Thanks to the keen eyes of the French calculator community, we discovered a lot more about the probable features of the TI-84 Plus CE, thanks to TI Education France's release of information about the TI-83 Premium CE. We have long seen TI Education France's propensity for what we'll charitably call creativity with handheld names: the TI-83 became the TI-82 Stats.fr, the TI-84 Plus Silver Edition became the TI-83 Plus.fr, and they even dropped the ubiquitous "Plus" in Flash-based models' names to designate the tiny TI-84 Pocket.fr. Therefore, we are all but certain that the TI-83 Premium CE is simply a re-branded TI-84 Plus CE, from the HP Prime-esque thin-and-light case design to the USB-only connectivity. Thanks to TI Education France's released information, we can now tell you a great deal more about the TI-84 Plus CE.

Most importantly to math teachers, the TI-84 Plus CE appears to offer a new exact math engine for the homescreen (not to be confused with a CAS). In one of the videos below, a student Lycee Rosa Parks de Montgeron in France computes a square root, and the answer printed on the homescreen is in the form A√B+C rather than the decimal answer a TI-84 Plus or TI-84 Plus C Silver Edition would give. Documentation indicates that there will be a special exam mode that presumably cripples this math engine for testing situations. Helpfully, TI-France gives us a helpful set of specifications and facts about the new TI-84 Plus CE via the TI-83 Premium CE:
  • LCD: 320x240, 2.8" diagonal
  • Archive/ROM: 3.5MB, likely indicating the calculator has the same 4MB of OS-accessible Flash as the TI-84 Plus C Silver Edition. Vernier's information continues to indicate 3.0MB.
  • RAM: 154KB user RAM. This suggests either banked user RAM, a la the TI-86, or a new 24-bit (ez80) or 32-bit (ARM) processor.
  • Ports: USB OTG port, no DBUS I/O port
  • Graphing Features: Similar to TI-84 Plus C Silver Edition. 10 functions, 6 parametric functions, 6 polar expressions, 3 sequential functions. 15 possible line colors, images usable as graph backgrounds.
  • Math Features: The same matrix, list, statistics, complex number, probability, finance, and math functions available on the TI-84 Plus C Silver Edition.
  • Standardized Tests: Allowed on the SAT, ACT, IB, AP, and presumably other standardized tests.
  • Programming Features: Supports both Assembly and TI-BASIC programming.
  • Preloaded Apps: Polynomial Root Finder and Simultaneous Equation Solver, EasyData 5.0, Cabri Jr, CellSheet, Conic Graphing, Inequality Graphing, Probability Sim, Transformation Graphing.

We also know that there will be a new version of TI-Connect, to be called TI-Connect CE, along with OS 5.0 for the TI-84 Plus CE and EasyData 5.0 to support Vernier hardware. As always, please weigh in in the attached topic. We expect to have official word from Texas Instruments with more information in a week or two, if possible, but for now, let the rampant speculation continue!

More Information & Sources
Promotional TI-84 Plus CE video from TI Education USA
Promotional TI-83 Premium CE video from TI Education France
Video of TI-83 Premium CE in the classroom at Lycee Rosa Parks de Montgeron, France.


From left: TI-Connect CE; math engine demo: "exact" mode answers, "approximate" mode answers
Anyone know someone from the video? Smile.

Either way the tech specs are interesting so far. I wonder if PaD has one yet?

Also an image of the top proving no io port:


And yaapt!:
Did I see Eeems? Good video there, the french one.
CalebHansberry wrote:
Did I see Eeems? Good video there, the french one.
Eeems is Canadian, not French. Our French friends are Adriweb and Lionel Debroux, neither of whom is in the French equivalent of high school. Smile

I see that the video overviews of the TI-83 Premium CE and the TI-84 Plus CE are now private rather than unlisted. Sorry about that, TI! Smile I didn't mean to jump the gun on whatever plan they have for rolling out information on these calculators. Yes, good spotting on the lack of a link port, tr1p1ea, although I was already pretty convinced from Vernier's information on Tuesday. Based on the current evidence, I'm strongly inclined to think that we're dealing with an ez80 or similar rather than an ARM emulator.

Edit: Oh, the TI-83 Premium CE video is still live; it's just the TI-84 Plus CE one that's private.
This is a small scale version of the hunt for Half Life 3 info
It's a bummer i missed the 84 Plus CE video, but around 1:35 into the 83 CE video you can see that the menus are much faster. Graphing also looks faster but still quite a bit slow.

