HP12C+



#66

According to DATAFILE V27 N5 a new edition of the 12C is going to be sold, featuring the opportunity for firmware updates and modifications like the HP20b offered recently. Does anybody know whether there are more changes included therein, e.g. a different LCD?


#67

Quote:
According to DATAFILE V27 N5 a new edition of the 12C is going to be sold, featuring the opportunity for firmware updates and modifications like the HP20b offered recently. Does anybody know whether there are more changes included therein, e.g. a different LCD?

If it uses the new ARM processor as per the 20b, that chip is perfectly capable of driving the existing display.

If this is true though, then a scientific version can't be that far away...

Dave.


#68

Would the 12C LCD allow for turning it into a 15C?


#69

Quote:
Would the 12C LCD allow for turning it into a 15C?

The main problem is the keys.

Dave.


#70

If HP knows we are willing to tear apart perfectly good hardware to get to the JTAG interface, perhaps they would be so kind as to manufacture generic or blank buttons for us to insert during the calculator surgery. Preferably the blank ones would be white, or light, ready to accept Sharpie ink.

(This is probably not news to many here): As seen in the link below, the keys appear to be individually placed in the case, which would allow for endless keyboard configurations.


http://www.decodesystems.com/hp11c-apart.html



PG


#71

In my opinion, the procedure shown at the link you provided makes disassembly look a lot easier than it is. The picture shown with step 3 is not what you would see after removing the screws and removing the back. At that stage, you have two pieces: the back shell, and a "sandwich" consisting of the printed circuit board, key contacts, keys and front shell. Separating that sandwich requires a rather nasty process of trimming 40+ heat stakes to free the front shell and allow it all to be separated. The stakes must be carefully trimmed, removing only the part of the stake that is "mushroomed" over. The center portion of the mushroom must be left intact and protruding through the PCB so that there is something to either “re-mushroom” or glue to when reassembling. Reassembly of the sandwich in a tight, sturdy and reliable manner after the keys are re-arranged and/or the keys and keyboard itself are relabeled is something which I believe would be beyond the abilities of most mortals.

But, since the subject 11C was built in 1985, perhaps new voyagers utilize a different type of construction. I do not have a new “ 12C+” or even a late model 12C (i.e. with a single 2032 battery) to take apart to check. I do have a 12CP 25th Anniversary model. Being the curious sort, I attempted to remove the back. First off – it has 5 screws, the four under the feet and one in the battery compartment. There are also several plastic catches around the sides (one each on the top and sides, two on the bottom) which were freed with some gentle prying to separate the back. Suffice it to say that it doesn’t look a lot like the 11C. However, this may be good - although the PCB and the rest of the stuff between it and the front shell do not readily separate, there do not appear to be a bazillion heat stakes. Without taking an exacto knife to them, it appears that there are just 11 heat stakes that hold things together. What’s more, these stakes look like the ones in the 35s (disassembly detailed here), except without the screws. So maybe they could be trimmed, then re-fastened with tiny screws. However, the keys may be on frames, like the 35s. If the 12C+ uses similar construction, perhaps disassembly and re-labelling may be viable. Of course all of the above is speculation, which Cyrille or one of our friends with a 12C+ are warmly invited to confirm or correct.


#72

Quote:
But, since the subject 11C was built in 1985, perhaps new voyagers utilize a different type of construction. I do not have a new “ 12C+” or even a late model 12C (i.e. with a single 2032 battery) to take apart to check. ... However, the keys may be on frames, like the 35s. If the 12C+ uses similar construction, perhaps disassembly and re-labelling may be viable. Of course all of the above is speculation, which Cyrille or one of our friends with a 12C+ are warmly invited to confirm or correct.

I do have a non-working "late model 12C" I took apart (s/n CNC5070...). I found still the good old 39 separate keys, placed individually in their frames as in 1981. No information about newer editions though.

#73

Was it held together with heat stakes? If so, how many? How did you hold it together upon re-assembly?


#74

Quote:
Was it held together with heat stakes? If so, how many?

41. Plus 6 screws.
Quote:
How did you hold it together upon re-assembly?

Not. As mentioned, it's a non working unit. I bought it for very few money just for curiosity to see what's inside. I'd not kill a living Voyager for this ;)
#75

If it's still the old LCD :-/ , it was the same for all Voyagers.


#76

Same LCD.


#77

Gene, I hope you get some royalties...

