▼
Posts: 55
Threads: 23
Joined: Oct 2007
This comes to my mind when HP decided to sell its calculator emulator on iphone. It sounds great, as one can have a machine with all the good old calculators inside it. But wait, would you ever feel comfortable "pressing the keys" all the way on the screen? I wouldn't. Try entering the Mach number equation on iphone and I'm sure it will drive you nuts.
So here comes a solution: HP builds a "barebone" calculator, with the keyboard, with your favorite HP calculator form factor like the HP41C or HP15C or whatever. The keyboard will be free from labels, of course, much like the HP-41C Option 001. Then much like the iphone, the firmware/software can be sold seperately and downloaded via a USB port on the calculator. Finally, you can buy the appropriate "label" or "faceplate" and attach onto the calculator.
What's the advantage of this?
1. Easy bug fixes.
2. You can upgrade easily to another "calculator".
3. Lots of people will contribute to the calculator software, whether HP likes it or not, and this makes the calculator better and better.
4. The software can be tailor made for you.
5. HP can concentrate to build better calculator hardware such as better key press.
And disadvantages:
1. Surely you cannot bring it to most of the examinations.
2. Emergence of "calculator virus".
Sounds great and ridiculous, isn't it?
KC
▼
Posts: 1,248
Threads: 33
Joined: Aug 2007
Sounds only great to me. I don't think viruses would be an issue; not enough "return on the investment" (glee for causing someone else problems) for the virus-writers.
The trick would be to get HP to consider it a worthwhile project.
Oh, and choosing a form factor that would gain widespread acceptance.
Posts: 4,587
Threads: 105
Joined: Jul 2005
We're not so far away from this. Take the HP-20b as an example.
Though its tactile response as well as its display can (and need to) be improved for sure, an interface is given at least.
Ceterum censeo: HP, launch a 43S (on your way to it you may launch something blank as KC suggests, if you prefer, but a good keyboard and display are inevitably necessary, as had been said here many times).
Walter
▼
Posts: 850
Threads: 10
Joined: Mar 2009
Has anyone tried reporposing a 20B to a 42S, to replace, as it were, the aging hardware of current user's 42s's? Doing the same as what HP has done with the 12C+ could give a faster 42S with new harware. Just a matter of re-labelling the keys, I remember someone posting that they were struggling to remove the original lettring from the 20B keys?
▼
Posts: 1,278
Threads: 44
Joined: Jul 2007
The big problem with trying to do that, or to make any scientific out of the 20b is the screen. The LCD driver built into the chip will only support 400 segments, and they are all used on the 20b. It could be possible. . . but I'm not sure how it would turn out. :-(
RW
▼
Posts: 875
Threads: 37
Joined: Jul 2005
While not optimal, it seems like the display could be useable for a decent scientific calculator. The dot-matrix area in the upper left side of the 20b diplay could be used as a rudimentary second line, to display the stack-y contents and/or to display the imaginary part or angle for a complex number representation. Last year after HHC 2008 when the 20b repurposing project was active, I played around to see how numbers might be represented in the dot-matrix area. Below is what I came up with. Of course I have neither the knowledge nor the equipment needed to repurpose my 20b, so that's as far as it got.
Edited: 25 Sept 2009, 7:26 a.m. after one or more responses were posted
▼
Posts: 758
Threads: 9
Joined: Jul 2007
This LCD would be completely unsatisfactory for an HP42S type of calculator.
▼
Posts: 875
Threads: 37
Joined: Jul 2005
Hi Mike,
I agree that it would not be ideal. Just to be clear, (and sorry if you are aware), the 20b has a 12 character 7-segment (plus 3 smaller 7-segment characters for the exponent) display area on the bottom of the LCD, with the dot-matix area in the upper left, with 11 or so annunciators in the upper right. So I was not suggesting my crude character display for the main numeric output.
But in any case, I'd certainly rather have this:
▼
Posts: 850
Threads: 10
Joined: Mar 2009
Nice one Jeff. I had been thinking that the 35s form-factor with a bigger screen would be good.
