HP Forums

Does anyone have any idea of what the Total Available Market (TAM) is for HP like calculators in either units for dollars? Perhaps there are some ex HP folks out there from the marketing side of the business that know these numbers? I am interested to know if the total units per year is over 100K, 250K, or 1M. Educated guess would also be appreciated. Specifically does anyone know how many units/yr of the following models are sold per year?

1. HP-12C

2. HP-19BII+

3. HP-48/HP-49

Don't read too much into these questions.

Best,

leb

Ex-HP employees would still be bound by confidentiality agreements (and ethics), and thus would not be able to divulge such information to you.

I doubt that there is any reliable source from which you can get such information without spending a lot of money to get it.

Quote:

---

Ex-HP employees would still be bound by confidentiality agreements (and ethics), and thus would not be able to divulge such information to you.

---

Ethics are another matter. I find it kind of bizarre that my inquiry has generated an ethics question from someone who has created a microcode simulator of HP calculators from HP patents. While it's legal and HP didn't copyright older ROMs aren't there serious ethical issues with that? Your homepage has a link to emu48 another HP calculator simulator that does use HP ROMs that DO HAVE HP copyrights and a number of these ROMS are available as a packaged file on SourceForge.

http://prdownloads.sourceforge.net/emu48mac/Emu48files.sit?download

http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/forum.cgi?read=75815#75815

To make my case. I am not really convinced that asking for older data about the size of the market based on memories is unethical. If I meet an ex-HP sales person at a trade show and I ask him/her, "How many calculators did you move back in '94?" I don't see that as an ethical problem. Having said that I certainly expect that those people who feel that it is an ethical issue would not answer anyway.

If this is going to turn into an ethical thread, I withdraw my post.

Quote:

---

I doubt that there is any reliable source from which you can get such information without spending a lot of money to get it.

---

Quote:

---

I find it kind of bizarre that my inquiry has generated an ethics question from someone who has created a microcode simulator of HP calculators from HP patents.

---

I completely fail to see the relevance. I used publicly available, public-domain source material. I'm not divulging any HP proprietary information to third parties. The information I do provide does not in any way affect HP's business interests.

That seems *completely* different to me than if I was an ex-employee of HP and agreed to give someone confidential sales information (even though it is possible that an employee confidentiality agreement might have expired).

Quote:

---

Your homepage has a link to emu48

---

That seems even less relevant. You're suggesting guilt by association?

Quote:

---

that does use HP ROMs that DO HAVE HP copyrights and a number of these ROMS are available as a packaged file on SourceForge.

---

As I understand it, HP granted permission for HP-48 and HP-49 ROM images to be distributed for use with simulators.

Quote:

---

I believe you also have participated in the creation of "skins" that mimic the look of HP calculators.

---

Yes. How is that a problem?

Quote:

---

Aren't you trading on HP's brand?

---

I don't think so. I'm not trying to sell anything, nor am I trying to convince anyone to use my software instead of buying an HP calculator. I don't expect that anyone would want to do that. I'm not trying to confuse anyone into thinking that Nonpareil is an HP product, and in fact I explicitly state that it is not on the web page.

I don't think my references to specific HP calculator models in any way serve to confuse consumers nor to dilute the HP brand name. Therefore I don't understand your apparent objection.

Quote:

---

Is that ethical?

---

Is it unethical?

Quote:

---

If I meet an ex-HP sales person at a trade show and I ask him/her, "How many calculators did you move back in '94?" I don't see that as an ethical problem.

---

I don't see asking the question as an ethical problem. I never intended to suggest that you have done anything unethical. But if I was an ex-HP employee, I certainly would not provide such information. I can only conjecture as to what reaction you're likely to get from actual ex-HP employees. (Actually, that's not quite true. I know several ex-HP employees who would have the precise figures you're looking for, and theoretically I could ask them whether they'd be willing to divulge the information, but I don't particularly feel comfortable asking that.)

Quote:

---

If this is going to turn into an ethical thread, I withdraw my post.

---

If you didn't want this to turn into an ethical thread, you shouldn't have impugned my ethics. I was only commenting on ethical behavior of a hypothetical ex-HP employee.