Just found this ;-) <<<CLICK

Edited: 11 Nov 2008, 4:36 a.m. after one or more responses were posted


#78

Ha!

Nope, none of those suggestions will make it into a 12c any time soon.

I don't think any model will actually be called a 12c+ anyway. It's just a faster 12c.

Much, much faster.


#79

I thought bank and mortgage people didn't trust 12c Platinum numbers because it was too fast, or is that folklore?

#80

So this is the knock out for any menus :-(

#81

Quote:
If it uses the new ARM processor as per the 20b, that chip is perfectly capable of driving the existing display.

If this is true though, then a scientific version can't be that far away...

Dave.


Seriously? I'd kill for a superfast 15C.

Just don't ask whom I would kill...

#82

Quote:
According to DATAFILE V27 N5 a new edition of the 12C is going to be sold, featuring the opportunity for firmware updates and modifications like the HP20b offered recently. Does anybody know whether there are more changes included therein, e.g. a different LCD?

I've asked this before--do we know what CPU is used in the 12C+? I've not found one in the stores yet, or I'd have already opened it up.

#83

hello

Quote:
I've asked this before--do we know what CPU is used in the 12C+? I've not found one in the stores yet, or I'd have already opened it up.

opening it up will not help you, the CPU is in die form under a blob of black stuff... but it's the same ARM7 from Atmel than on the 20b.

Unfortunately, the HP 12C+ does NOT have the Jtag interface (sorry) due to ESD reasons... BUt it still has the serial interface, so you can still reprogram it! I will be probably releasing data on the HP 12C+ soon including schematics and an example application.

regards, cyrille


#84

Quote:
Unfortunately, the HP 12C+ does NOT have the Jtag interface (sorry) due to ESD reasons... BUt it still has the serial interface, so you can still reprogram it! I will be probably releasing data on the HP 12C+ soon including schematics and an example application.

regards, cyrille


No JTAG certainly puts the brakes on my reverse engineering fun, but releasing schematics will be appreciated.

I'm liking this trend toward open hardware...

#85

Hi Cyrille,

May I ask your honest opinion an why you think HP or their subcontractors aren't able to produce a decent keyboard anymore. Please don't take this as an offense but rather as a constructive comment. I've just bought a HP-20b and deeply regret it. Ok I bought it mainly as I was interested in ARM CPU programming. I don't think I'm spoilt by good old and expensive HP calcs but I expect a minimum usability. My claim is that it's almost impossible to do some error free calcs on these new HPs due to many missed keystrokes. And I can hardly believe that it's impossible to design a reliable keyboard on a budget. I don't expect molded keys but a good tactile feedback.

I hope the new 12C+ is better in this respect.

Regards from Switzerland
Daniel


#86

Don't ask this of Cyrille. You know perfectly well he can't answer while he works for HP.


#87

Hi Bruce,

Actually it was my intention to ask Cyrille as he works for HP and I hope he's got something to say on harware evaluation. I also like to mention that I like the software of the 20b and the fact that this is the first business calc to also have scientific functions. But I had it used by some of my coworkers in the aircraft R&D dept and they all struggled over the unusable keyboard. I haven't given up all my hopes.

Regards Daniel


#88

I do not have a 20b but a 35s, 17bii+ silver, and 12c anniversary edition. Last 3 models feature keyboards with IMHO reasonable tactile response, and my samples record my keystrokes reliably. So, my conclusion is: YES HP CAN produce "right" keyboards still :)

Seems to be impossible for them, however, to merge all advantages into one model. You get firmware access (20b) OR a row of soft keys (17bii+) OR landscape design (12cp) OR scientific (35s). You MAY get a good keyboard (3 out of 4). You will get 16x131 dot matrix of 1988 (17bii+), 7 segments display of 1981 (12cp), 2x14 pixel blocks (35s), or crippled dot matrix (20b). You will NOT get a state of the art display. Hard to understand for me. Must be marketing ;)

Ceterum censeo: HP, launch a 43s (or a 15s or ... well, you know).

Walter

Edited to add the display variants.

Edited: 11 Nov 2008, 2:33 p.m.