On the subject of the 20B's LCD: I had looked at it and thought maybe to substitute it for a same physical size generic LCD which means finding another way of displaying the annunciators, or replace it with a custom made screen which might be viable if there were enough takers. But then this would defeat the whole object of repurposing existing hardware. I also wasn't aware of the LCD driver's capability limitations in the 20B's hardware. 'twas but a simple idea.
▼
Posts: 875
Threads: 37
Joined: Jul 2005
Quote:
Nice one Jeff.
Thanks!
Quote:
I had been thinking that the 35s form-factor with a bigger screen would be good.
For some reason I spent (or perhaps wasted, depending on your point of view) considerable thought and effort on that concept a while ago.
...
Posts: 758
Threads: 9
Joined: Jul 2007
Quote:
Has anyone tried reporposing a 20B to a 42S,
Sounds fishy. :-)
Quote:
to replace, as it were, the aging hardware of current user's 42s's? Doing the same as what HP has done with the 12C+ could give a faster 42S with new harware. Just a matter of re-labelling the keys,
IMHO, such alterations to existing hardware would be a big-time kludge of interest only as an experiment. How would the 20B display be made completely compatible with that of the 42S? What about generating 42S sounds, which can be important to prompt for input? What about IR printer support?
If I lost both of my HP42S units, I'd use my disdained HP32SII before falling back on a Franken-42S. Actually, I'd accomodate the large size and just use a 50G. But I've used one 42S daily for 12 years and it still looks and responds like new, while the other is yet new in the box. The 42S has kept its youthful appearance much better than my HP-15C did after only a few years of routine use.
But a new HP42S-type machine designed from the start in the HP20B-type hardware (but with sound and IR) would be interesting indeed. Much better than any mythological HP-15C re-issue, IMHO.
Posts: 24
Threads: 2
Joined: Oct 2008
Quote:
So here comes a solution: HP builds a "barebone" calculator,
The idea is good, but change "HP" for "someone else".
Quote:
5. HP can concentrate to build better calculator hardware such as better key press.
If HP has a real interest in developing better calculator hardware, how do you explain what has happened in last decade? Nobody convinces me that HP has a deep interest in doing it
Sílvio
▼
Posts: 1,248
Threads: 33
Joined: Aug 2007
Quote:
If HP has a real interest in developing better calculator hardware, how do you explain what has happened in last decade? Nobody convinces me that HP has a deep interest in doing it
Well, to be fair, HP did give us the 50g and 35s, not perfect but attempts to return to their roots, and the 12c series is still pretty good...
Posts: 1,830
Threads: 113
Joined: Aug 2005
I would have argued with this if Sam Kim had stuck around. You could at least point to quality improvements in the scientific and graphing models under his watch Now the question is open again in my mind.
Focusing on the high school market is presumably the right move for their business, but what sort of quality standards do you need to compete effectively there? What shortcuts in quality would make economic sense in that market and others HP competes in?
Regards Howard
Posts: 115
Threads: 16
Joined: Jul 2007
If I want to drive a beetle I don´t need a Mercedes Benz chassis. Will say the still expensive calculator hardware needed to implement and run advanced functions is a must for the typical 50g customer, but by far an overkill for the 10s type of guy. So where will the 10s mate buy his next basic scientific calculator? Would the 50g guy be happy to subsidize the lower end market fraction by paying higher prices than necessary?
Further, the 50g customer expects first class keys, the 10s customer expects nothing from the keys. The latter is used to the simple equation "cheap calculator = poor keys". Will he be happy to pay a higher price tag for decent keys?
Posts: 185
Threads: 20
Joined: Apr 2007
You mean something like this?
Actually I think the technology could be developed to place some kind of character display on each key and the keyboard so that the function labels could be dynamically changed electronically.
Then one would say, why not just have the flat touch screen like an iPod with the complete keyboard electronic? Two reasons. First as you say there is something inherently superior to a real keyboard. Second, a blend of real and electronic keyboard would have a cool factor - flat touchscreens are so yesterday.
▼
Posts: 887
Threads: 9
Joined: Jul 2007
Quote: You mean something like this?
I believe one of the early ideas or plans for the 41 was to make language ROMs available. That could still be done, couldn't it?