Quote:

---

In Silicon Valley if you can find the right person you can probably get it for the cost of dinner or a drink.

---

Finding the "right person" would be quite a challenge, especially in Silicon Valley. Most of the people who have those numbers live in the Pacific Northwest and Singapore.

I'm sure you can get someone to give you some numbers in that manner. but having any confidence in the numbers is another matter entirely.

Eric

Quote:

---

I don't see asking the question as an ethical problem. I never intended to suggest that you have done anything unethical. But if I was an ex-HP employee, I certainly would not provide such information. I can only conjecture as to what reaction you're likely to get from actual ex-HP employees. (Actually, that's not quite true. I know several ex-HP employees who would have the precise figures you're looking for, and theoretically I could ask them whether they'd be willing to divulge the information, but I don't particularly feel comfortable asking that.)

Quote:
If this is going to turn into an ethical thread, I withdraw my post.

If you didn't want this to turn into an ethical thread, you shouldn't have impugned my ethics. I was only commenting on ethical behavior of a hypothetical ex-HP employee.

---

It seemed to me your were questioning the ethics of me asking the question.

For the record, I did not [mean to] impugn your ethics, I just questioned them. If you or others took it the wrong way, I apologize. Peace.

Look. I have an idea for a new kind of "calculator". I have a hardware and software platform in development that would be perfect for this. It's would not be based on any HP technology although RPN, SOLVE, and supporting many of the HP functions is an important goal. The rate of progress with calculators has certainly slowed over the years but hardware, software, and development technology has advanced a great deal. The concepts, the software, and the UI has not really been rethought since HP designed it. It's all been incremental. The combination of much better hardware and much better software development tools allows the bar to raised substainly. The question for me is; "Is there a large enough market for this kind of product?". To be honest I have no idea.

Fancy calculators aren't as important as they once were. I know many people at various labs, schools, and businesses that used to have and use HP calculators every day and no longer use them or even bring them to work. Laptops can do a lot. Excel is great and I use it all the time. My HP 19BII auto leasing program is better in Excel than it is on the HP-19BII, although the HP is faster because of the dedicated variable and numeric keys.

To get reasonable pricing on things like LCDs and ASICs you need to have a run rate of at least 100K units/yr. Since it's very unrealistic to think a new company could capture 100% of the TAM, the TAM probably needs to be at least 500K units/year and you hope you can capture 20% and get a foothold. To be honest I don't think the market is that big, but maybe it is. It would be nice to know which of the 2 big markets is biggest business or scientific/engineering? I assume scientific/engineering but again I don't really know.

If anyone has any thoughts about this or would be interested in talking with me, I'd be interested in doing that in either in public or private. I do know about Qonos.

The biggest market for advanced calculators is neither business nor science/engineering. The biggest market, by far, is education.

TI began marketing graphing calculators in 1990, primarily to the education market. TI announced that their 20,000,000th graphing calculators was sold in 2000. According to the book "TI-83 Plus Graphing Calculator for Dummies," which was published in 2003, the TI-83 model alone was selling at a rate of more than 2,500,000 units per year.

The numbers for other markets, and for HP calculators, are not known, but are probably much, much lower. It seems likely, for example, that a significant fraction of all HP-33S calculators are sold to people taking US professional licensing exams in engineering and surveying (the 33S is the only RPN calculator allowed on such exams). But the total number of new examinees is less than 50,000 per year, and the number of 33S sales to such examinees can't be more than 10,000 to 20,000 units per year at best. If the professional exam market is indeed a significant fraction of all 33S sales, then it seems unlikely that total 33S sales are at the 100,000 unit per year level.

Now that's some useful insight for me at least. Thanks. 2.5M units x $150/unit = $375M for one product isn't too bad, eh? Now the question would is "How elastic is that price?" Could you capture 100K to 500K units @ $300 with a better product? In my experience better doesn't sell too well and price is really the big determiner which is how TI killed HP anyway. Most of my cost is in the LCD.

Quote:

---

To get reasonable pricing on things like LCDs and ASICs you need to have a run rate of at least 100K units/yr.