#89

You created a lot of excitement in the community with the open hardware developed recently. Some projects are on their way. Assessing the opportunities, however, it comes down the display driving means are rigidly limiting any custom developments based on these platforms. On the other hand, the largest LCD available so far in an HP RPN calculator is the one seen recently in the 17bii+ Silver again, featuring some 2100 elements. I don't know if you are willing (and allowed) to answer the following 2 naive questions, but I ask them anyway:

  1. Is there a chance for a calculator with the form factor of the 17bii+ Silver *with an JTAG interface or similar*?
  2. Where are the limits of this LCD driver?
If my request shows a significant lack of electronic understanding then I apologize right now. I'm no electronic engineer, I'm just wondering and trying to understand. Thanks in advance for your response, if possible.

Walter


#90

Walter,

Sorry, I'm not Cyrille:-) Perhaps he will chime in but if not...

Was not your second question answered here?

.


#91

Jeff,

IIRC these 400 elements were the limits of the chip in the 20b. I want to know the limits of the 17bii+ ... or did I miss anything?


#92

Quote:
Jeff,

IIRC these 400 elements were the limits of the chip in the 20b. I want to know the limits of the 17bii+ ... or did I miss anything?


As far as I know, the 17bii+ uses a 6502 based core with mask or OTP rom, not flash. That means we can't hack new code into the microcontroller.

OTOH, the 20B uses the Atmel AT91SAM7L128, which can be updated in the field with new code. The internal LCD controller is limited to 400 segments.

#93

OK, sorry, I misunderstood. But from what I have read, any and all new designs will be based on the new chip. So if you want more than 400 elements, you need the outboard driver circuitry which increases costs, etc. etc.

.


#94

Fully agree it increases costs, but - just in case it didn't become clear yet - I don't want a mushy calc for 40US$, I want a serious scientific instrument instead.

To make it simple: Right now the top of HP's calculator line sells for some 130US$ retail, with a huge LCD and tremendous power many of us don't need, and a RePeLling OS and a large housing many of us don't want. Somewhere in between the 20b and the 50G, presumably next to the upper limit, must be a niche for the heir of the 42S. According to the survey most of us participated, many are willing to pay even more for such a device than for a 50G.

So if you claim

Quote:
any and all new designs will be based on the new chip

I doubt it, since the 17bii+ Silver contains something different, and since I doubt HP will abandon intentionally all medium level RPN activities and reduce its own market to cheap calcs with low margin. Maybe I'm terribly wrong, however. Dunno. Anyway, DaveJ and Hugh Steers have proven they can build a very nice device within limited time from scratch with standard material and homegrown SW, and for reasonable costs, though far away from mass production. So, if HP will provide a feasible platform, it may well start something like "personal calculating" which may be not only a big boost for its renommee in the sci-tech world but also a direct economic success. It may even become "cool" to taylor one's own calculator. I'm no marketeer - but also the decision for the 35 was not taken by marketeers, nevertheless it became a great (unexpected) success because it started something new fulfilling the needs of professionals.

So, please, HP, after creating the 20b as a guinea pig for the new concept of open calcs, and after extending that approach to the 12C forced by short supplies, please do the next step and provide a platform allowing to make something serious - if you don't want to make it completely on your own.

<end of rant - or in German: Das musste mal gesagt werden!>

Ceterum censeo: ... you know.

Walter

Edited: 12 Nov 2008, 6:12 p.m.


#95

Quote:
Fully agree it increases costs, but - just in case it didn't become clear yet - I don't want a mushy calc for 40US$, I want a serious scientific instrument instead.

Rest assured, your desires are very clear :-)
Quote:
To make it simple: Right now...many are willing to pay even more for such a device than for a 50G.

I agree with everythign you said here.
Quote:
So if you claim
Quote:
Quote:
any and all new designs will be based on the new chip


Not a claim, just an assumption.
Quote:
I doubt it, since the 17bii+ Silver contains something different, and since I doubt HP will abandon intentionally all medium level RPN activities....

I certainly hope you are correct.


#96

Quote:

Rest assured, your desires are very clear :-)


It's going to be tough to have it all. The 20B CPU is maxed out at 400 segments, and it only has 6kB of ram. You'd have to add both an external LCD controller and external memory before it would be capable of replacing the 35S or 17bii+ guts.

You'd gain the ability to do field firmware updates and maybe a better toolchain than with the 6502 core. Doesn't seem like enough in the pro column to make up for the cons.

The 12C+ switching to the '7L128 CPU makes a lot of sense to me--the chip can drive the existing display and has enough ram for the 12C platform, it's proven from the 20B project, so why not? Hackability and field updates are a nice side effect.


#97

A short calculation of the number of LCD segments in the 35s display results in more than 1000. So no chance to use the built in controller of the processor in the 20b exclusively.

