Like most people here, I have a TI-84 plus CE. (It's what got me interested in programming.) Since I am taking calculus this year, I thought it would be a good idea to get something with CAS. I ended up with the TI 89T my brother left behind when he went off to college. After using it for a little bit, I started looking for how to program it.

Well, it looks like I was late to the party... Almost twenty years late!

I looked up the TI89T specs, and it is superior in every way to the TI84 series!(not the CE) The processor is faster, (And the 68K is way better than the z80) it has more RAM, more flash, a better screen, a better operating system, better BASIC programming, a better C, blah blah blah. You get the point. Why isn't there a community for these calculators any more? I know there was one twenty years ago, but I can't piece together what happened to it. All the old sites have links to other sites that no longer exist. They all haven't been updated in years. What happened? I know there was a huge flame war, something to do with the c compiler, but how did that destroy an entire community? Why did the z80 series become more popular than these obviously superior 68K calculators?

I have way too many questions...

Could someone who knows these things tell me some more?

Should I even bother programming for the 89? Or should I stick with the CE?
Quote:
Could someone who knows these things tell me some more?

I got my 89 HW2 20 years ago (possibly to the day, I'm pretty sure it was in August 2000), started attending the international community on the TICT message board in late spring 2001, and the French community on yAronet/TI-FR in the autumn of 2001.

Quote:
Well, it looks like I was late to the party... Almost twenty years late!

You're indeed late to the party, though not by as much as 20 years Smile
The community belittled a lot around 2006-2007, i.e. in the same time frame as the dead Nspire CAS+ and the production Nspire Clickpad, and slightly before the first iPhone, so not at all because of those. As for why exactly the community belittled and never picked up... well, despite the fact that flamewars in the years before 2006-2007 did discourage some people from participating to the community, it could also be the effects of a generational change, with wider availability of much higher-powered computers and other computing platforms (including some game consoles) whose price tags had decreased, and people moving on from calculators to these platforms.

Quote:
I looked up the TI89T specs, and it is superior in every way to the TI84 series!(not the CE) The processor is faster, (And the 68K is way better than the z80) it has more RAM, more flash, a better screen, a better operating system, better BASIC programming, a better C, blah blah blah.

Pretty much correct, apart from "better C" nowadays: the simultaneous advent of a LLVM-based toolchain for the eZ80 and the continued decrease of the correctness and optimization strength of m68k toolchains, make the comparison less clear-cut.

Quote:
All the old sites have links to other sites that no longer exist.

Some of them have disappeared, indeed. However, let it be said that some of the old sites which still exist, and have re-hosted content made by other people with attribution but without permission, are the only remaining sources for some great pieces of work whose original sources have disappeared. For instance, Samuel Stearley's excellent, unmatched CAS-related work. Sadly, there hasn't even been enough re-hosting going on, other programs have disappeared "forever" from public download. All the more open source wasn't the norm back then.

Quote:
I know there was a huge flame war,

There were always huge flamewars with some people. And yes, as a matter of fact, I took my fair share in those, both before and after understanding that the main perpetrator abused my youth and naiveness...
Obviously, I'm not neutral when telling the story, but my argumentation is based on facts, which could be checked, usually easily enough, since main sources such as yAronet remained available Smile

Quote:
something to do with the c compiler, but how did that destroy an entire community?

Nah. The community was already dead for two years before GCC4TI was announced publicly.
GCC4TI brought various improvements (bugfixes, new functionality, new library and header functions, optimizations, better portability, some preparation for the "slow death" life cycle phase, etc. - some of these pieces had been contributed years earlier to GCC4TI's dead ancestor, but the maintainers didn't act on them), but it didn't help reviving the community, it was already too late. That's something I realized only after seeing hardly any user response to the announcements.
Neither did NewProg, which you'll find is probably the closest thing to a TI-68k counterpart of Axe Parser (or rather, if NewProg is older, which it may well be, Axe Parser is some form of TI-Z80 counterpart to NewProg), help reviving the community in any amount remotely resembling what Axe Parser achieved for the TI-Z80 series.

