can be picked up for very cheap on the used market.

...

I bought one [Ti-89 titanium] on eBay. The screen is terrible. The font is too small, contrast is terrible, the screen surface is shiny instead of matte, so you are constantly angling it away from a light source.

LOL, I am about to find out! The good/bad news is, I made a mistake on my sniping engine, so I've won TWO of them. Fortunately, both cost me less than $30. I think I will keep one, sell the other, and grab the 83+SE to run your library. Unless I really hate the 89, in which case I sell both.

johnb · (This post was last modified: 03-08-2024 06:41 PM by johnb.)

(03-07-2024 01:06 AM)johnb Wrote:
(03-03-2024 10:21 PM)bxparks Wrote: I recommend the 83+ Silver Edition as your best bang for the money. It's basically identical to the 84+SE (same CPU, same clock speed, same flash and RAM capacity, same features), and can be picked up for very cheap on the used market.

LOL, I am about to find out! The good/bad news is, I made a mistake on my sniping engine, so I've won TWO of them. Fortunately, both cost me less than $30. I think I will keep one, sell the other, and grab the 83+SE to run your library.

Well, either there are a BUNCH of 83+SE's that aren't labeled as an SE on the faceplate, or else there's a LOT of sellers on TAS [That Auction Site] who don't know what they've got (or are dishonest). I am guessing it's the faceplates that are accurate, not the auction descriptions: other readers, beware and look carefully before buying, the 83+SE has 9 times the flash RAM and 2x the processor speed of the non-SE 83+, and most (wow!) of the calculators being sold as an SE in the description are not SE's.

BTW, further evidence TI regarded this as purely an educational product: that hideous translucent case! When I was in 8th grade, I would have thought that was the coolest thing in the world. By the 11th, I wouldn't be caught dead with the thing.

Fortunately, I found a black one (sans cover) for under $20.

johnb · (This post was last modified: 03-13-2024 07:46 PM by johnb.)

Two of my three "new" TI's have arrived.
Notes:

Nice enough for a high schooler, but they really just don't have the...
gravitas... of any of my HP calculators!
They're not flimsy, but they're not solid like an HP.
Ti-89 Titanium:
- Maybe I lucked out, the display is matte on mine.
- However, you're correct on the TINY, TINY font!
  Not operable without wearing 9.5+ diopter old-fart magnifying glasses.
Ti-83 Plus Silver Edition:
- With the black case it is not unattractive.
- Just actually seems more usable than the 89-T.
  So I'm looking forward to trying your RPN suite.
- WHOOPS -- I didn't get a cable! Read below.

Does anyone have a spare TI graph link cable for an 83+SE that they'd be willing to sell cheaply? I can't try bxparks' cool RPN software until I can load it on the machine! :-D

Also... I think I'll play with the 89-T for a while but -- unless I find some kind of project I can use it for -- I think I won't be keeping it.

Dave Britten · 03-13-2024, 08:26 PM

Somehow I missed the discussion of this when it came up a few months ago. Definitely going to load this up on my 83+ Silver and 84+ Pocket. Adding RPN, base conversions, and a few unit conversions could actually push the 83/84 into daily-driver territory for me. It looks like there's a decent selection of unit-conversion functions, but my one request would be to make it as complete and flexible as the 19BII, TI-85, etc.

sunpazed · 03-13-2024, 10:48 PM

This a great tool given that any good HP RPN is hundreds of dollars.

I tested your app with an emulator, and it’s really quite good, well done!
It motivated me to purchase a second-hand TI-83+ with Arduino converter cable for $30. I look forward to checking it out on a real device.

bxparks · 03-14-2024, 12:53 AM

(03-13-2024 07:43 PM)johnb Wrote: Does anyone have a spare TI graph link cable for an 83+SE that they'd be willing to sell cheaply?

If you are referring to the USB-to-2.5mm SilverLink cable, I accidentally bought an extra one a while ago. Send me a PM with your email and we can discuss offline.

Yeah, the 89T seems like a great calculator. But the screen is unusable for me. You're right, it's slightly matte, but shinier than the 83+/84+.

bxparks · 03-14-2024, 01:02 AM

(03-13-2024 08:26 PM)Dave Britten Wrote: Somehow I missed the discussion of this when it came up a few months ago. Definitely going to load this up on my 83+ Silver and 84+ Pocket.

The 84+ Pocket SE has come up a couple of times previously. I don't think anyone has explicitly confirmed that RPN83P works on it. If you can verify that, that would be useful.

(03-13-2024 08:26 PM)Dave Britten Wrote: Adding RPN, base conversions, and a few unit conversions could actually push the 83/84 into daily-driver territory for me. It looks like there's a decent selection of unit-conversion functions, but my one request would be to make it as complete and flexible as the 19BII, TI-85, etc.

I'm not familiar with the 19BII or the TI-85. Can you expand? The most flexible solution, of course, is to support keystroke programming with custom menu bindings and custom key bindings. I did a quick estimate of how long it would take to implement keystroke programming: I got something like 1000-2000 hours. I might decide to just do it, for the technical challenge, but it will take a while, 1-2 years.

Dave Britten · 03-14-2024, 11:40 AM

(03-14-2024 01:02 AM)bxparks Wrote:
(03-13-2024 08:26 PM)Dave Britten Wrote: Somehow I missed the discussion of this when it came up a few months ago. Definitely going to load this up on my 83+ Silver and 84+ Pocket.

The 84+ Pocket SE has come up a couple of times previously. I don't think anyone has explicitly confirmed that RPN83P works on it. If you can verify that, that would be useful.

Seems like it's working fine on my Pocket. It is, after all, just an 84+ SE in a smaller form factor.

(03-14-2024 01:02 AM)bxparks Wrote:
(03-13-2024 08:26 PM)Dave Britten Wrote: Adding RPN, base conversions, and a few unit conversions could actually push the 83/84 into daily-driver territory for me. It looks like there's a decent selection of unit-conversion functions, but my one request would be to make it as complete and flexible as the 19BII, TI-85, etc.

I'm not familiar with the 19BII or the TI-85. Can you expand? The most flexible solution, of course, is to support keystroke programming with custom menu bindings and custom key bindings. I did a quick estimate of how long it would take to implement keystroke programming: I got something like 1000-2000 hours. I might decide to just do it, for the technical challenge, but it will take a while, 1-2 years.

Certainly! On these calculators, you select a category of units (length, area, temperature, etc.), and are given a menu with various units to choose from. To convert, you first press the unit you are converting from, and then press the unit you are converting to. So if I wanted to convert 25 ft to m on the 19BII, I would type 25, press "FT", and then press "M", which would convert the value in the X register. The TI-85 is similar in that you get a menu of units, and press two in sequence to enter a complete conversion function. So to enter the expression 25[ft]>[m], you would type 25, press "FT", then press "M". You can include as many conversions as needed in a single expression.