EDIT: Also, what could the difference in Archive memory be due to?
You assume it's not just a sped-up video to show the features for marketing reasons.
bhtooefr wrote:
You assume it's not just a sped-up video to show the features for marketing reasons.
As far as I can tell, the non-live-action videos are just a series of screenshots presented on the rendered calculator's screen as fast as the video's designer decided was appropriate, so I'm not putting too much stock in that.

In other news, DJ_O pointed out that the TI-83 Premium CE in France has a number of keys mapped differently: Apps requires a shifted keypress, there's a [trig] key instead of [sin]/[cos]/[tan], and a few other differences. Here's the image he shared:
KermMartian wrote:
a new CAS-like engine

Eh, no, it's not a CAS engine (TI would have insisted on that quite heavily if it was...), it's an exact math engine - big difference. (By the way, the Nspire CX, for instance, doesn't [even] have an exact math engine...)
And well, a calc from the 83/84-named series with a CAS would be crazy, even with TI's weird naming habits.

KermMartian wrote:
at Lycee Rosa Parks de Montgeron, France.

To be picky, that should be "at [the] Lycée Rosa Parks in Montgeron, France." since Montgeron is a city Smile

KermMartian wrote:
are now private rather than unlisted.

They were both public, not unlisted - I wouldn't have found them otherwise Wink
Looks like TI may have forgotten to check in the upload form...
But anyway, that video doesn't reveal much (or at all, anyway). But from it I was able to get some HD version of images we kind of had already : see the first few images of the album there : https://tiplanet.org/forum/gallery/album.php?album_id=306

KermMartian wrote:
DJ_O pointed out that the TI-83 Premium CE in France has a number of keys mapped differently [...] Here's the image he shared

That would be from critor (especially the image, since he made it), right there: https://tiplanet.org/forum/viewtopic.php?f=41&t=15882#p175812 (in the ≈ middle - right)
The fact that the calc is allowed on the ACT proves that it has no CAS functionality because all forms of a CAS-equipped calculators are banned. Handling certain return values specially doesn't require a CAS.
AHelper wrote:
The fact that the calc is allowed on the ACT proves that it has no CAS functionality because all forms of a CAS-equipped calculators are banned. Handling certain return values specially doesn't require a CAS.
Indeed, I said "CAS-like" hoping to get across the idea that it performs some intelligent processing of the input, rather than just collapse it down to a single value. If the correct term for that is an "exact math" engine, then so be it. Smile
The first sentence of the second paragraph needs to be fixed for a major factual error, beyond the misleading "CAS-like" term: the 84+CE doesn't offer the exact math engine. Only the 83PCE does Wink
Lionel Debroux wrote:
The first sentence of the second paragraph needs to be fixed for a major factual error, beyond the misleading "CAS-like" term: the 84+CE doesn't offer the exact math engine. Only the 83PCE does Wink


How do you know that the 84+CE doesn't have it?
The day-to-day usage of the exact math engine is based on the specific "<>" key, known from the TI-Collège Plus.
The 84+CE doesn't have this key, as shown by the above pictures.
Lionel Debroux wrote:
The day-to-day usage of the exact math engine is based on the specific "<>" key, known from the TI-College Plus.
The 84+CE doesn't have this key, as shown by the above pictures.

Fascinating. I wonder if there's an OS difference, and if there's anything that would prevent such features from being used on the English edition.


Anyway, I was thinking about what sort of tests we could have somebody with a test unit run to give us an idea about what CPU there is. So here are some test programs that can be run to figure out if it's a Z80.

Test 1: Basic "is it a Z80?" test
This test attempts to determine if a very basic Z80 opcode works.

Code:
:AsmPrgm
:C9          RET

Expected output: Program returns immediately without any apparent output.

Test 2: More advanced "is it a Z80?" test
This program tests more complicated Z80 opcodes. It screws with a bunch of RAM, and then fixes it.

Code:
:AsmPrgm
:F3          DI              4
:010000      LD BC, 0        10
:DD21FFFF    LD IX, 0FFFFh   14     
:            loop1:
:DD3500      DEC (IX)        23
:DD2B        DEC IX          10
:10F9        DJNZ loop1      13/8
:            loop2:
:DD23        INC IX          10
:DD3400      INC (IX)        23
:10F9        DJNZ loop2      13/8
:FB          EI              4
:C9          RET

Expected output: Program returns immediately without any apparent output. If it crashes, the it's not a Z80 and there's no point in the rest of the tests.