---

To get dirt cheap pricing you need that many units. And for an ASIC you might need that many units. But you can get decent LCD pricing and masked-ROM microcontroller pricing at much lower volumes than that. And if you're not trying to compete with low-end calculators, flash-based microcontrollers are probably inexpensive enough. I've seen custom LCDs in $100 appliances that are only made in about 5K/year quantities.

AFAICT, the main entry barrier to calculator manufacturing isn't the electronics at all. It's the tooling for the plastics. If you don't mind rubber keys with silkscreened legends, even that cost can be mostly eliminated. Of course, most of us here despise rubber keys with silkscreened legends.

Eric

IMHO, Eric's points in the previous post are correct, let me emphasize:

1. Use a mask-ROM off-the-shelf microcontroller instead of ASIC.
If you choose a manufacturer which has a compatible FLASH part,
you can start production with these and lower your risk.

2. There are so many small China based LCD manufacturers looking for
business you will find someone who makes your custom LCD at very low
cost (if ordered at least by the 1,000s). If you agree to pay all
NREs up front it is likely they will even make you a lot of 1000 or
even less.

3. Now the keyboard issue (hah !):

Except for HP, no pocket calculator manufacturer EVER has managed to build a good keyboard. The superior HP quality has several factors:

a) double injection molded keys with eternal and virtually indestructible legends that simply won't wear off

b) keys are hinged which gives them a unique feel of precision and quality (and increases operational life of the contact). They don't wobble under the finger. Just compare tactile key feedback of any Classic, Woodstock and Spice HP with a TI of the era (TI keys not hinged)

c) the snap-action contact strips themselves - not too much snap, just right to feel "easy", and not too soft (to feel "mushy").
Casio calculator keys always feel mushy, like poking your fingers
in soft s... (you guess what I mean).

d) the gold-plated counter-contacts on the keyboard pcb for reliable contacts

e) the overall mechanical assembly (with calculator shell) so the keyboard does not twist and bend.

f) plastic membrane under keys to protect contacts from contamination.

So far the factors which made HP calculator keyboards absolutely superior.

One historical anecdote: when the TI scientific calculators came out (and were somewhat cheaper than HP) the purchasing department of a major defense contractor tried to talk their engineers into buying TIs instead of HPs. These however already knew the poor reliability of TI keys (often leading to miscalculations) and responded: with a HP, I need to do my calculations just two times, get two identical answers, and can use the result in full confidence, while with TI, I need to do them five times or even more and then, among the different results, pick the numbers which appear more often, as they seem to be the most likely ones, all just due to keyboard bounce or keys not responding. (It was the 1970s when they made very long calculations such as center of gravity of missiles using these pocket calculators, because PC's and CAD software didn't exist).
Needless to say they got their HPs.

Bottom line: if you want to make a superior calculator for serious quality work, then you ought to have the best keyboard possible.

Before you find an answer how to manufacture such a keyboard (with the legendary HP look and feel and reliabilty) at affordable costs,
your calculator project most likely won't succeed.

-----> Good keyboard is the key to calculator success ! <------

(The only other road is to make cheap throwaway disposables, like other calculator manufacturers do, even today's HP (sigh), but I'd doubt this road leads to sustainable success)

Good luck,
Bernhard

Edited: 7 July 2005, 10:57 a.m. after one or more responses were posted

The 33s and latest 49g+ do have such a keyboard and I have no hesitation in using them. I feel the 41 a bit mushy though so I guess lots of people here won't trust me.

Arnaud

If you wouldn't mind elaborating on your ideas I'm sure everyone here would be interested. Please send me an e-mail, I may be able to lend some insight and/or assistance.

Best Regards,
HDE

Note that selling to the education market is *not* like selling to the business or scientific/engineering market.

You would have to convince teachers, professors, and administrators to adopt your calculator, instead of TI's. But education professionals are not simply interested in the capabilities of the calculator. They want to look at the entire educational package.

Can you supply dozens of student activity books and teacher guides so that calculator can be used in different classes and grades? Can you provide PC software that allows teachers to view the screen of any student's calculator? Do you offer teacher training on the use of the calculator? Do you sponsor annual teacher conferences?