GCC4TI did break the community free from an improductive / counter-productive stranglehold on the toolchain. Following growing frustration with the way the toolchain was maintained in dictatorship style, the fork was triggered by an announced plan, in the summer of 2008, to scrap the Delphi IDE entirely and switch entirely to KDE 4, since that was supposedly portable to Windows.
Fast forward 12 years: Delphi still exists and evolves, FreePascal+Lazarus has improved, and KDE (whose internal interfaces have kept breaking multiple times in backwards-incompatible ways, forcing applications to adjust time and again) remains nonexistent in terms of real-world usage on Windows.
IOW, the plan of switching entirely to KDE was mind-numbingly disconnected from reality in 2008, it still is, and I can't foresee how it will... just about ever become a good idea, in fact. Not to mention that the KDE 4-based IDE which was produced during a GSoC project has, uh... lackluster coding style and maintainability. You could look for yourself in the GCC4TI repository if you fancy doing that.

The stranglehold on the toolchain also materialized in the spokes thrown in GTC's wheels by the maintainer of GCC4TI's dead ancestor. GTC's quite an engineering feat if you ask me, an on-calc C compiler with a higher-powered computer version, quite a bit better than the older cc/as tools. The dirty play delayed GTC by enough time for Pollux to become busy with other stuff outside the TI community, and TI to stop replying to FlashApp signing requests, eventually prompting the development of Flashappy.

I could have spent much more time working on GCC4TI, whose code base I knew better than average after contributing to the dead ancestor (I'm in the credits for that), but it made more sense to work on libti*/gfm/tilp, because they had more users... so that's what I did, most of the time, since 2009.

Quote:
Why did the z80 series become more popular than these obviously superior 68K calculators?

First and foremost, I'd blame the f* stupid standardized tests - especially, but not only, USA standardized tests - which forbid the usage of a CAS for stupid reasons. TI-68k calculators were more popular in France, where there were (and still are, despite the now mandated stupid exam mode) no restrictions on CAS functionality.
However, the higher price tags were certainly major contributors to this outcome as well, let's face it.

Quote:
Should I even bother programming for the 89? Or should I stick with the CE?

Well, if you want your work to be useful to more users for the years to come, the choice is clear-cut: stick with the CE (and depend on jailbreaks when intermittently available).
Or buy a NumWorks calculator: at the time of this writing, that is an open platform, whose hardware characteristics, on average, surpass those of TI-68k and TI-eZ80 calculators by a wide margin, apart from the amount of RAM. With the Omega firmware, a modern CAS (giac, same as xcas on computers, the HP Prime's standard OS, KhiCAS on the Nspire and fx-CG50/Graph 90+E) is available. But of course, NumWorks calculators are much less popular than calculators from the best-known graphing calculator manufacturers established long ago.
The TI-68k series remains a good learning platform for yourself. Besides, it was a source of inspiration for the TI-eZ80 series, both by TI (the TI-eZ80 FlashApp and OS formats, as well as transfer protocols for those, are close derivatives of those for the TI-68k series) and the community (the CE toolchain and libs took a number of clues from TI-68k practices). Just don't expect TI-68k creations, no matter how polished and impressive, to become popular Smile

On IRC, DrDnar mentioned TI's SDK, TI Flash Studio. There are two pieces in TIFS:
* the IDE, which requires Microsoft's incompatible Java which Sun burned to the ground in the Win XP era. The embedded emulator is outright terrible: even VTI, with its documented emulation inaccuracy and incompleteness, does a much better job emulating a TI-68k calculator;
* the Sierra C toolchain. The C compiler is poor: it only supports a dialect of C even more outdated than GCC 4.1 does, fairly often generates code which makes experienced assembly programmers shake their heads, and sometimes generates invalid code. However, that the toolchain is the only way to produce TI-68k FlashApps. Not that anybody in the community did that since... before the validation keys were factored and the signing keys deduced, in 2009. TI-68k FlashApps were always unpopular among developers because of comparatively lower restrictions than TI-Z80 FlashApps, and the need to send FlashApps to TI for signing, which was an annoying procedure, and as I mentioned above, one day, TI stopped replying to such requests.
The calculator version of GTC was a FlashApp, in order to both avoid the RAM consumption of the compiler's code, and skip the ~64 KB size limit (without jumping through loopholes). IIRC, it was partially compiled with GCC4TI's dead ancestor to take advantage of its better optimization, but I couldn't find a trace of that, the public archives I could find for GTC seem to be incomplete.
The TI-89T is the best calculator of all time.
  
Register to Join the Conversation
Have your own thoughts to add to this or any other topic? Want to ask a question, offer a suggestion, share your own programs and projects, upload a file to the file archives, get help with calculator and computer programming, or simply chat with like-minded coders and tech and calculator enthusiasts via the site-wide AJAX SAX widget? Registration for a free Cemetech account only takes a minute.

» Go to Registration page
Page 1 of 1
» All times are UTC - 5 Hours
 
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

 

Advertisement