TI Bans all Assembly Programming, Casio Limits Add-ins. Please don't follow suit HP.

[Jean-Baptiste Boric]

(05-27-2020 06:36 PM)HP67 Wrote:  

(05-26-2020 09:01 PM)Jean-Baptiste Boric Wrote:  Wikipedia says the old 512 bit RSA has been factored for more than a decade. I recall reading somewhere TI introducing a 2048 bit RSA key as a response. Unless you have contacts at the NSA or many millions of dollars to spare, I doubt that key will be broken anytime soon.

People said the same thing about 3DES and 1024 bit keys and MD5 and were proven overly optimistic before too long.

This issue just doesn't seem as hopeless and dire as advertised.

I'd very much like to be proven wrong on this, even if just to join everyone in collectively saying "Told you so" to Texas Instruments. After all, there could be a leak of the private key, a catastrophic bug lurking inside TI's cryptographic code, somebody winning the one-in-a-million-years lottery while brute-forcing the key or some fundamental flaw in RSA that we do not know about.

I just find any of these options very unlikely to happen in the near future at the time of writing.

This does not mean that I consider a jail-break of TI's calculators very unlikely to happen too ; but that it's much more likely to happen through a random but exploitable bug lurking somewhere in their firmware as usual.

[CyberAngel]

(05-26-2020 10:43 PM)medwatt Wrote:  X
I wish I can sell my HP Prime 2 without too much loss because the more it stays around, the more my blood boils when I realize I burned more than €120 on it. I absolutely have no use for it and I refuse to learn HP's version of Basic. I have no intentions of wasting my time on a programming language that works on a single device.

I might buy it! - I just got a laudatur in High School Math and head to the uni.
I have absolutely no use for another HP Prime, but I'd like to compare G1/G2.
- -
VPN

[Kevin Ouellet]

https://www.youtube.com/watch?v=Myh9UsFMcy8

This video showcases the performance of python on the TI-83 Premium CE compared to other calculator models. Python on the CE is limited by speed, program size (I think 17.7 KB max?) and if its proprietary commands are used then the program cannot be ported to other calculator models. Even TI-BASIC is better.

[grsbanks]

(05-28-2020 02:28 AM)CyberAngel Wrote:  I have absolutely no use for another HP Prime, but I'd like to compare G1/G2.

In a nutshell, there is a 3x increase in speed from the G1 to the G2.

http://techy.horwits.com/2018/12/hp-prim...hmark.html

[Jean-Baptiste Boric]

(05-30-2020 07:10 PM)grsbanks Wrote:  

(05-28-2020 02:28 AM)CyberAngel Wrote:  I have absolutely no use for another HP Prime, but I'd like to compare G1/G2.

In a nutshell, there is a 3x increase in speed from the G1 to the G2.

http://techy.horwits.com/2018/12/hp-prim...hmark.html

More importantly, an 8x increase in RAM from 32 MiB to 256 MiB. Anecdotally, a 2x increase in Flash from 256 MiB to 512 MiB.

[debrouxl]

(05-26-2020 10:43 PM)medwatt Wrote:  Meanwhile, TI gave them Python instead which will probably make up for the lack of Assembly and even more.

Obviously not, otherwise the community of users would be fuming somewhat less.
Others have replied before I did: the trivial graphical operation of setting a pixel on TI's Python is horrendously slow - 2-3 orders of magnitude slower than the competition - and the smallest Python heap size on the market limits Python scripts to like 2-3 KB of raw source code. That's outright unacceptable - the most limited school algorithms work, but anything slightly outside that narrow box won't work.
TI added lots of proprietary, non-portable Python modules.
The suggestion by TI EdTech head Peter Balyta to use Python instead of native code adds outright insult to the severe injury, and makes the community resent TI's move even more strongly.

At least, your Prime, the one you're lamenting about in the rest of your message, has a chance to get a proper implementation of Python some day. The TI-eZ80 series... never, especially as long as TI insists on using the current hardware

(05-26-2020 06:22 AM)cyrille de brébisson Wrote:  Anyhow, the main issue here for TI and Casio is that when a government or minister tells you: "If you can't guaranty that the test mode is secure, you will not be allowed on the national exam"... well, you do it. Point end of story. High school education is 99% of their market.

The people who regulate these national exams are flat out incompetent, as the way the hardware and software of the calculators are designed and implemented, and the way calculators are used in the first place, guarantee the insecurity of the test mode.
Users have unrestricted physical access to the calculators, which is considered a "game over" condition by various threat models.