▼
Posts: 4,587
Threads: 105
Joined: Jul 2005
Quote:
I believe one of the early ideas or plans for the 41 was to make language ROMs available. That could still be done, couldn't it?
Do you mean replacing SIN (English) by SIN (German)? ;)
Posts: 528
Threads: 40
Joined: Dec 2008
I think HP could get 90% of the benefit of your proposal just by providing keyboard overlays for their calculators, just like they did with the HP 41's. I'm not sure which models currently offer user-assignable keyboards, but certainly the 50g does.
On future models, I think they could also make a handsome profit if they made the machines expandable (again, like the 41). Lots of us would pay to add cool extras (like a printer).
Posts: 291
Threads: 63
Joined: Jan 1970
This Message was deleted. This empty message preserves the threading when a post with followup(s) is deleted. If all followups have been removed, the original poster may delete this post again to make this placeholder disappear.
Posts: 1,619
Threads: 147
Joined: May 2006
Quote:
What's the advantage of this?
1. Easy bug fixes.
2. You can upgrade easily to another "calculator".
3. Lots of people will contribute to the calculator software, whether HP likes it or not, and this makes the calculator better and better.
4. The software can be tailor made for you.
5. HP can concentrate to build better calculator hardware such as better key press.
And disadvantages:
1. Surely you cannot bring it to most of the examinations.
2. Emergence of "calculator virus".
Sounds great and ridiculous, isn't it?
With the exception of #5 I get this with an iPhone and soon any number of touch screen devices such-as other phones, touch screen netbooks, tablet PCs, and eBook readers.
If you survey the landscape of calculators out there today and their respective users you'll find that key press doesn't even register as something of importance. However, even small markets can generate revenue. What is the size of the nit-picky calculator market?
I avoided the iPhone for a year because of a bias towards Blackberry style keyboards--rigid, inflexible, reliable. However I--as have many others--have adapted to touchscreen technology. The possibilities of virtual keyboards IMHO outweighs the static (even if you can put your own stickers on it). Look at the work done by i41CX+ and 42s. I like the dynamic labels, the on-demand QWERTY keyboard, the swipe for a 4 line stack, etc... I feel that as a whole it makes me more productive.
It's the end of an era, time to change.
▼
Posts: 1,248
Threads: 33
Joined: Aug 2007
Quote:
It's the end of an era, time to change.
Well, fortunately, I do not have to, since I already have sufficient physical HP calculators that I like.
Posts: 536
Threads: 56
Joined: Jul 2005
Hi Egan,
The issues of this thread overlap what I am planning to talk about in Denver next month, so I won’t comment.
However, your point on virtual keyboards is interesting; I’ve been experimenting with free42 on the HTC touch pro 2. On this device and other latest gen handsets (like iphone), the calculator appears on screen almost life sized – at least big enough to actually press the virtual buttons with your finger.
One of the big drawbacks is that you can’t feel the keys. Consequently, I’ve been trying to get some tactile feedback going with the virtual keys. I don’t know if the iphone version does this, but it’s fairly effective (compared to nothing!). Phone people like to call it haptic feedback, but its not really because true haptic feedback involves fingertip sensation (would be nice to add!).
What Im doing is wiggling the vibration mass inside the phone. I’m picking a number for the duration that gives it about 1 spin or so. For me, its 25ms, for other handsets it could well be different.
Anyway, this is a real step forward for virtualised keyboards. It would be even better to have special hardware, like a mass that could move forward and back rather than rotate.
if you’re in Denver, I can show you and that goes for anyone else interested.
▼
Posts: 1,619
Threads: 147
Joined: May 2006
Quote:
One of the big drawbacks is that you can’t feel the keys.
That is because you were conditioned to expect to feel it. The next generation will not. Visual and auditory feedback may be sufficient and in some cases mandatory, i.e. a poorly designed mechanical keyboard can give the user a false positive--you'll find multiple threads of missed keystrokes on the 50g and 35s. I'd argue that most pocket calculator users rely mostly on the visual feedback of the LCD.
However that all said I firmly believe that there are applications where a real keyboard cannot ever be replaced. E.g. the weighted keys of a piano.