#98

hello

Quote:
  1. Is there a chance for a calculator with the form factor of the 17bii+ Silver *with an JTAG interface or similar*?

    > sorry, but I can not answer that question... however, I can tell you that doing so would require a redesign of the 17BII+ with a new chip which is non trivial

  2. Where are the limits of this LCD driver?


> for the 20b, it's 400 segments. for the current 17BII+ CPU, is't in the 2500 range if my memory serves me well.



regards, cyrille

#99

The 400 segments the new chip can drive would allow for a 15C+ featuring an LCD almost like the one the 32sii had. Just 10 blocks of 5x7 pixels, corresponding to 10 digits as in the Voyagers before, instead of 12 as in the Pioneers. This would allow for everything a Voyager did show so far, plus some alpha display (e.g. mnemonics instead of key codes!) and up to 5 menu keys.

A poor-man's solution may be using Pioneer displays and dropping the last 2 blocks, provided the LCD would fit geometrically into the Voyager housing.

Just crossed my mind ...

Edit:
Forget the last paragraph: Recycling Pioneer displays needs the Voyager housing being 6mm higher. But a new versatile 400 segment LCD of this kind is worth thinking about, when we need a PCB redesign for the new chip anyway.


Edited: 17 Nov 2008, 5:03 p.m.

OK, so if we keep the old display, we won't get menus nor alpha prompts nor mnemonics in program listings, what can still be done despite all these awkward limitations? I tried a really careful redesign of the 15C, to avoid - beware - frightening any conservatives, and eventually came to this:


I'm not sure about the use of blue shifted COS, one may place something like Re<>Im there instead. Please, however, find the conversion key as blue shifted CHS, to be combined with the shifted functions of keys 4 through 9. But don't tell them.

Remember: no menus, no mnemonics, no alpha. Just a big step into ... the past.

Ceterum censeo: HP, launch a 43s or 15s.

Walter


Based on the 12C+ discussion -- that it's running the original 12C ROM image -- if there is a 15C+ on the horizon it's going to be an original 15C only a lot faster. The only additional cost to produce this given, the 12C+, would be printing the keys and the faceplate (and perhaps more money to Eric). I'll bet that one is on the way pretty soon.


Quote:
Based on the 12C+ discussion -- that it's running the original 12C ROM image -- if there is a 15C+ on the horizon it's going to be an original 15C only a lot faster. The only additional cost to produce this given, the 12C+, would be printing the keys and the faceplate (and perhaps more money to Eric). I'll bet that one is on the way pretty soon.

Yep, betcha Cyrille already has one too.

I wonder if the market for original 15C and 11C's will plummet if/when such a calc comes out?

Dave.

To complete the information:

BASE 0  = integer base 10
BASE 2 = BIN
BASE 6 = HEX
BASE 8 = OCT
BASE . = DECM

MODE 0 = ALG
MODE 1 = RPN
MODE 2 = proper fraction mode (HR or BASE . will return to DECM)
MODE 3 = improper fraction mode
MODE 4 = /c in 32sii (set maximum denominator)
MODE . = toggles decimal radix mark

Edited to add HR.


Edited: 20 Nov 2008, 2:22 a.m.

Nice work as usual. But alas it has two of my pet peeves - a dedicated HYP key, and a shifted 1/X key.

Dave.


Dave, thanks for your feedback. The simplest solution may be to swap these 2 function labels. The result will be:

Function    keystrokes now    and after swapping

1/x g 1/x 1/x
sinh HYP SIN g HYP SIN
arcosh INV HYP COS INV g HYP COS

I simply shrinked from requiring 4 keystrokes for 3 "plain" functions, but I understand your request. I may even change it after another night ;)

Walter

Edited: 19 Nov 2008, 6:21 a.m.

Quote:
OK, so if we keep the old display, we won't get menus nor alpha prompts nor mnemonics in program listings, what can still be done despite all these awkward limitations? I tried a really careful redesign of the 15C, to avoid - beware - frightening any conservatives, and eventually came to this: ...

As a long-time HP-15C owner/user, I'll admit I find this new layout a little bit disorienting. The true purist would complain you're not allowed to move ANY keys, but that's unrealistic and limits the ability to sneak in extra features in a sensible manner.

What are the added white markings for? Parentheses are pretty obvious if you're going to include a non-RPN mode, and I'm guessing the upper-left/lower-right combination replaces the SST and BST keys for stepping through programs. How about the dashes which seem to be grouping keys?