[medwatt]

(06-01-2020 07:27 PM)debrouxl Wrote:  At least, your Prime, the one you're lamenting about in the rest of your message, has a chance to get a proper implementation of Python some day. The TI-eZ80 series... never, especially as long as TI insists on using the current hardware

You might be right, but I don't see the point in waiting for 5 years for something like this when something like the Nspire CAS II will be having it soon. I actually looked at the history of the Prime and was quite shocked that the current Prime is almost the same as the Prime that was released in 2013. The only relevant addition is probably the 3D graphing app and that came after many many years of waiting. Meanwhile, the triangle solver app was the highlight and still is the highlight of the calculator.

There's a function for the continuous Fourier Transform on XCAS but is not included on the Prime because of licensing. I asked Tom about it and he told me the only way such a function could be included is for the author of the code to change his license. Believe it or not, I asked the guy and he he agreed to change the license and that he would contact the HP team directly. I wrote back to Tom informing him about it. He didn't bother replying again which means the reply he gave me back then (why he cannot add the function) was based on the fact that he didn't believe the guy would allow his code to be used on the Prime and it's a nice excuse for Tom to not include the function.

[pinkman]

Yes I followed these conversations here: https://www.hpmuseum.org/forum/thread-14...#pid126108

Did you contact Bernard Parisse to ask him to include the CFT with its more opened licence in HP Prime CAS package and deliver it to Tim (not Tom) and HP teams?
Bernard answers quickly to mails, and there is also a XCas forum: https://xcas.univ-grenoble-alpes.fr/forum/

Regards,
Thibault

[parisse]

(06-02-2020 04:38 AM)medwatt Wrote:  I actually looked at the history of the Prime and was quite shocked that the current Prime is almost the same as the Prime that was released in 2013.

This is wrong. For example in the CAS area, a lot of bugs have been fixed, and some new commands/features have been added, like domain or CAS programming in Python syntax.

[parisse]

(06-02-2020 05:41 AM)pinkman Wrote:  Did you contact Bernard Parisse to ask him to include the CFT with its more opened licence in HP Prime CAS package and deliver it to Tim (not Tom) and HP teams?

This is not a technical question (it's fairly easy to change one #ifdef in the source code), it's a licensing question where I'm not involved.

[debrouxl]

(06-02-2020 04:38 AM)medwatt Wrote:  You might be right, but I don't see the point in waiting for 5 years for something like this when something like the Nspire CAS II will be having it soon.

Not only you certainly won't be waiting for 5 years before the Prime grows Python capability (a build containing an early Python implementation become available for a short amount of time months ago, and there's high market pressure for Python... I have no inside information, I'm not a beta-tester, but I wouldn't be surprised if a HP Prime OS version made available around the BTS season provided something)... but also, the Nspire series is actually a great example of users having to wait for years for TI to implement highly wanted features, you know
Namely:
* 12 years for the Nspire TI-Basic to provide some graphics drawing commands. Users had been requesting them right from the beginning, because the Nspire's TI-Basic was inferior in that regard to the TI-81's Basic from the early 1990s;
* nearly, or more than, 6 years for Python... the feature was requested directly to TI management at the end of 2014 (I participated in writing the presentation), and I can't know whether we were the first ones requesting it.

Both features are only provided on the CX II (CAS), despite the fact that the CX I would be able to run all of the improvements that TI is intentionally reserving for the CX II, albeit somewhat more slowly. The CX II (CAS) hardware has neither more RAM, nor more Flash, nor AFAIK more 2D graphics acceleration than the CX I (CAS) hardware. The sole improvement is on the CPU speed: ~2.5x, from 150+ MHz (on newer OS versions - TI changed the LCD and had to add a task which constantly copies the screen data to the rotated format used by the new LCD, at significant CPU cost) to 400- MHz. Clearly, a 2.5x CPU speedup is nothing to sneer at, but the '2011 CX I would already be so much faster at Python than the '2015 TI-eZ80 series that it isn't funny. Not providing Python and better TI-Basic onto the CX I series is a non-technical decision.

If you don't have one yet, buying a CX II (CAS) would be supporting TI's loathsome policies, and I'm not sure you want to do that, either.

[medwatt]

(06-02-2020 06:20 AM)parisse Wrote:  

(06-02-2020 05:41 AM)pinkman Wrote:  Did you contact Bernard Parisse to ask him to include the CFT with its more opened licence in HP Prime CAS package and deliver it to Tim (not Tom) and HP teams?

This is not a technical question (it's fairly easy to change one #ifdef in the source code), it's a licensing question where I'm not involved.

I guess that answers the question regarding who's responsible.

[medwatt]

(06-02-2020 06:18 AM)parisse Wrote:  

(06-02-2020 04:38 AM)medwatt Wrote:  I actually looked at the history of the Prime and was quite shocked that the current Prime is almost the same as the Prime that was released in 2013.

This is wrong. For example in the CAS area, a lot of bugs have been fixed, and some new commands/features have been added, like domain or CAS programming in Python syntax.

I mean no disrespect to you, but I would hardly consider bug fixes as the best feature of your device when the device is supposed to be bug free.

[debrouxl]

Well, basically, no device (not just graphing calculators) is ever bug-free
Even software developed using formal methods - which is far more difficult and costly, to the point that it can only be done for code bases much smaller than giac is - is subject to bugs in the specification, the formal validation code, and/or of course the underlying hardware. seL4's proofs assume that the processor registers, the MMU and various other hardware parts are working according to the published specification. Then there are the side channels, vulnerabilities related to speculative execution, etc.

[pinkman]

(06-02-2020 07:14 AM)medwatt Wrote:  

(06-02-2020 06:20 AM)parisse Wrote:  This is not a technical question (it's fairly easy to change one #ifdef in the source code), it's a licensing question where I'm not involved.

I guess that answers the question regarding who's responsible.

PM sent to Tim.

[parisse]

(06-02-2020 07:16 AM)medwatt Wrote:  I mean no disrespect to you, but I would hardly consider bug fixes as the best feature of your device when the device is supposed to be bug free.

We are not living in an ideal world where software are bugfree, in the real world bugs are slowly fixed once they are discovered, and many companies are *paying* for support and upgrades. Looking at the TI Nspire, you should really consider that it is a feature of the Prime to get free upgrades for the G1 despite the fact that the G2 is now here for a couple of years!

[debrouxl]

(06-02-2020 01:49 PM)parisse Wrote:  Looking at the TI Nspire, you should really consider that it is a feature of the Prime to get free upgrades for the G1 despite the fact that the G2 is now here for a couple of years!

Right. That said, both TI and HP have done short duration support, or no support, for some of their models

* HP gave up on the 39gII, but I know it's not their fault;
* TI-68k series: the 89 & 92+ received OS upgrades for only ~1 year after the V200 came out, the V200 received OS upgrades until 1 year after the 89T came out - and the 89T itself received only two OS upgrades, the last one a bit less than two years before the Nspire became available;
* TI-Z80 series: the 82A and 84+-T, of which are minor, recent, crippled derivatives of the decade-old 84+, received no OS upgrades whatsoever. There are probably other examples, but I'm less familiar with that series;
* Nspire series: TI had previously supported the Nspire Clickpad & Touchpad for a while after the CX I came out, until their 32 MB Flash memories were maxed out by the OS upgrade, which occurred after OS 3.9 - but the CX I is not about to hit such a limit. I don't remember about a previous occurrence of TI giving up on OS upgrades for a model immediately after the next model hit the market...

[parisse]

(06-02-2020 09:57 PM)debrouxl Wrote:  I don't remember about a previous occurrence of TI giving up on OS upgrades for a model immediately after the next model hit the market...

The reason migh be that the CX II is not a big hardware upgrade, if they want users to buy a CX II (instead of a used CX), they must give the CX II a software advantage.
That's why I decided to port the enhanced KhiCAS user interface I made for the Numworks on the Nspire CX with ndless, this should give a second youth to these Nspire CX. It would be great to have it on the Prime too but there are licensing issues.

[CyberAngel]

(06-02-2020 04:38 AM)medwatt Wrote:  

(06-01-2020 07:27 PM)debrouxl Wrote:  At least, your Prime, the one you're lamenting about in the rest of your message, has a chance to get a proper implementation of Python some day. The TI-eZ80 series... never, especially as long as TI insists on using the current hardware

You might be right, but I don't see the point in waiting for 5 years for something like this when something like the Nspire CAS II will be having it soon. I actually looked at the history of the Prime and was quite shocked that the current Prime is almost the same as the Prime that was released in 2013. The only relevant addition is probably the 3D graphing app and that came after many many years of waiting. Meanwhile, the triangle solver app was the highlight and still is the highlight of the calculator.

There's a function for the continuous Fourier Transform on XCAS but is not included on the Prime because of licensing. I asked Tom about it and he told me the only way such a function could be included is for the author of the code to change his license. Believe it or not, I asked the guy and he he agreed to change the license and that he would contact the HP team directly. I wrote back to Tom informing him about it. He didn't bother replying again which means the reply he gave me back then (why he cannot add the function) was based on the fact that he didn't believe the guy would allow his code to be used on the Prime and it's a nice excuse for Tom to not include the function.

Could you yet again check the situation, please?
While I just got the best grade in High School Advanced Math and thus the need for Prime went close to zero
I'm still interested in any advances in the functionality expansion via license restriction release. Bring them in!
- -
VPN

[TheLastMillennial]

I just published a video today that summarizes all the events that led to TI's decision. It contains tons of information not covered in this thread!