Test 3: Is it a Z80 at 15 MHz?
This test attempts to approximate the CPU speed the program is running at. If it's still the same Z80 CPU and still running at the same clock speed, then the test will take a really long time. But, if it's a different Z80-compatible CPU, or it's a Z80 at a much higher speed, then the test won't take very long.

Code:
PROGRAM:A
:checkTmr(0)->X
:Asm(prgmB)
:Disp checkTmr(X)


Code:
PROGRAM:B
:AsmPrgm
:F3          DI              4
:010000      LD BC, 0        10
:DD21FFFF    LD IX, 0FFFFh   14
:            loop1:
:DD3500      DEC (IX)        23
:DD2B        DEC IX          10
:10F9        DJNZ loop1      13/8
:            loop2:
:DD23        INC IX          10
:DD3400      INC (IX)        23
:10F9        DJNZ loop2      13/8
:AF          XOR A           4
:            loop3:
:DD3400      INC (IX)        23
:10FB        DJNZ loop3      13/8
:0D          DEC C           4
:20F8        JR NZ, loop3    12/7
:3D          DEC A           4
:20F5        JR NZ, loop3    12/7
:FB          EI              4
:C9          RET
:            ; 470,707,553 clock cycles

Run prgmA. Expected output:

  • If it's a Z80 at 6 MHz (OS 4.0 defaults to 6 MHz for asm programs and apps), then nothing will happen for over a minute, and then prgmA will display a number near 78-79.
  • If it's a Z80 at 15 MHz, then nothing will happen for half a minute, and then prgmA will display a number near 31-32.
  • If it's a Z80-compatible CPU, then it may display a much smaller number. Or not.


Test 4: Is it a faster Z80? Is it a Z80-compatible?
This test attempts to differentiate between a Z80 CPU and more advanced, pipe-lined Z80-compatible CPUs. It does so by looking at the difference in execution time between two different loops. In an un-pipe-lined CPU, the second loop version will take about 50 % longer than the first (that is, the ratio between the first and second execution times will be about 2:3). In a pipe-lined CPU, the difference will be smaller (that is, the ratio will be closer to 1).

Code:
PROGRAM:A
:checkTmr(0)->X
:Asm(prgmB)
:Disp checkTmr(X)
:checkTmr(0)->X
:Asm(prgmC)
:Disp checkTmr(X)



Code:
PROGRAM:B
:AsmPrgm
:F3          DI              4
:010000      BC, 0           10
:21FFFF      LD HL, 0FFFFh   10
:AF          XOR A           4
:            loop1:
:34          INC (HL)        11        8 (6)
:10FD        DJNZ loop1      13/8      6
:0D          DEC C           4
:20FA        JR NZ, loop1    12/7      6/4
:3D          DEC A           4
:20F7        JR NZ, loop1    12/7
:FB          EI              4
:C9          RET



Code:
PROGRAM:C
:AsmPrgm
:F3          DI              4
:010000      LD BC, 0        10
:DD21FFFF    LD IX, 0FFFFh   14     
:AF          XOR A           4
:            loop1:
:DD3400      INC (IX)        23        12 (10)
:10FB        DJNZ loop1      13/8
:0D          DEC C           4
:20F8        JR NZ, loop1    12/7
:3D          DEC A           4
:20F5        JR NZ, loop1    12/7
:FB          EI              4
:C9          RET


Run prgmA. Expected output: Program will wait a while, display a number, then wait again, then display a second number. Record these two numbers.

  • If it's a Z80: The ratio between the two numbers will be between .62 and .71, provided the first number is at least 20.
  • If it's a Z80-compatible with pipe-lining: The ratio will be greater than .71
  • If the ratio is less than .62, it's a WTF.


Test 5: Does it have MLT?

Code:
:PROGRAM:A
:AsmPrgm
:21FF00      LD HL, 00FFh
:ED6C        MLT HL      / 8-cycle NOP
:7D          LD A, L
:21FFFF      LD HL, 0FFFFh
:AF          XOR A
:            loop1:
:34          INC (HL)
:10FD        DJNZ loop1
:0D          DEC C
:20FA        JR NZ, loop1
:3D          DEC A
:20F7        JR NZ, loop1
:FB          EI
:C9          RET

Run prgmA. Expected output:
  • If there is MLT: The program will return almost immediately without any apparent output.
  • If there is no MLT: The program will pause for whatever time the first number in test 4 was, and then return without any apparent output.


Test 6: Is it the same video driver?