Finally, I've had very little use for the hyperbolic functions and would much rather see (1/x) as a primary key.

Chris,

Quote:
What are the added white markings for? Parentheses are pretty obvious if you're going to include a non-RPN mode, and I'm guessing the upper-left/lower-right combination replaces the SST and BST keys for stepping through programs. How about the dashes which seem to be grouping keys?
Finally, I've had very little use for the hyperbolic functions and would much rather see (1/x) as a primary key.

You're right. The upper left quarter of the design looks this way now:

The white dashes there are between all keys which's primary function may be combined with INV.

The white marks in the upper right quarter are between all keys allowing mode conversions. I.e. > DEG, RAD, GRD, POL, REC, H:M:S, HR, REAL, CPX. You set e.g. polar mode by pressing f POL. Within this mode you convert the contents of x and y (representing the magnitude and the polar angle) to Cartesian components by pressing g > g REC.

And the upper-left and lower-right keys are for BST and SST, as you guessed, and also for scrolling e.g. large binary numbers, and for leaving a parenthesis in ALG mode. Parentheses are inserted in pairs always. They are for indirect addressing, too. For example, pressing STO + () 07 adds x to the memory register 07 is pointing to.

Edited the last paragraph and moved some keys to achieve even less difference to the 15C.


Edited: 20 Nov 2008, 6:40 p.m. after one or more responses were posted

I would rather change the [ f ] key to [ALPHA] and use the left-hand side of each key to show a neon-green letter, except for [ENTER] [<=] and the shift keys. what about only [ g ]-shift then? Let's make it a ROTating shift: f,g,none,f,g,none,...[f,g] What about other "alpha" than the letters A..Z? You could put them under the first alpha A..Z using another color and naturally you press [ALPHA], then the [f,g] shift key then the appropiate key. This way we can really clutter up the keyboard! :-)

Edited: 19 Nov 2008, 9:42 a.m.


Veli-Pekka,

please explain the use of [ALPHA] without an LCD capable of displaying them.

Kiitos,

Walter


You use what you can... :-/

Meant for writing secret messages...

SCNR;-)

Not something I would endorse, but the TI-59 had it: some alpha capability without an alpha display, an external printer was needed to take advantage of it.

In a modern model, it may be of use when interfacing to a PC via USB or so... Just thinking...


Alpha capabilities could be useful, if the display were a reincarnation of the segmented 41C variety. It should have no more than 400 single elements and fit easily the new processor.

I wouldn't want a programmable calculator with numeric code display.


100% agree!

BTW, my car's radio+cd player has such 14-segment LCD, very similar to the HP 41C and I actually find it rather readable and nice (of course, it also calls fond 41C memories).

Luckily, it never showed "MALFUNCTION", "NONEXISTENT" or "MEMORY LOST" yet!

:-)

Quote:
I wouldn't want a programmable calculator with numeric code display.

I double that, I can't be bothered memorizing numeric codes anymore, that page should be closed at any expense. New calculator is expected to have a lot more memory either...

Cheers,
Reth

Do I recall correctly that Eric had a patch for the 15C firmware so that it could use more memory? I think there was some problem with displaying the step numbers, but it worked in nonpareil. So more mem might no be a problem.

I second your opinion. And such a segmented LCD would allow for at least 2 more digits/characters. Compare an alternative here, also within the limits of 400 elements.


the 41C has 14-segments plus a (semi)colon eg 17 together for each character. some annuciators like flags 0..4 ALPHA USER ... with 400 elements thats 2*11 characters. If we leave out extra dot for semi(colon) and satisfy ourselves with just comma/point eg, 16 segments => 2*12 characters and some extra annuciators! - I want another CPU and a real dot matrix LCD...


Displays are the cornerstones.

This thread has shown this again. HP, please equip your calculators accordingly -- at least those you want to sell for good money. There are enough cheap calcs in this world, please proof you can make a difference. I think every member of this forum is at least willing to support this process.

Quote:
According to DATAFILE V27 N5 a new edition of the 12C is going to be sold, featuring the opportunity for firmware updates and modifications like the HP20b offered recently. Does anybody know whether there are more changes included therein, e.g. a different LCD?

When will HP launch the HP 12C+ ?

Apart from using new CPU, any additional features?


Already selling overseas.

No new features other than the serial connection.

Sorry, no cable for that in the box.