TI does all these things. See education.ti.com. That's why TI has more than 95% of the graphing calculator market in the US (statistic is from "TI-89 Graphing Calculator For Dummies").

It isn't just a matter of price. Casio has been trying unsuccessfully to penetrate the US education market with cheap, capable graphing calculators for years (granted, they have been more successful overseas). HP isn't even a player.

Quote:

---

-----> Good keyboard is the key to calculator success ! <------

(The only other road is to make cheap throwaway disposables, like other calculator manufacturers do, even today's HP (sigh), but I'd doubt this road leads to sustainable success)

---

The only road to sustainable success, I would agree, with hope in my heart. It seems to me as though TI is selling baskets and barrels full of calculators, though I have no hard data to back it up. So for the education market, it may be enough for a machine to last one to four years. And what are students supposed to compare the keyboards on the throwaways to? If they can't touch a vintage HP calculator, they have no chance of judging the current crop fairly.

The "sustainable" qualification comes from the hope that someone will somehow find a way to engineer a production process that will allow keyboard quality comparable to the old HPs, but for a price point that doesn't rule out competing in the main calculator market. If they can do that, and produce a pricey high-end calculator that I would like to own with the profits from the good-quality bread-and-butter models, why I'd be delighted. Perhaps new materials will make this possible? Combine that with up-to-date electronics and brilliant software engineering in the spirit and tradition of Dr. Wickes et al, and my hopes for old age geekery would be fulfilled. 8)

Norris,

You are very right. The educational markey needs to vendor to romance the buyers (teachers and students in this case). TI has done a very good job, but it did not happen overnight. My wife sells calculators at a big college in town. Some professors are attached to specific models and will move to new ones with great difficulty. One case she told me, is a professor that hesitated to use the new TI-84 Plus because he was so used to the TI-83 Plus!!

In the HHC 2004 in San Jose I heard a presentation by Wlodek Mier-Jedrzejowicz about how HP tried to attack the educational market in a teachers' conference in Europe. Wlodek recounts how poor HP's efforts were compared to TI and Casio. I remember Richard Nelson addressing the folks from HP at the conference and commenting that it might be a wise move for HP to come out with an inexpensive HP-45-like non-programmable calculator that can be used to get teachers and students into RPN calculators. Of course if that succeeds (heaven forbids!) there is no law that says TI cannot make RPN calculators!!! That will be the day!!

Namir

To satisfy the fanatic/hobbyist market, design to a proven keyboard component -- say, the HP-41 keyboard. They're generally available, still durable, and should be easily incorporated in an otherwise brand-new kit-type product.

And if the product takes off and the price of used HP-41s goes even FURTHER through the roof, making a new keyboard substitute will suddenly become economically viable.

We would expect that quality of merchandise has a distribution function much as people have ... the majority being mediocre, with
a minority at each extreme being best or worst. It does not matter which characteristic considered for classification, be it beauty, IQ, muscles, wealth, etc., everywhere we find such a distribution
function. Real people are a mix of all this. The unique mix makes
people individuals. Each one is equally worthy (see the US Constitution), so don't misunderstand the above as a judgement.

From this, we would expect most products sold to belong to the mediocre variety. Not bad, but not good either. Not expensive,
but not cheap either. Just somewhere in the middle. As are most
people. A really great fit.

In the "good old times", this was true indeed. But now, we witness
some development where people either buy:

a) the cheapest product, or

b) the best product (in most cases, the most expensive)

This even applies to food. Class a) buyers literally will eat
s... or Soylent Green or whatever as long as it is cheap,
tasty (= artificially flavored) and saturates the stomach.

How come ? I really have no idea why. Fact is, the mediocre ("normal") products are crushed between these two extremes. Just go to your local shopping mall and watch what merchandise is offered
and what is really bought by the masses.

Manufacturers opting for a) must have strict cost controlling
as profit margins are so thin there is no margin for error...
as errors are inevitable (and would kill a small business on
the spot) this is for the big international corporations only.

Manufacturers opting for b) must have strict quality control.
They must offer the best quality technically possible. Period.
Profit margins (without NRE) are high and cost controlling
blunders are more easily absorbed. The catch is in the NREs:
any underestimation about the selling numbers may be lethal.

In my personal opinion (and vision for the future), the class
a) manufacturers will reign as long as energy and ressources
are abundant and cheap, and then they will go under. Class b)
manufacturers will face hard times (but many will make it) and
they will reign the post-industrialized era (I guess > 2025 AD).

Maybe then we will be able to buy brand-new Classic HPs again.

Their technology seems to be sustainable for countless millennia.

They are useful tools that don't eat much energy (vs. a PC), could run on solar power alone, and would last for 30 or 40 years. Then you replace the worn keyboard PCB the first time and continue to use the calculator for the next 30 or 40 years. How much do you need to pay for such a calculator ? One month of an engineer's salary ? No problem - just look how much time it will save (compared to calculating with pencil and paper) and how many decades it will last. Actually, it's a bargain.

Regards,
Bernhard

Yes, but I don't want to wait until after the flood, dammit! 8)
And of course I'm a cranky, spoiled boomer who helps drive the crazy economy of over abundance of which you speak.

All good points. I'd like to add another factor that I think is driving the throw-away/luxury culture besides overabundance. That's the rate of change that accellerated (2nd derivative of innovation?) during the last quarter of the 20th century. Future shock, and all that is a component. But so is the expectation that some marvelous and new gizmo will emerge next year, running on ever more capable hardware and incorporating more and more clever programming. This obviously discounts today's gizmo, let alone yesterday's. On the flip side, many people of means are willing to throw unheard of piles of cash at anything perceived to have lasting value against a background of unremitting novelty.

A real energy crunch is surely coming, I agree. That will divert the smarty pants of the world into looking for technical fixes for that problem. These may or may not be forthcoming, but the diverted effort will absolutely mean less glitzy new stuff. This will not only be because of the depletion of natural resources, but also because of the diversion of brain power to more urgent problems. But one beneficial side effect will clearly be a saner yardstick of value and quality, as you lay it out. Perhaps it won't be merely a side effect, come to think.

It's both fortunate and unfortunate for me, but I see the crunch happening later than you do. Fortunate, because I get to continue to ride the shockwave of change for the rest of my professional life, at least. I have grown up welcoming and adapting to change. It would be a change for me not to change, so to speak. Unfortunately, I will have to continue to cope with the skewed values engendered by the profligacy and genius of my age. Lack of a decent new calculator is nowhere near the worst of the consequences of that excess, but it's emblematic.

Edited: 7 July 2005, 7:02 p.m.

I'd like to thank everyone for their input. It is all much appreciated.

As far as the keyboard "issue" is concerned I just would like to say that I think understand that issue completely. I've been involved with HP products since 1973, I used to receive the HP Journal, and I have owned a variety of HP calculators starting with the HP-65. I understand double-shot molding, the hinged keys, the metals involved, and so on. I believe I know how to create a quality product with a quality keyboard and I understand the NRE that would be involved in doing so.

Quote:

---

If you wouldn't mind elaborating on your ideas I'm sure everyone here would be interested. Please send me an e-mail, I may be able to lend some insight and/or assistance.

---

I have a hardware platform based on a SOC that I have access to. I have a software substrate platform based on a variety of technologies that I don't want to reveal right now. The hardware platform is probably overkill even for a next generation calculator but it is has superb power management capabilities. I have a number of candidate concepts for deploying the platform and a next generation calculator is one of them.

It would be "nice" for someone to offer a next generation calculator platform. I wanted to explore the issues involved in in doing so. As is often the case it's not the technical aspects that are the gating factor but the business details. Most people are happy with what TI sells. The price is right. They control the education market. HP doesn't sell many units anymore and the current models are just incremental retreads of the previous models. The barriers to entry in this market are large and many.

Having said that, I have no desire to build a huge company. If I could sell 50K units a year and make $5M ($100/unit) I'd be happy. I could even be happy with 10K units the first year, but I need some volume to get good LCD pricing.

As far as what the system would look like I have some ideas on that but I would prefer to keep those to myself.

My email address is available here (remove the spam blocking characters) if someone wants to contact me. Please use the word "calculators" in the subject.

Thanks. -- leb