Code:
:PROGRAM:A
:AsmPrgm
:F3          DI
:            ; Set color
:16F0        LD D, 0F0h
:            ; Set top window bound
:0E10        LD C, pLcdCmd
:AF0650      XOR A \ LD B, lrWinTop
:D310ED41    OUT (pLcdCmd), A \ OUT (C), B
:D311D311    OUT (pLcdData), A \ OUT (pLcdData), A
:            ; Set bottom window bound
:04          INC B
:D310ED41    OUT (pLcdCmd), A \ OUT (C), B
:D311        OUT (pLcdData), A
:3EEFD311    LD A, colorScrnHeight - 1 \ OUT (pLcdData), A
:            ; Set left window bound
:AF04        XOR A \ INC B
:D310ED41    OUT (pLcdCmd), A \ OUT (C), B
:D311D311    OUT (pLcdData), A \ OUT (pLcdData), A
:            ; Set right window bound
:04          INC B
:D310ED41    OUT (pLcdCmd), A \ OUT (C), B
:3CD311      INC A \ OUT (pLcdData), A
:3E3FD311    LD B, (colorScrnWidth - 1) & 255 \ OUT (pLcdData), A
:            ; Set cursor row
:AF0620      XOR A \ LD B, lrRow
:D310ED41    OUT (pLcdCmd), A \ OUT (C), B
:D311D311    OUT (pLcdData), A \ OUT (pLcdData), A
:            ; Set cursor column
:04          INC B
:D310ED41    OUT (pLcdCmd), A \ OUT (C), B
:D311D311    OUT (pLcdData), A \ OUT (pLcdData), A
:            ; Tell it we're going to write data
:0622        LD B, lrGram
:D310ED41    OUT (pLcdCmd), A \ OUT (C), B           
:014B00      LD BC, 75
:7A          LD A, D
:            loop: ; Main output loop
:D311D311    OUT (pLcdData), A
:D311D311    OUT (pLcdData), A
:D311D311    OUT (pLcdData), A
:D311D311    OUT (pLcdData), A
:10EE        DJNZ loop
:0D          DEC C
:20EB        JR NZ, loop
:FB          EI
:C9          RET

Expected output: The screen should flash a sort of pink color, or some sort of color. If it doesn't, or you don't catch the color, try appending the contents of prgmB or prgmC from test 4.

Test 6: Are there multiple CPU speeds?

Code:
PROGRAM:A
:checkTmr(0)->X
:Asm(prgmB)
:Disp checkTmr(X)



Code:
PROGRAM:B
:AsmPrgm
:F3          DI
:3E00        LD A, 0
:D320        OUT (pCpuSpeed), a
:010000      LD BC, 0
:DD21FFFF    LD IX, 0FFFFh
:AF          XOR A
:            loop1:
:DD3400      INC (IX)
:10FB        DJNZ loop1
:0D          DEC C
:20F8        JR NZ, loop1
:3D          DEC A
:20F5        JR NZ, loop1
:FB          EI
:C9          RET

Run prgmA, and write down its output. Then, edit the second line of prgmB from 3E00 to 3E01, run prgmB again, and write down its output. Test again with 3E02 and 3E03.
Thanks for writing these test programs, DrDnar. If I happen to get my hands on a TI-84 Plus CE before the calculator's official release, I'll definitely do my best to do some sleuthing. If anyone else out there happens to have a TI-84 Plus CE or a TI-83 Premium CE (critor, perhaps?), then testing those and posting the results here would also be helpful. Smile
Apparently, it's a moot point, because Critor said it's an eZ80. Well, the LCD test and CPU speed tests should still be considered.
Yeah I just saw the TI-Planet article now and was surprised. I was sure this would be an ARM machine running some emulator. At least it's much more adequate than a 15 MHz Z80 for screen refreshing.


I am curious how well ASM programs will work? I know some CSE APPs will be available for the calc, but will stuff such as hardware scrolling and 160x240 still work? We all know that on the TI-Nspire 84+ emulator, hardware scrolling (z-adress) didn't work.
DJ_O wrote:
I am curious how well ASM programs will work? I know some CSE APPs will be available for the calc, but will stuff such as hardware scrolling and 160x240 still work? We all know that on the TI-Nspire 84+ emulator, hardware scrolling (z-adress) didn't work.
It appears to be exactly the same LCD. It's likely that LCD routines and techniques will not have to change at all.
ez80? Im ecstatic! I wonder when someone will get their hands on one?
  
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 2
» 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