How to tell the difference? new 12c uses two 2032 coin batteries.


Gene,

if I understand you correctly, then there will be no difference visible looking on the keyboard (or top) side of the 12c. Y/N?

Quote:
Already selling overseas.

Where?


Yes, no visible difference on the keyboard side of the update.

Being in the USA, I can't answer the second question.

But, if you find a 12c on the shelf somewhere that has a second coin-battery, then you have it.

Or, a 12c that does around 60,000 counts a MINUTE in the " + GTO 01" loop, then you have it. :-)


Will both the original 12c and the 12c 25th anniversary be updated?


No algebraic on this guy. It is a 12c, not a 12cp.

The 12C+ raises an interesting question.

They have completely revamped the hardware architecture on their highest selling and longest running calculator, with no name change or other visible marketing difference - Why?

Were existing components about to become obsolete and they were forced into a change? (unlikely I'd say, but there have been a lot of silicon fabs going under recently)

Is the new solution cheaper to manufacture? Given that they have been selling the old one in large volumes for a long time, and the fact that they'd have to amortise new development cost again, this seems a bit unlikely. Perhaps in the long run?

Is it an investment in a new strategic direction? Where they can more easily churn out new models for niche markets based on common platforms. i.e. they plan to bring back the Voyager platform (and Pioneer like 20b) in scientific and maybe other flavours. This seems likely.

Financial models may have come first because they are a better cash-cow to get the concept off to a flying start perhaps?

Perhaps another reason(s) entirely?, or a combination?

Dave.


My guess is that they will start using just 2 AR'M CPU's
so it is a more strategic directive plus I guess it also lowers the overall cost on the long run, so it's therefore - well - financial decision. just my 2 eurocents...

Nice theory. Let me quote from the Editorial of Datafile V27 N5:

Quote:
This new model (12C+) ... was first mentioned at the HHC2007 conference when (Cyrille) mentioned that the processor used for the 12C Platinum and 25th Anniversary Edition was no longer in production and that stocks were coming to an end, forcing HP to re-design the calculator around a new processor.


Well there you go. Thanks for the clarification.

As a designer I hate it when the "last orders taken" call comes form the manufactuer. Do you buy 10 squillion of them to last you 20 years, or do you redesign...

It's often a big ask to take an ASIC to another fab, so it was probably a no-brainer for them to re-design.

Dave.


The problem with redesigning the 12C is that bugs can/will creep into the firmware. Much of the financial community depends on this calculator and trusts the results. A serious bug will erode that trust and HP could loose the market for this calculator.

Also, we're not the only group of users that are picky about the feel of the keys. I get comments from all my clients (I consult to wall street firms) that they hate the feel of cheap plastic keys on the newer 12c's. I sure hope that the 12C+ doesn't make this even worse.


NOW there is finally hope for new fast and large-memory 
HP-15C
with memonics instead of keycodes,
maybe A..Z labels (and matrices)
and 35s-style row-number addressing
and...
<drool>

Hyvää päivää Veli-Pekka,

You were right IF they would spend a reasonable display. But with the traditional one, displaying alphabetic information will be very difficult :(

Ceterum censeo: HP, launch a 15s (with a dot matrix LCD!!)

Walter


Does anyone know when the new 12C will be available in the U.S.? Thanks.


I do not know, but my guess would be that the new 12C will become available when the old 12C stock that is currently in the pipeline is exhausted. So encourage all of your real estate agent, mortgage broker and stock broker friends (who will not care about the new features) to buy up the old stock. (Oh yeah, the real estate agents, mortgage brokers and stock brokers are having a tough time these days.)


.

At least now we can agree on the model number :-)
Also the dot matrix display - even a 20b style - is a must


Peace for our time?

Though I must call for a *real* dot matrix display, i.e. 16x131 or higher, as you know ;)

The new 12C (with two coin cells) should get arithmetic results identical to the original 12C (one coin cell or three button cells) for all calculations, as it is using exactly the same algorithms as the original. If you find any cases where it differs, please let Cyrille and me know about it.

This is different than any version of the 12c Platinum, which uses entirely different code than the 12C, and may get different results. I'm not sure about the early 12c Platinum, but the newer versions (with parentheses and undo) should generally have higher precision than the 12C.


Quote:
The new 12C (with two coin cells) should get arithmetic results identical to the original 12C (one coin cell or three button cells) for all calculations, as it is using exactly the same algorithms as the original.
If you find any cases where it differs, please let Cyrille and me know about it.