Being conditioned myself for tactical feedback keyboards I'll admit I do miss the feeling at times and that I cannot substitute certain keyboards with a touch screen, e.g. my computer keyboard--the ridges on the J and F are critical for touch typing at speed. I imagine that accountants with their 5 key feel the same way. Perhaps the difference lies in speed. The piano, computer keyboard, Blackberry, etc... need tactical feedback for speed. Most calculator and mobile phone users use a single finger and without thinking register each keystroke from the display--relatively slow work.
Quote:
What Im doing is wiggling the vibration mass inside the phone. I’m picking a number for the duration that gives it about 1 spin or so. For me, its 25ms, for other handsets it could well be different.
Anyway, this is a real step forward for virtualised keyboards. It would be even better to have special hardware, like a mass that could move forward and back rather than rotate.
if you’re in Denver, I can show you and that goes for anyone else interested.
Very interested. I think there is a high probability I will make it this year.
▼
Posts: 901
Threads: 113
Joined: Jun 2007
Quote:
Being conditioned myself for tactical feedback keyboards I'll admit I do miss the feeling at times and that I cannot substitute certain keyboards with a touch screen, e.g. my computer keyboard--the ridges on the J and F are critical for touch typing at speed. I imagine that accountants with their 5 key feel the same way. Perhaps the difference lies in speed. The piano, computer keyboard, Blackberry, etc... need tactical feedback for speed. Most calculator and mobile phone users use a single finger and without thinking register each keystroke from the display--relatively slow work.
What I have found for touch typing is that the positioning of the rows of keys on a slanted keyboard and audible feedback from the keys hitting the platen were more important than the tactile feedback. I went through that when I switched from a mechanical typewriter to an electric typewriter, and again to a modern computer keyboard. When I learned to type on a mechanical typewriter I could do forty to fifty words a minute with very few errors. I held that speed with an electric typewriter but lost it with modern computer keyboards.
In the old days we also had key punch entry machines. Operators never looked at the keys, often (at least as I recall) didn't even have a display, but could achieve phenomenal speeds and accuracy. Engineers didn't need to apply.
Posts: 55
Threads: 5
Joined: Jul 2007
Couldn't agree more with you, Egan.
Had a personal iPhone (plus almost every RPN calc that run on it :-) for about a year, and I got outfitted now for work with a Blackberry. The tiny keyboard is totally unusable.
I found out the best way use the various calcs for me was to hold the iPhone in the left hand and "press" buttons with the left thumb, while I can write with the right hand at the same time. YMMV.
Everybody has different needs and ways to use the machine, and the touch screen seems the easiest to use and customize. Look at the discussions on this forum and OpenRPN.org in the past on keyboard layout. Anybody can potentially make their own with a programmable device like iPhone, it's not limited by the hardware.
Edited: 19 Sept 2009, 1:10 a.m.
▼
Posts: 887
Threads: 9
Joined: Jul 2007
Does anyone have any numbers on how long a touch screen lasts? Sometimes I can hardly make the ATM work, and it's not even mechanical-- you can see the conductors for the capacitive sensors when the sun is at the right angle.
One problem with consumer products like the iPhone is that the product life is extremely short-- they're here today and gone tomorrow. If the battery goes out in a few years, there probably won't be one available to fit it anymore. The 41 OTOH was in production for over ten years IIRC, and even today, 30 years after its introduction, you can still get things like Clonix41, MLDL 2000, and of course the older products are available on eBay. I've had programs on mine continuously for over 20 years. In the consumer market you have to move all your programs to another device every couple of years and learn to use it.
▼
Posts: 1,619
Threads: 147
Joined: May 2006
Quote:
One problem with consumer products like the iPhone is that the product life is extremely short
I'd argue that most if not all electronic consumer products. My first HP (15C) is still in use and will probably continue to function longer than my 50g and 35s. They just don't make them like they used to. The market has changed to wanting change.
Quote:
In the consumer market you have to move all your programs to another device every couple of years and learn to use it.
Yep, I did this with my Palms, then Zaurus, then Windows Mobile, now iPhone. Hmmm... also did this with VCR -> DVD -> BlueRay and Vinyl/Tape -> CD -> MP3 and rotary dial -> touch tone -> voice dialing.
And, I also did this with my 15C -> 48GX -> 50g. "Couple of years" in this case was ~10 years.
Change is good. Nothing wrong with learning something new.
▼
Posts: 887
Threads: 9
Joined: Jul 2007
Quote: Change is good. Nothing wrong with learning something new
Sometimes. But when new doesn't mean better, it's a waste of time. I got a 50g, and totally lost interest after going through the paper manual. The 41cx meets my needs better. So does my 71.
Posts: 302
Threads: 34
Joined: Aug 2007
Neat idea, but given the cost of development for such an endeavor will keep it from getting off the ground. From some of the posts I've read from Cyrille it could theoretically be possible to get HP to make calculator hardware to the specs of the community... Bankrolling it would be the big challenge, though. How many hundreds of thousands of dollars can HP enthusiasts pool to make that happen?
I'll leave it at that, this could be an interesting thought experiment.
▼
Posts: 887
Threads: 9
Joined: Jul 2007
Quote: How many hundreds of thousands of dollars can HP enthusiasts pool to make that happen?
As one who has brought many electronic products to market, I am almost tempted to make such hardware myself, if I could give it the time. It would not take hundreds of thousands of dollars. For NRE (non-recurring engineering costs), figure a few thousand for a custom LCD (we looked into it 20 years ago and were quoted about $2K, and I suspect that greater efficiency today means the price has not increased much), or maybe better, use an off-the-shelf LCD to avoid the NRE, a few thousand for the keyboard (we looked into that too, with double-shot keys so the labels are actually in the keys like the HPs' of the early 80's so they won't wear off), a couple thousand for lexan overlays (we've done those), a few hundred dollars for the first run of PC boards (we've done loads of those, up to 12 layers, but 2 to 4 would be plenty for this), and a few thousand for the case molds. I lay out very dense PC boards all the time, SMT and thru-hole, and up to 500 parts on a single board. Add in the parts for the first run, and we're only up to about $20,000, plus my own time.
Edited: 22 Sept 2009, 1:51 a.m.
▼
Posts: 1,619
Threads: 147
Joined: May 2006
Quote:
plus my own time.
Much of that time is probably spent on software and documentation. I believe that is where the hundreds of thousands comes from. A single engineer's salary can easily exceed $100K US. And, you need to double that to get the actual carrying cost of an employee (at least in large companies).
I have a dozen incomplete personal projects sitting on my desk (I'll bet Eric has 100 :-). Time is a rare and expensive commodity.
▼
Posts: 887
Threads: 9
Joined: Jul 2007
I was figuring that the community of enthusiasts would do the software. I'm happy on a fraction of your $100,000/yr though.
▼
Posts: 302
Threads: 34
Joined: Aug 2007
I've pondered this issue before. Looking at the body of work already available from nonpareil, free42, x48, to name a few. Creating hardware capable of running said software would be relatively trivial.
Posts: 302
Threads: 34
Joined: Aug 2007
My quote of hundreds of thousands was in response to HP taking on the development of said hardware. Your figures are correct. LCDs in many cases waive tooling fees now, injection molding costs vary more than I'd like. As a cost cutting tool, I worked on a procedure for OpenRPN to accomplish the effect of double-shot keys using laser etching with a resin back-fill for key legends. It's a trade-off in terms of cost vs. flexibility.
If you ever want to pursue such an endeavor I would be happy to donate my efforts to see it happen. OpenRPN, for me, was always about bringing at least one more great no-holds-barred piece of calculator hardware to the market.
Posts: 8
Threads: 2
Joined: Dec 2007
It looks like TI doesn't like the idea of people running a custom OS on their hardware.
http://www.ticalc.org/archives/news/articles/14/145/145316.html
▼
Posts: 2,309
Threads: 116
Joined: Jun 2005
I imagine that a few of the reasons that TI isn't happy about installation of non-standard firmware on their calculators might be:
- risk of losing certification or approval for students to use the calculator model on tests, which would probably dramatically reduce sales
- support costs - if people have calculators with modified firmware, they're going to get some amount of support requests relating to firmware problems
- reputation - suppose someone buys a used TI calculator and doesn't know that it has modified firmware; if it doesn't work right it makes TI look bad
|