Does this means that it's firmware is based on Nonpareil?

Congratulations if that's true, Eric!

Do you have any (financial) agreement with HP?


I suspect he can't say even if he hasn't got any contracts. Another thing is that the original interest rate calculations of the original calculator (which is now fatefully duplicated) is "better" than the more accurate Platinum version. ??? Because it is The Official Wall Street Calculator and it has to do the calculation exactly the same way - forever! This is surely gonna be the longest in-production model ever - even if it's now emulated on new hardware. If HP brings back the HP-15C+ redesigned with 32KB RAM and full alpha labels/matrix names from A..Z, I guess it could also be successful as a true pocket size scientific non-graphing calculator.


Quote:
If HP brings back the HP-15C+ redesigned with 32KB RAM and full alpha labels/matrix names from A..Z, I guess it could also be successful as a true pocket size scientific non-graphing calculator.

IF. But you'll need a different LCD for this, and recent communication doesn't support the assumption of such a capable (or call it normal) LCD :-/

HP bring back the HP-42...with SD-card for surveying
LCD should be graphing (18C/19BII, 28C/S)

Quote:

If HP brings back the HP-15C+ redesigned with 32KB RAM and full alpha labels/matrix names from A..Z


Why look a gift horse in the mouth?

Addressing more memory would be great, but adding "full alpha" and/or other bells and whistles to the original ROM might be asking too much. Why ask the developers to tinker with something that is already tested and true.

Some of us cannot afford eb*y HP 15Cs, so holding a new 15c+ in my hand (vs. clicking on an emulator), running the same firmware that so many have used and trusted forever, would be nice for me, and hopefully countless others.


Thanks,

PG


Pal,

Quote:
Why look a gift horse in the mouth?

Heh, I'd have not guessed to see a German idiom nearly literally translated in English ("Einem geschenkten Gaul schaut man nicht ins Maul") :)

For your other statement, I take the liberty to disagree. As I've said here earlier already, the 15C was a great machine in its time. Time didn't stop then, however, so we've seen further progress after the 15C, meaning models with advanced user interfaces. So, returning to displaying programs in keycodes would fall back behind the 32S. Returning to split complex objects and double stacks would do the same. It would be disappointing for the educated user, who experienced better things meanwhile. IMHO a relaunched 15C+ with a display of 1981 would confine the market to the community of vintage calc fans, while an LCD at least on the level of a 32S LCD will add some extra push to the C+ and gain new users.

Just my 20 Milli-Euros.

Ceterum censeo: ... (you know)

Walter


Quote:
Heh, I'd have not guessed to see a German idiom nearly literally translated in English ("Einem geschenkten Gaul schaut man nicht ins Maul") :)

I'm guessing multiple languages might have taken that from the Latin original Equi donati dentes non inspiciuntur. Have a look here

>>>in the horse's mouth<<<


Edited: 15 Nov 2008, 2:07 p.m.


George,

Thanks for the link. I didn't know it's going back to Noli equi dentes inspicere donati! It's all Caesar's fault ;)

Edited: 15 Nov 2008, 4:26 p.m.

Walter, I agree an improved 15c, like the 4-line LCD models you fathomed, and I modeled, would be incredible, powerful, and desired. However, all the talk about the "12c+" and a "15c+" makes me think HP might just use their new processor to emulate old ROMs. Since the 12c+ is presumably only getting a new PCB (but same display and chassis (w/more batteries)), but not new firmware (classic 12c only, please), why not roll a 15c+ with its own new PCB and famous 15c firmware (already written and waiting)?

Know what I mean?

Trust me, I would love a 4-line Voyager, but that may be dreaming..

Cheers,

PG


Pal,

Quote:
I agree an improved 15c, like the 4-line LCD models you fathomed, and I modeled, would be incredible, powerful, and desired.

Agreed. Though I only remember my 3-line 15s in Voyager dimensions, and don't recall your models right now (my bad memory).
Quote:
why not roll a 15c+ with its own new PCB and famous 15c firmware (already written and waiting)?

Absolutely no problem -- but *zero* progress as well. This would be just an edition for us greybeards as mentioned already, nothing to gain new users. So I'd rate this as a missed opportunity, though I understand some reasons for doing it this way.

Ceterum censeo: HP, launch a 15s (take the chance to pimp your 15C -- don't try to tell us you've got no ideas for improvements after 27 years of development in the world of electronics).

Walter

Edited to add a call for action.


Edited: 16 Nov 2008, 5:20 a.m.

Assuming that this is true and the 12C+ is running an unmodified ROM dump of the original 12C, that means that it's going to be back to 99 program steps/20 registers. Although that's probably fine for 99% of 12C users it seems like a step backwards from the 12CP versions -- and I won't be able to use all that extra speed to get 632 digits of pi :(

OTOH, I suppose when you go through all this effort to create such great algorithms it's worth keeping them.

Quote:
The HP 12c team concluded that they would design the calculations in the 12c to be so precise that the calculator could receive certification by the federal Bureau of Standards (now the National Institute of Standards and Technology). Hence, HP 12c’s results were legally accurate for the banking industry.

The algorithms used to perform calculations such as bond interest and partial payments on home mortgages were critical if the calculator was to be trusted by the financial world and meet the U.S. standards. HP consulted with experts from various countries to be certain that the calculator would work in markets using different methods of calculation all over the world. HP worked with William Kahan, UC Berkeley’s renowned professor of mathematics, electrical engineering and computer science, to develop and test the complex algorithms.


Edited: 15 Nov 2008, 1:48 p.m. after one or more responses were posted


Remember, modifying the 12c to add more program steps and algebraic is what gave us the problems with the 12cp.

I think better to have a 12c that is > 60X faster is better than a 12cp with issues.

My 2 cents (or what is it worth with the bailouts these days?)

Quote:
Assuming that this is true and the 12C+ is running an unmodified ROM dump of the original 12C, that means that it's going to be back to 99 program steps/20 registers. Although that's probably fine for 99% of 12C users it seems like a step backwards from the 12CP versions -- and I won't be able to use all that extra speed to get 632 digits of pi :(

OTOH, I suppose when you go through all this effort to create such great algorithms it's worth keeping them.


THAT is the difference between HP-12C(+) and "HP-15s". You just can't chamge the algorithms in the 12C, but 15C could become 15s, if not, then it's better to release HP-42 instead.

Quote:

Does this means that it's firmware is based on Nonpareil?


From the Nonpareil news page: "Voyager calculator models have been removed in release 0.79 due to licensing issues. They will be made available in a separate package in the near future." (August 23, 2008).

Coincidence?


Quote:


From the Nonpareil news page: "Voyager calculator models have been removed in release 0.79 due to licensing issues. They will be made available in a separate package in the near future." (August 23, 2008).

Coincidence?


I doubt it's a coincidence, well done Eric!

The original ASIC was no longer obtainable, so that either meant new software almost from scratch for a new platform, or an original ROM emulator already proven. Seems like a no-brainer decision to me.
Makes sense from HP's point of view for such a vital calculator as the 12C

The reasoning is not really the same for a scientific version (i.e. 15C), but if it's so easy, why not?

I'd be very surprised if Cryrille doesn't already have a new 15C prototype on his desk...

Dave.


Possibly Related Threads...
Thread Author Replies Views Last Post
  HP12c Present Value of $1.. Inaccurate??? Edward Dixon 12 645 11-12-2013, 11:32 AM
Last Post: Edward Dixon
  HP12C Platinum PC software activation fails Russell Clinton 0 187 07-02-2013, 02:32 PM
Last Post: Russell Clinton
  HP12C Limited Edition 30 anniversary keyboard Revan Ng 0 152 01-11-2013, 12:49 AM
Last Post: Revan Ng
  HP12C 'complete' collection Keith Midson 9 743 08-07-2012, 07:51 PM
Last Post: Keith Midson
  HP12c Anniversary Seft Test Nick Mihiylov 2 208 01-04-2012, 07:23 AM
Last Post: Koralatov
  The HP12c Anniversary edition has finally arrived to Europe Jose Gonzalez Divasson 1 161 11-12-2011, 08:16 AM
Last Post: hpnut
  hp12C capability Todd Israels 8 369 10-14-2011, 12:20 AM
Last Post: Todd Israels
  HP12c 30th AE keyboard Paulo MO 4 276 10-01-2011, 03:26 AM
Last Post: Paulo MO
  HP12C Article in the Financial Times Charles 4 235 09-24-2011, 06:54 PM
Last Post: Mark Harman
  HP12C 30th Anniversary Keith Midson 2 175 09-15-2011, 07:33 PM
Last Post: Keith Midson

Forum Jump: