[43S]Top row poll (until Monday, 4.2.13, 13:13 UTC)



#2

Apparently, the following question is open still: Assume the bottom six rows of keys looking like introduced here recently. Then there are two major alternatives for the top two rows shown here in order of their appearance:

a) This covers the 'basic' trig and log functions in two primary menus, so you get the left display pressing TRIG and the right pressing LOG.

b) This follows the 'traditional' approach of the HP-42S with a populated row of 'top' functions which will be overwritten by each menu displayed and reappear after some EXITs.

May I ask for your opinions? Do you vote for a) or b)? (For the usual suspects: One vote per head, please!)

Thanks in advance for your participation.

d:-)

Edit: BTW, the key [>x y] is the same as [>REC], and [>r theta] is the same as [>POL] - both new labels save some space and add information.

And Gene is right: The advantage of B is you get six default primary functions more than with A. Press SIN and you get it - if no menu is displayed.

The disadvantage is you may be as confused as you may be with the top row of the HP-42S.


Edited: 31 Jan 2013, 4:15 p.m. after one or more responses were posted


#3

Absolutely a.


- Pauli

#4

A, zillion times. Whoever likes b should get WP34S.


#5

Nah, all voters for b should be sentenced to the salt mines or shot :-)

- Pauli


#6

Hmmh, if you're looking for a punishment, how about fracking?

d;-)

#7

B please.

#8

A
Günter

#9

A

#10

a) of course! ;-)

PS: I see that Pauli also voted "a", so you can stop the poll here - as developer his vote counts at least 100-times! :-)

Franz

Edited: 31 Jan 2013, 4:58 a.m.


#11

Franz,

if it helps you, I know some other developers who didn't vote yet.

d#-/


#12

Quote:
Franz,

if it helps you, I know some other developers who didn't vote yet.


I guess you mean Marcus, right?

Well, I'm quite sure he will also vote the only correct way, and that is "a". ;-)

Oh, I forgot - there's still another 'developer': Walter!

But since he only 'developed' the manual, I would count his vote only 50x (I hope you don't feel offended ;-))

Franz

#13

A

#14

A

(Hallo Walter, könnten wir in Sache dieses Projektes 'mal direkten Kontakt (gerne auch telefonisch) aufnehmen? Ich bin in diesem Kreis ja ziemlich neu und möchte gerne meine Möglichkeiten ausloten, wie ich diese Entwicklung (sofern Bedarf besteht) unterstützen kann. Beste Grüße und Wünsche von der physikalisch-mathematischen Forschungsfront nahe Marburg! Martin)

#15

A - because having a key labelled "sin" (for example) when it may do something completely different is confusing. If it can be avoided, it should be.

Nigel (UK)

#16

A for me.

#17

A

#18

A.

#19

A

(wobei ich mich frage, wofür soll das denn sein? scheint ich habe was verpaßt ...)


#20

Oooch, macht nix. Wir sind nur gerade dabei, den Nachfolger für den WP 34S zu entwerfen. Stöber mal ein bisschen unter dem Suchwort 43S ab Oktober.

d:-)


#21

auf welcher basis?


#22

Die Basis basteln Eric und Richard, Schlagwort DIY5.

d:-)


#23

wow

#24

B

#25

While (A) may be better in concept, I believe (B) would be better in actual use.

My vote is for (B).

-Bill

#26

"A" for sure.
It's sad but unavoidable that the H.MS & H.d functions burn up another line on the screen but they need to be grouped where they are used.


#27

Don't be too sad - it seems we'll have sufficient display space.

d:-)

#28

I prefer some option C lol. :-)

Keystroke efficiency is important. I *like* soft menus to bring functions to the keyboard and out of long long catalogs if possible.

But not at an expense of keystroke efficiency.

Option B puts more functions on the keyboard with access in 1 or 2 keystrokes.

Option A takes functions off the keyboard and makes more functions take 2 keystrokes compared to 1 and some take 3 now rather than 2. For a lot of manual use, those extra keystrokes will add up.

Why not option B with the same 3 level softkey menus for functions that are not on the keyboard?

Is the real issue whether to have the top row labeled F1 through F6 or to have functions printed on them?

If the arrows are present, I **really** believe most people who would use this can figure out what is going on.

So, I want some option C which is option B WITH some menus added that bring out commonly used functions and show them like option A does.

I know... I'm a problem. :-)


#29

Did you really think the menus would vanish with option B? Be assured, they won't. They will look different, that's also sure. But menus will behave in A like in B.

Everyone owning a scientific Pioneer knows the situation. Sometimes there are softkeys on the screen. Just take the HP-42S (it was mentioned in the OP): sometimes the key LN doesn't act as such but calls some other function as indicated in the softkey line on display. That's almost the same functionality as built in the HP-27S though this has arrows printed on its top row. And that's the only difference between A and B: one shows particular functions printed on the top row, the other one shows just a neutral print.

I admit I didn't expect that being such a difficult topic as it appears now. Please, everybody who didn't dare to ask yet, ask now whatever you want to know. I will try to answer according to my present knowledge, but I cannot know what you don't know (or presume or guess or ...).

d:-)


#30

No, I didn't think they would disappear. Not at all.

I am simply trying to point out that option A will require more keystrokes to access the functions in the menus compared to option B for those same functions. Logs and Trigs are used a alot, so that extra keystroke each time will add up, IMO.

I would like option B but with even MORE functions beyond those on the keyboard put into some add'l menus. Think the X^3, etc. Those are not on the keyboard, but it really seems like Sin, etc, should be.

2 cents. :-)


#31

Perfectly understood, Gene, and I didn't mean you (sg.) but you (pl.) - the ambiguity of the English language caused mathematical and scientific literature (i.e. text) being written in Latin (or similar languages with an elaborated grammar) for long time.

d:-)


#32

Ah, in the northern USA, that's "you's guys"

In the southern USA, it's "ya'all"


#33

Pittsburgh, PA: Youns. (or is it spelled yoons?)

As for the poll: I grew up with an HP-25, so I'm used to the basic trigs and logs being f and g shifts on the number keys.

The suggestion for keeping the top row keys "soft only" and the menus "permanently on display" really appeals to me. I especially liked the idea that was mentioned of a "user" menu that you (each user) sets up to do your (user's) work best. (Sorry, I forgot who posted it, and you <poster> did it much more eloquently than what I just typed. Thank you <poster>, whoever posted it.)

Walter, I know you (sing.) and you (developers pl.) wanted to avoid g-shift functions on the number keys. As for me: I'd prefer "A" with the basic trigs (sin, cos, tan, their inverses) and basic logs/exponentials (base e and 10) as (I know I'm gonna hear back on this!) g-shift of the number keys. (Ten functions, ten number keys.)

As we say in the U.S.: "So shoot me." (Or is it "So sue me."? -- Either way. We've got lots of guns and lawyers here. The latter currently arguing loudly about the former....)

Me: I don't use hyperbolics or logs/trigs with bases other than e or 10 enough to justify putting on the keyboard. Access to these via menus is sufficient for my purposes.

Again, the above is influenced heavily by the fact that I got an HP-25 instead of a car for my 18th birthday. (So was my career choice.)

That's MY personal usage / preference.

You (Walter) asked.

Dale


#34

Quote:
The suggestion for keeping the top row keys "soft only" and the menus "permanently on display" really appeals to me. I especially liked the idea that was mentioned of a "user" menu that you (each user) sets up to do your (user's) work best. (Sorry, I forgot who posted it, and you <poster> did it much more eloquently than what I just typed. Thank you <poster>, whoever posted it.)

Dale, that was my suggestion here:

http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/forum.cgi?read=238240#238240

Unfortunately it seems that either this posting has been overlooked by most members, or it has not been understood what or how I meant it. :-(

The keyboard layout "A" together with my suggested "user menu" has all advantages of both choices ("A" and "B") and this permanent user menu can additionally be filled with all functions that any user would like most (even 18 instead of only 12 for "B"). And of course there's no need for any additional keypresses - if this user menu is filled with the same functions that are assigned to the keys in the layout "B", then one single keypress would give all standard math functions like e.g. SIN, COS, TAN, 10^x, e^x, y^x (and their inverses by the prefix [f] and 6 other ones by [g]).

If it would be made this way then this poll won't be necessary at all, because it's definitely the best option and combines all advantages of "A" and "B", can even be user-defined and has no disadvantage at all.

But as I already said: it's not my business anymore ...

Franz

Edited: 1 Feb 2013, 9:21 a.m.

#35

Quote:
I am simply trying to point out that option A will require more keystrokes to access the functions in the menus compared to option B for those same functions. Logs and Trigs are used a alot, so that extra keystroke each time will add up, IMO.
2 cents. :-)

That is plain wrong. ASIN for example needs 2 keystrokes from default state. One "extra" keystroke to invoke TRIG gives you single keystroke access to SIN COS TAN and their inverts which is obviously more efficient.

#36

Ah, you are correct on the ASIN...but let's try some math, shall we?

Option B: 1 keystroke for SIN
Option A: 2 keystrokes for SIN

Option B: 2 keystrokes for ASIN
Option A: 2 keystrokes for ASIN

So certainly any functions that are primary on B will take more keystrokes if they disappear from the keyboard and are placed in a menu. Agreed?

Now, obviously, these menus are subject to change, but given exactly as Walter has drawn them out, consider the function ->HR

Option B: 1 keystroke
Option A: 3 keystrokes (TRIG blueshift F6)


1) I believe that there are a number of circumstances where Option A would require more keystrokes. I have shown some here.

2) I haven't yet seen a situation where Option B requires more.

3) There are other circumstances where Option A and Option B require the same number. You showed those.

So, on average Option A will require more keystrokes.

Help me out if this is not correct? I think it will require circumstances on what we have been shown where Option B requires more keypresses than option A.

Remember, Option B can still have menus to bring up X^3, etc, which show up on menus on Option A.


#37

Quote:
2) I haven't yet seen a situation where Option B requires more.

Here is one: ASIN ACOS ATAN is 4 for A and 6 for B


Edited: 1 Feb 2013, 8:42 p.m.

#38

Quote:
So, on average Option A will require more keystrokes.

Help me out if this is not correct?


On average ONLY if the functions on the keyboard are the ones one uses most. Polls here showed there is no consensus about that.

Secondly, keyboard efficiency is not all about number of keystrokes given the fact that they wouldn't differ significantly anyway. It's more about logical, intuitive and uniform approach and clear, clean interface. That's why most of the participants here are looking out of the square of their own preferences and vote A.

#39

Also remember that anytime you have a menu already displayed, option B will require a keystroke to clear it before you can use the top row of keys for their printed functions. In this case B can actually be worse than A. e.g. SIN-1 could take three key presses -- one to clear the existing menu, then shift and finally the function. Here option A only requires two.

- Pauli


#40

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.

#41

Walter-

I'll take your dare and ask a question (or two).

What happened to 10^x? I know that I suggested it might be a waste, but somebody pointed out that calculations involving decibels use it heavily. I had forgotten about decibels, so I think your "weekend pleasure" layout is the better option B approach.

Also (and not to argue against myself again), option B would probably benefit from a TOP.FCN menu similar to the HP 42S. That way the regular menus can be persistent while still maintaining reasonable access to, for example, complex number entry. I think the 42S handles this fairly gracefully. Your thoughts?

-Bill


#42

Quote:
What happened to 10^x?

10^x (and LOGxy) was dropped following your reasoning :-) The layout is a default setup - it shall please as many as possible but won't please all. The latter isn't dramatic since every user may reassign everything. Yes, 10^x is nice to have as default but not more IMHO. The HP-35 didn't feature it ...
Quote:
... option B would probably benefit from a TOP.FCN menu similar to the HP 42S. That way the regular menus can be persistent while still maintaining reasonable access to, for example, complex number entry. I think the 42S handles this fairly gracefully. Your thoughts?

Yes :-) And we certainly have to think over the whole menu business in detail as soon as we have some toys to play with.

d:-)

#43

Quote:
Please, everybody who didn't dare to ask yet, ask now whatever you want to know.

Concerning the 43s-project I find myself very much in the midst of a fog. I find some information here and there popping up in some posting or another, being virtually contradicted in a third. It seems to me many of these information bits are taken as a given by many, only that I can't piece them together. So to make a long story short:

Please, Walter, could you write up a list of specifications that are certain, expected, wished for, debated, dreamed of etc. Hardware, software, possible costs, ways of purchase, whatever. I would be grateful!

Edited: 1 Feb 2013, 3:33 a.m.


#44

Hallo Alexander,

I understand your foggy feelings ;-) Seriously, kind of what you're requesting can be downloaded here (for 5 weeks). It's less than 5% of the WP 34S documentation yet, and it's a bit outdated. I'll update it when we know about the top row.

Quote:
... could you write up a list of specifications that are certain, expected, wished for, debated, dreamed of etc. Hardware, software, possible costs, ways of purchase, whatever.

  • Certain: Nothing will be available unless Eric and Richard finish their HW work at DIY5 :-/
  • Expected: When they do, the device will have 8 rows of keys in 6 / 5 columns as shown. I've to dig for the specs of the dot matrix LCD and the outer dimensions. Anyway, the display will allow for softkeys :-)
  • Wished for: Everything, as you can read on this forum. Even a calculator guessing the right command after keying in a sequence of characters ;-)
  • Debated: Pretty much - again, you'll see most of it on this forum.
  • Dreamed of etc.: Please see 'wished for'.
  • Hardware: Please ask Eric and Richard for that.
  • Software will be based on the WP 34S project. Pauli may be able to tell you more but we're really right at the beginning.
  • Possible costs: Frankly no idea yet. I even don't know the HW costs.
  • Ways of purchase: dto.
  • Whatever ... else do you want to know?

d:-)


#45

Quote:
Wished for: Everything, as you can read on this forum. Even a calculator guessing the right command after keying in a sequence of characters ;-)

I "guess" that was an indicator related to my question about a real command line. If yes: The HP 48 doesn't have to guess, because it uses a command line parser, like virtually every computer featuring a command line.

Even with autocompletion (of which I did not talk in the other thread, but which your CAT seems to feature internally) the machine does not have to guess.

#46

  • Display: B&W, 400x240, 2.7" diagonal (about 173 dpi)
  • I/O
    • USB (device only)
    • serial (maybe)
    • infrared, output only, 82440 printer protocol
  • CPU: ARM Cortex-M3
  • Speed:
    • 32.768 kHz watch crystal
    • 1, 7, 11, 14, 21, or 28 MHz RC oscillator
    • 48 MHz crystal for USB
  • Memory: 1MB flash, 128KB SRAM
  • Mass storage: 128KB FRAM (maybe), MicroSD card
  • Double-shot injection molded keys: probably not, unfortunately.
  • Price: unknown, but very expensive by today's calculator standards. Possibly around USD 300. Possibly even more.
  • Ways to purchase: direct sales
  • When: unknown

Prototype hardware exists and has been demonstrated at HHC 2012 running Free42.

It should be noted that the flash memory will have no wait states at frequencies under 16 MHz, one wait state from 16-32 MHz, and three wait states above 32 MHz. This means that running at 21, 28, or 48 MHz does not perform that much better than 14 MHz, though drawing significantly more power. 14 MHz will be the default except when USB is connected, since USB requires 48 MHz.


Edited: 1 Feb 2013, 1:04 p.m. after one or more responses were posted


#47

A price of USD 300 may seem expensive, until you compare the ratio of the cost of the HP calculators in the 1970's to the average monthly salary in the 1970's. Even at USD 300, the cost ratio compared to the current average salary is much cheaper - unless you're still making $3.50/hr. Personally, I would defer other discretionary expenditures and save up the money to buy one at a cost of USD 300, just because I think it would be cool to have one. That's why I bought a WP-34S and a DM-15cc.

#48

Quote:
Walter - "I've to dig for the specs of the dot matrix LCD and the outer dimensions. Anyway, the display will allow for softkeys :-)"

Eric - "Display: B&W, 400x240, 2.7" diagonal (about 173 dpi)"


So something like this should be possible with the display?:


#49

I just scaled your image to match key sizes of the WP34s and the font on the screen looks too small, almost unreadable. I hope it is better in reality.


#50

Have no fear, I was only illustrating a concept, not suggesting a font size. I made no attempt to optimize or maximize. Rather than scale to the wp34s key size, it would probably be better to scale the display diagonal to the 2.7" size specified by Eric.


#51

Yes, I did that too, still...

Thanks for your rendering.

#52

Quote:
So something like this should be possible with the display?

Exactly: something - like - this ;-)
#53

Walter and Eric,

thank you very much for taking the time. Amazing, indeed! But I guess, if $300 is a realistic figure, I will remain an admirer but not a purchaser.

#54

Another comment on "ways of purchase". I don't know that we'll be able to offer them for sale outside the US, because of the cost of required conformance testing. We have to do FCC testing for the US, but if we have to do it for the EU and for other countries, the retail price of the calculator will go up noticeably, since we expect this to be a low-volume product and can't amortize the conformance testing over a large number of units.

Of course, if we only sell them in the US, and someone else were to buy one and ship it abroad at their own risk without telling us, we wouldn't know. Naturally we'd never approve of such a thing, and would advise anyone not to do it.

Edited: 1 Feb 2013, 3:12 p.m.


#55

Quote:
Of course, if we only sell them in the US, and someone else were to buy one and ship it abroad at their own risk without telling us, we wouldn't know. Naturally we'd never approve of such a thing, and would advise anyone not to do it.

I never knew that one would need approval (from the previous owner) to resell goods paid for in full. What "risk" are you talking about? Being put in jail for sending $300 calculator abroad? Lol

#56

You don't need any approval from the previous owner to resell goods you own. That's a form of the "doctrine of exhaustion"; once you've paid for the good, I have no further say in what you do with it, unless we've signed a contract to the contrary.

However, if you were for some strange reason to ask me for permission to ship a product you purchased without a CE mark to an EU country (to choose an arbitrary regulatory domain for exemplary purposes), I would not grant it.

The risk is that the customs people might seize a product without the necessary certification on import, and I'm not willing to accept liability for that. In other words, if you do it without my knowledge or consent, and thus not acting as my agent, it's your $300 at risk, not mine.

#57

Hmmh, a bit early for such kind of thoughts IMHO. But since you started it - how about distributing kits for DIY assembly over the world? Noone can prohibit I might assemble a calculator from some plastic parts plus some electronics (plus some locally supplied screws) for my own use ... but IANAL.

d:-/

Edited: 1 Feb 2013, 5:14 p.m.


#58

Actually, conformance testing is a serious legal concern even regarding prototypes that I provide for development, let alone anything shipped for revenue.

I don't know about the laws in the EU and elsewhere, but if you were in the US and I provided you with a prototype for development, I couldn't charge any money for it, and it would have to bear a label stating "This device has not been authorized as required by the rules of the Federal Communications Commission. This device is not, and may not be, offered for sale or lease, or sold or leased, until authorization is obtained." (47 C.F.R. § 2.803(c)).

The chances of the FCC finding out that I sent a prototype calculator to someone and accepted payment or omitted that label are small, but the penalty for being caught is high.

#59

Contrary to the popular wisdom, selling something as a kit does NOT put you in the free and clear with regard to conformance testing, at least in the US. If it is found that the device is inherently non-compliant, you can still face hefty fines.

#60

Quote:
Hmmh, a bit early for such kind of thoughts IMHO.


Not early enough, IMHO. I would think that all the non-US contributors all over the world would considerably lose interest if they weren't able to purchase the final product.

#61

Shall we set up an assembly line in good ol'Europe?

d;-)


#62

Columbus's egg stories will be there forever

#63

It doesn't so much matter where they are produced. It's a matter of where they are sold. If someone wants to be a distributor in a region, and pay for compliance testing for that region, I'm not opposed to it.

#64

May I humbly ask if ever a poll was made on how much potential buyers would pay and what the outcome was?

My upper limit would be 100€. Tax and postage included.


#65

I think there's little point in having a poll. The price of the calculator is largely determined by the manufacturing cost.

I could certainly make a calculator that would sell for 100€. However, it wouldn't be *anything* like the 43s being discussed. It would be more like the 34s.


#66

Quote:
The price of the calculator is largely determined by the manufacturing cost.

Is this hardware really so much better than e.g. the HP-39gII?

Won't it be much cheaper to use an existing powerful calculator (like the 39gII) and simply write a new ROM for it, i.e. repurpose it as it was done with the 20s/30s -> WP34s?

Franz


#67

Hardware costs depend on scale effects. Small batches are expensive. Repurposing is a lot cheaper.

Alas, AFAIK there is no calculator HW on the market which can be repurposed for the 43S. Even if we take the electronics out. The HP-17bii Silver comes closest in size and keys, but features just 7 rows and a small display window. The HP-35s has 8 rows of nice keys but the cursor cross would allow for a row of 4 softkeys only - suboptimal (and has the same small window). The HP-39gii is simply too large - a sledgehammer, no pocket calc. That's real life.

d:-(


#68

Quote:
The HP-39gii is simply too large - a sledgehammer, no pocket calc.
And the HP-48/50g? Also too large?

Well, I guess you simply want the "Mission impossible": lots of keys and a big display, but a small overall size. ;-)

Franz

Edited: 2 Feb 2013, 2:24 p.m.


#69

A repurposed smartphone maybe? Downside is probably the short battery time, but smartphones are readily available and of course insanely powerful in comparison.

Plus, since they are essentially linux pc, infinite software is available to build on.

Plusplus: it even has a double wide enter key!

Edited: 2 Feb 2013, 3:31 p.m.


#70

For sure a nice display. The keys, however, and the overall keyboard are ... hmmh ... abysmal for our purposes. Sorry.

d:-(

#71

Franz,

As you can clearly see on your 119th mission on this friendly forum, I'm doing everything to shoot this project down - and that for months #-( Yes, IMHO also the HP-48/50g are made to impress fellow students but not for serious work of professionals beyond an academic environment. And I don't like the keys of the recent calcs of this family. YMMV, but you're not designing this device.

d:-I


#72

Quote:
Yes, IMHO also the HP-48/50g are made to impress fellow students but not for serious work of professionals beyond an academic environment.

What do you mean with serious work of professionals? I have a PE (as in Professional Engineer) and I use the HP48GX in my daily work (not in academia), would that make my work less serious? And for your objection to the calculator size, just buy a shirt with a bigger pocket.

#73

The longer I look at it the more I think it's a damn pity that HP cut that cursor cross into the keyboard of the 35S. Else it could have been an almost ideal host or at least parts donor for the 43S :-(

Jeff, do you still remember how to open that device? I don't want to destroy it permanently yet ...


#74

Sure. Paraphrasing this old post:

Quote:
Remove the rubber plugs in the four corners of the battery compartment, then peel off the rubber strip on the bottom of the calculator, and remove the six screws. The front and back halves of the case are also held together by five plastic catches, one in the middle on the top edge, one in the middle on each side, and one near the bottom on each side. These catches are not exceptionally tenacious; some gentle tugging at the case halves and careful prying between them will fairly easily separate the front and back. It certainly appears that the PCB is held in by the 25 small screws, but after their removal, inspection will reveal that the posts into which the screws are screwed are mushroomed over at the top, just like the heat stakes in other HP calculators. I found it necessary to trim the posts off flush with PCB, and with a little gently prying, was finally able to free up the PCB. Once the PCB comes out, next is a thin rubber membrane, then the keys themselves. The keys are in two groups with a number of keys connected to a plastic frame, with the four cursor keys adhered to their own separate rubber membrane. The key contacts are adhered to a Mylar or some other type of plastic sheet stuck to the top side of the PCB, very similar to the 30b if you have ever fully disassembled one of those.


Edited: 3 Feb 2013, 12:42 p.m.


#75

Thanks a lot, Jeff! I remembered it was a long time ago when I read your post.

d:-)

#76

... this original drawing of 2007:

Not too far away from what may be viable today.

d;-)


#77

bx ???????


This is one I wish we could have repurposed.


- Pauli


#78

Quote:
bx ???????

b = base, see also f-shifted SQRT and g-shifted yx.

d:-)

#79

Quote:
39gII
The 39gII may have nice and powerful innards and LCD, but it has at least one big disadvantage: The awful keyboard layout.

I wouldn't buy a calc with that keyboard layout.

The 43S looks promising, and it has a much better physical keyboard layout. If the 42S housing doesn't have the slanted and "sharply shaped" edges as on the 30B, chances a good for me buying a unit:-)


#80

Quote:
If the 42S housing doesn't have the slanted and "sharply shaped" edges as on the 30B, chances a good for me buying a unit:-)

If you're talking about the 43S - AFAIK it will look more like a Pioneer than like a 30B.

d:-)

#81

I'm definitely not planning to copy the physical styling of recent HP calculators.

#82

According to people in the know, the 39gII boot process (determined by the chip vendor) is absurdly complicated, and documentation is not publicly available, so it will be difficult to figure out how to build replacement firmware that will work on it.

If I had all the time in the world, I'd be happy to reverse-engineer the 39gII hardware and get something working on it. Perhaps someone else will do that.

I fully expect that the price of the hardware Richard Ottosen and I are building will greatly limit its uptake. I'm more interested in building the best calculator I can, than building a bargain basement device that's not much better than what's already out there.

For those people that aren't willing to spend a fair bit of money on a fancy calculator (though still *MUCH* less money, adjusted for inflation, than HP's early calculators), I suggest using the WP-34s (based on the HP 20b or 30b hardware) instead. The WP-34s firmware is absolutely amazing, given the limitations of the hardware.

#83

Don't call it poll then - but market research (or whatever the correct term is), which you hopefully have done before investing money. I wouldn't produce any product if there were too few customers to make it profitable.

I for one wouldn't even pay $300 for a smartphone, let alone a calculator, no matter what it could do. But I surely wish you all the luck and high profits!

Edited: 2 Feb 2013, 1:41 p.m.


#84

It doesn't matter whether it's called a "poll", "market research", or something else entirely. The fact of the matter is that regardless of the outcome of a poll, I can't change the selling price of the calculator to the under $100 price you want, because I would lose a *lot* of money on each unit.

Without doing a poll, I think it's safe to say that more than 99.9% of the people in the world won't pay more than $150 for a calculator (the approximate upper bound of the pricing of mainstream graphing calculators). Probably more than 99.9999% won't pay $300. Nevertheless, that has no bearing whatsoever on my project. I'm not doing this as a mass-market product.


#85

Even the WP 34S is far from being a "mass-market product", although it may cost just an HP-30b (<40 US$) plus a cable, an overlay, and a bit of time. Maybe some 300 WP 34S are in the world. That's the feedback from our project site at least - I don't have any better figures, and those won't be off by an order of magnitude.

d:-/


#86

My assumption has been that Richard and I should plan to sell [redacted] units over perhaps two years.

Edited: 5 Feb 2013, 11:51 a.m. after one or more responses were posted


#87

I'd buy one if layout A is chosen. Even it is $500. As a token of my appreciation of the efforts put in.

*Many* people wouldn't buy Ferrari or Lamborghini because Toyota does the job too, besides the latter is affordable. That doesn't mean the former ones don't sell, in lesser numbers.

$300 for state-of-the-art, unique and boutique calculator is *cheap*. Any one of my ute tyres is more expensive.


#88

Quote:
I'd buy one if layout A is chosen. Even it is $500.

In this price range, there is a world market for perhaps 5 calculators, I would say :-)
I would pay $150 at most, if B is chosen (up to $250 for the original B version).


Quote:
$300 for state-of-the-art, unique and boutique calculator is *cheap*.

Sorry, but since 1987 no calculator with keystroke programming can be considered state-of-the-art anymore.


#89

Quote:
I would pay $150 at most, if B is chosen (up to $250 for the original B version).


Ok, let's up the ante: I'd pay up to $350 for the original (- + * / on the left side)...

:D :P

Massimo


#90

:-D Hey folks, please let this poll finish before you start the next!

d:-b

#91

This is where difference is. I put my money where my mouth is. I stated I'd buy ONE HP15c and I did. Thousands lied. I never trusted them and made it clear.
If someone struggles paying $300 for a state-of-the-art never-to-be-repeated gem, too bad for him.


#92

Ditto, Reth.

#93

Quote:
This is where difference is. I put my money where my mouth is. I stated I'd buy ONE HP15c and I did.

Same here, rest assured. I signed the petition and said I'd but three HP-15C and so I did:

#94

Heh... I said I'd buy one, but then I bought *two*... so I offset the others a bit! :)
Still, $300 (plus shipment plus import taxes and so on) for a calculator can be steep for many people, especially in some countries. No matter how state-of-the-art it is, it would still be not much more than a "nice toy" for me. I *hope* I'll be able to afford one, but I won't know for sure until the time comes...

Cristian


#95

Quote:
Heh... I said I'd buy one, but then I bought *two*... so I offset the others a bit! :)

Same here! But I will never pay 300$/Euro for anything with a grey LCD. Not in this life.

#96

I've never signed the HP-15C-LE initiative, simply because I was not aware of it---like (probably) many other buyers. But I've bought three of them: two as a gift and one for every-day use instead of my time-honored 15C from 1985.

As far as I see from the ongoing discussion (and especially the substantial contributions form Walter, Pauli, Eric, and Richard) the upcoming 43S will be a serious and unrivaled 'scientific instrument' like the HP classics were for decades and generations of engineers and scientists. One should think about an effective marketing strategy: I'm convinced that many potential customers didn't know anything about the existence of this forum, a poll, or even the existence of this project at the present.

I will buy at least two---and, yes, with my current budget I've to save money for them. [Im sorry for my "Denglish".]

#97

Quote:
Maybe some 300 WP 34S are in the world. That's the feedback from our project site at least - I don't have any better figures[...]

You might get another estimate by looking at how many keyboard overlays Eric has distributed. Off hand, I'd say maybe 85% of the V3 layouts would be a fair estimate (the rest accounting for people who needed more than one try to get it right).

Dave

#98

I don't doubt any of this. One can make a profit from any product at any price, as long as enough customers are willing to pay that price. That alone is the question, I hope, you can answer with a firm yes!
If you are willing to share, I would be interested to know as to how many would actually pledge to purchase at $300.


#99

Quote:
I would be interested to know as to how many would actually pledge to purchase at $300.

I don't believe in such pledges. Look how many people signed the petition for bringing back the HP-15C, for example, and how many of those bought the amount they 'promised' then. Public polls of that kind may give you some idea, maybe even an order of magnitude - provided the population you asked is in any way representative for your market. And also this last point is not given when you ask on the forum only.

Just my 20m€,

d:-)

P.S.: And far less people "bought" a WP 34S than an HP-15LE although the first is a better and far more powerful calculator. So it obviously depends on brands etc., too.


Edited: 4 Feb 2013, 4:44 a.m.


What I meant was: how many buyers do you expect? I'm only curious, not planning to steal your customers... ;-)

Eric, thanks for talking about your estimate (#84). I must confess that I have serious trouble believing that 100 people might actually buy at $300 when only 300 got themselves the WP-34s which could be had at around €30 with the added bonus of some DIY work that might have made it even more attractive to some...


Edited: 4 Feb 2013, 5:47 a.m.


Quote:
I'm only curious, not planning to steal your customers... ;-)

I din't expect you doing that ;-)
Quote:
Eric, thanks for talking about your estimate (#84).

FYI, message numbers after #2 may change most easily.
Quote:
I must confess that I have serious trouble believing that 100 people might actually buy at $300 when only 300 got themselves the WP-34s which could be had at around €30 with the added bonus of some DIY work that might have made it even more attractive to some...

I concur. Now there are two possibilities: either the number of WP 34S is too low or Eric's is too high.

d:-)

I don't quite agree with your analogy to the WP-34s. Stick-on key labels, well made though they are, are still just stick-on labels.

From what I've seen so far, this looks like a fine calculator. And if this calculator is anything like the HP-48 series hardware wise (but much faster of course :-), I would be interested in purchasing one, even for approx. 200 GBP. (I am not a person of great financial means so I had better start saving :-)

The stunning functionality of the WP34S software exceeds the capabilities of the display for several orders if magnitude.

May it possible that many interested buyers are hesitating and waiting for a scientific RPN calculator (numeric inclusive Special functions without RPL, CAS, and graphic features) of high qualitiy with a state-of-the-art display and a robust keyboard labeling?

Quote:
I would be interested to know as to how many would actually pledge to purchase at $300.

Add me to that list. If for no other reason i'll get one for the sake of tradition. I bought this a few years ago and it wants a brother.

Hey Reth; I agree. RE your car analogy: As far as budgeting goes; since my Toyota truck does what no Lamborghini can and this is the only Ferrari i ever wanted - I'll put $300 into the perfect calculator. It's a question of priorities.

i see your point gene, but for how i have used trig on a calculator:
1) almost 100% of the time i can get the angle i need plus a check distance, a distance plus an angle to mentally check reasonableness, or both quickly using P<>R and those would be on the hard keyboard.
2)When i was in school i'd need to be using the regular SOH-CAH-TOA trig functions a lot and in a case like that then the trig functions would be the softkey menu and always be up.
3) the reason i'd like to see P<>R on the keyboard is that i have used those functions a lot while running a program that would take up the softkeys.

I chose "A". that's just me and your mileage may vary.


FYI, I added a little remark at the bottom of message #1. >POL and >REC are on both keyboards, A and B.

d:-)


yes. but my point was refering to gene's post.
that:
1) standard trig functions aren't even the best way to get an answer for me personally, most of the time
so
2) putting all of them in their own suite of assigned softkeys doesn't hinder one user (me) most of the time. if they were printed on the top row keytops as in option B; they would be lost to use while a program has assigned those keys anyway.
and
3) that suite of keys is sort of a one stop shopping center for all things trig and can be left as a default when needed.

both A & B look good for easy use of P<>R. so either gene's or my preferred choice gives me what i mainly want. i'd re-assign the log keys on either with the the user keyboard and a stick-on label anyway.

now that you've pointed it out though; gene's idea has a certain economy that appeals to me.

Quote:
Option B puts more functions on the keyboard with access in 1 or 2 keystrokes.

Option A takes functions off the keyboard and makes more functions take 2 keystrokes compared to 1 and some take 3 now rather than 2. For a lot of manual use, those extra keystrokes will add up.

Why not option B with the same 3 level softkey menus for functions that are not on the keyboard?


I suggested the same thing about "Option B with the softkeys" as well in my vote (below, off the original post). :)

I will expand your comment that Option A requires at least 3 keystrokes if the menu is not already up: G-TRIG-SIN. If you're swapping between Trig & Exp functions a lot that could really add up. (Granted, if you're doing that a lot, you're probably also working with complex numbers.)

Cheers,

Doug


Where does the G come from?

To bring the TRIG menu up in option A is just the TRIG key. Likewise the log menu is just LOG. No prefix keys required.


- Pauli


Quote:
Where does the G come from?

In Option B, there is not TRIG or EXP key. I suggested putting those menus as G-shifted keys where they are unshifted keys in Option A. Is that more clear?

Oh, I see. Yes, you're right, in Option A, TRIG is unshifted so it takes only 2 keypresses. My mistake.

Cheers,

Doug

Edited: 2 Feb 2013, 6:31 p.m.

B

B as it puts the main trig functions on one keystroke and allows more room for further functions in the menu.


B for the very same reason even though ->HR, ->H.MS, H.MS+ and H.MS- on the top row is plain nonsense (or they just have a hidden machiavellian purpose there).


Hi Gerson; I don't know why the trinity is considering that setup but it would be useful to surveyors. Back before field computers were marketed which (badly) do all the surveyors thinking for him; we would have to add and divide angles all the time. Mostly for laying out curves but also for spinning multiple angles. Most guys would have those same four functions assigned to the shifted arithmetic keys on their 41. HMS + & - were , of course, always on the shifted + & - keys.


Hi Dennis,

By no means I meant to belittle those important functions. I only meant there are more general use functions that should be there instead, like the ones in the following configuration, the result of a previous poll, if I am not wrong:

Best regards,

Gerson.


Gerson; Since the calc will have a user keyboard, this is all a moot point. However i can tell you what every surveyor, gradesetter draftsman, and civil engineer who owns one of these will do if it has the keyboard from your last example. He will reassign the hr<>hms and P<>R conversion functions to the log and natural log keys and HMS+ & HMS- to the sigma key because he sees those keytops as a waste of ink and wants functions he uses to not be overwritten by a program right when he wants to calculate an entry with them. But since each year there are fewer surveyors who have to do basic math and since we will probably be able to get clear key overlays for errant keys from Eric Rechlin; it's all good. The Trinity should go with what the majority finds most convenient so there are fewer folks having to change their basic format.

PS; You didn't run down any mathematical function, but I belittle math functions all the time. For example, I believe that whatever financial functions enabled Enron, AIG, Goldman Sachs, Lehman Bros, the Federal Reserve Bank, the US Treasury and that ilk to do what they did to the global economy should be taken off the 12c keyboard and out of the textbooks. -db


I can't agree more!

Somehow we, the surveyors, put up with everything, shut up and dance ;) , but this doesn't apply to other professional people. So we are trying to find workarounds while others insist to be served first.

I can't remember when was the last time I used trig function as standalone not in a program. Log-menu ones - never, same with complex numbers, summations - in programs only. AFAIC everything that's not on a standard 4 banger should be hidden in menus and leaving more space for everything including finger travel (hp41 to be the best example).

Cheers,

Quote:
there are more general use functions that should be there instead,

According to whom? You? So you disregards this one as well as previous polls?

This is the real problem here, some reckon their views/preferences are more important/valuable than everyone else's. Layout "B" supporters are all in that category, unless they are under the impression that this calculator is being designed for them personally.

Edited: 1 Feb 2013, 9:56 p.m.

A

A. By a mile... excuse me, 1.609344 km

I'll take "B"

B.

I've been lurking here for years, but this is the first time I've felt compelled to post.

When I graduated from college (BSEE) in 1987 I received an HP 28C to replace my aging 41C. I used the 28C for about six months and then put it into a drawer, never to be used again. The problem was that all the trig and log functions were embedded in menus similar to option A, above. The extra keystrokes and visual scanning for the desired function made it unpleasant to use. It felt like using a Windows program without a mouse. If I have to search through a menu for my most commonly used functions then I might as well use Excel.

I vote for option B because I've already got option A sitting in a drawer somewhere.

-Rob


The "visual scanning" part is at least partially alleviated by having a display that has much higher contrast and resolution than that of the HP-28, so that the soft menu labels are highly legible. That's probably hard to believe without seeing it in person; even the still photos I've posted of the prototype don't do an adequate job of conveying just how legible this display is. That's the reason I'm willing to use an absurdly expensive display in the calculator, even though it pushes the manufacturing cost, and thus the retail price, up quite a bit.

Well said, Rob, and welcome.

(I gave my 28 away just to be rid of it.)

-Bill

I vote for A.

I'm abstaining from the A vs B vote, but...

If it's A, I'd rather do without the F1 through F6 legends, and just have the keys blank. The arrowheads are sufficient to convey that they are softkeys.


FYI, the only reason for introducing these labels was to identify these keys by unambiguous names instead of calling them the third from left or the second from right.

d:-)


Quote:
FYI, the only reason for introducing these labels was to identify these keys by unambiguous names instead of calling them the third from left or the second from right.

d:-)


Thats fair enough and so it should be. Just another little cosmetic thing scratches my eyes: that white background for the arrows above top row - was alot better on some previous renderings - white arrows on the main faceplate colour.

Cheers

Interesting. When will they ever need to be referred to as "F3" or "the third from the left", rather than based on the relevant menu label? In hundreds of pages of documentation for the HP-28C and HP-28S, I don't recall that they ever had to do that.

Referencing them by the menu label has the advantage that it is inherently error detecting. If you tell someone, "do this, then do that, then press F3", F3 might not mean what you think, either due to a mistake, or due to the menus changing due to a different firmware release or some other transient condition that affects the menus.If you tell them "do this, then do that, then press ACOS" (with ACOS printed as white text in a black rectangle), then there's no ambiguity.


Quote:
...If you tell them "do this, then do that, then press ACOS" (with ACOS printed as white text in a black rectangle), then there's no ambiguity.

Which reminds me the possible scenario with variant B having TAN on the key and ACOS (clearly and brightly displayed (unlike the Pioneers)) on the menu label right on top of it :)
How is that easy to make it unambiguous? Let alone the manuals.

It would be just as clear as the much-beloved HP 42S was and its documentation.

Menu labels are referenced by menu-label symbols and keys by keys.

That's how I did the HP49g+/50g learning modules and how a group of us edited the Advanced User Reference Manual for the 49g+/50g calculators.

Methinks it is easily learned. Not as clean, sure.


So does that mean that we don't need the Fn legends on the keys, since there's no reason to refer to a specific soft menu key without regard to the displayed menu label?

A because it gives me (I presume) 18 user definable functions (f1..f6 unshifted, [f] f1..f6 and [g] f1..f6).

Everybody could then have his most used functions available with a single key stroke, making it actually more key stroke efficient than B.


Ah, but those are redefinable even in B, so there is no difference. :-)

Keystroke efficiency outside of USER mode is my only concern.

I will buy one whether A or B.


"Progress is a snail" as one of our Nobel laureates once wrote.

d:-)

Quote:
Ah, but those are redefinable even in B, so there is no difference. :-)

So a key with COS hard legend on could actually do ASIN as displayed on the soft menu? Charming!
Or is it an absurd?

Quote:

So a key with COS hard legend on could actually do ASIN as displayed on the soft menu? Charming!
Or is it an absurd?


Did you ever hold a Pioneer in your hands? Recognized anything?

d#-/


Bad example. HP42s has the top row keys FAR AWAY from the screen therefore no way to get confused. HP48 is close to what you are proposing with plan A now and that is the right thing to do.

Ps. I mentioned hp42 because some ther Pioneers have nothing to do with the issue.

Btw the hp27s had it right, with those big arrows on the top row keys pointing towards the menu labels. It is an example for perfect key layout design - clean, intuitive, logical, beautiful.


Edited: 1 Feb 2013, 6:17 p.m.


Perhaps, but a LOT of HP 42s users had no problem with its arrangement for years of daily use. I was one of them.

Again, my focus is on reducing keystroke usage aka increasing keystroke efficiency in non-USER mode.

B is more efficient AND allows for 3 levels of menus etc.

A is a cleaner interface, which is is primary (only?) advantage over B.

It might well be a BIG advantage but there is no functional advantage to A over B. B can do everything A can do with fewer keystrokes in non-USER mode...at the disadvantage of a less clean interface.

I think that's a correct summary... ?


No, it's not. Read above my answer in post #34.

Quote:
B is more efficient AND allows for 3 levels of menus etc.

No, that's definitely not true. This sounds like you could have 3 menu levels AND at the same time being able to use e.g. SIN, COS etc.

Of course that won't work, as soon as you call any other menu, the 6 top row keys lose their original functions (i.e. the functions printed on the keys).

My suggestion of a default (but changable) user menu has exactly the same functionality as option "B", there's not a single main function which would require one keystroke more than in the "B" layout!

Think about the following predefined 'user menu':

>HR   >HMS  HMS+  HMS-   >POL  >REC
ASIN ACOS ATAN LOG LN LOGxy
SIN COS TAN 10^x e^x y^x
Here you have all 18 functions which you also find on the "B" layout (of course a bit rearranged, because I don't think that e.g. functions like HMS+ would really be needed on a primary key).

If this 'user menu' is displayed permanently by default, then you can access all the main math functions (trig & log) by just one single keystroke (F1..F6) or with prefix [f] - that's exactly the same behaviour like option "B". The only difference is that in my 'user menu' layout these function names are not printed on the keys themselves but in the menus instead - and that's indeed much more flexible and certainly also looking better.

And additionally it has the advantage that every user can rearrange (or even redefine) all these 18 functions to his own needs!

Franz


Edited: 2 Feb 2013, 6:19 a.m.


This is a brilliant idea, Franz! A colleague of mine would die to have your top row menu line as first (at the bottom). He would also like that hard coded on the top row keys to reach nirvana. Older surveyor's dream. I'd be happy with keeping top row keys clean.

Cheers,

Edited: 2 Feb 2013, 6:40 a.m.


Quote:
This is a brilliant idea, Franz!

Thanks Reth, at least one here who appreciates a good idea. :-)

I've already posted this suggestion twice, but so far there was almost no reaction:

http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/forum.cgi?read=238240#238240

http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/forum.cgi?read=238318#238318

If this 'user menu' can be switched on/off (e.g. after having called any other menu) for example by [g][USER] (which is still a free key position), then this would indeed be a superset of both options "A" and "B", and we won't need this useless poll here at all, because it includes both versions.

Franz


That's what I'd call "Columbus’s Egg"!

I think the user should be allowed to rearrange (by special procedure to avoid unintentional changes) this "user menu" too, to fit any particular taste?


Edited: 2 Feb 2013, 7:38 a.m.


Quote:
I guess the user should be able to rearrange (by special procedure to avoid unintentional changes) such "user menu" too, to fit any particular taste?

I would even go further: the user should be able to completely re-define his own 18 functions to this 'user menu'! For example there could be an internal variable containing a list like {SIN COS TAN ...} with all 18 entries.

Even more flexible: this internal list could be a nested list, so you couldn't only define usual functions as entries, but even sub-menus (which contain another set of menu functions).

It's just like the HP-48/49/50 allows directories and sub-directories of variables, functions etc.

Franz


I'm going dizzy, apparently sky is the limit.
Cheers,

Edited: 2 Feb 2013, 7:58 a.m.

Just to see if I get this right:

- the keys would have no hard labels, only f1..f6

- when no menu is active, the screen would display a user definable menu.

- functions then would be accessible with f1..f6 or [f]/[g] f1..f6

- if a menu (eg. TRIG or LOG) is activated, the user menu disappears and the other menu takes its place

- when the other menu is no longer needed, the user menu reappears

I think I like this, but admittedly I am not familiar with the USER mode of the 42s, so maybe there is something I miss.

One more thing: would it not make sense to connect the user menu with programme labels? Users then could launch both basic functions (SIN etc.) and full programs from there.


Quote:
Just to see if I get this right:

- the keys would have no hard labels, only f1..f6

- when no menu is active, the screen would display a user definable menu.

- functions then would be accessible with f1..f6 or [f]/[g] f1..f6

- if a menu (eg. TRIG or LOG) is activated, the user menu disappears and the other menu takes its place


Yep, so far it's just like I had it in mind, too. :-)
Quote:
- when the other menu is no longer needed, the user menu reappears


That's something I'm not yet sure: should this other menu be removed automatically after any of its function is called, or should it stay until I call any other menu again (e.g. the 'user menu' with [g][USER])?

That's certainly a matter of personal taste. I rather tend to let each menu stay until any other is called - think about you want for example do some statistics: then it's certainly more comfortable that this 'stat' menu doesn't close automatically whenever you use any of its functions, because in this case you would have to call [g][STAT] again and again for each new function call.
Quote:
One more thing: would it not make sense to connect the user menu with programme labels? Users then could launch both basic functions (SIN etc.) and full programs from there.

Yes, that's what I mentioned in my last post: you should be able to assign ANY object to this 'user menu', i.e. not only functions but also variables or own programs.

PS: IMO such a powerful and flexible user menu (possibly even with sub-menus) is much more important than the idea of a fully user-assignable keyboard. I've never understood what sense it should make to assign any other functions to the main keys (e.g. the lower 6 rows of this WP43s layout)? What should it be good for to redefine e.g. ENTER, the number keys 0..9 or the arithmetic operators +-x/ ??? Who would really like to press [+], but instead of getting the sum of Y and X he would get any other (redefined) function? It's just crazy IMO ... ;-)

Franz


Edited: 2 Feb 2013, 8:49 a.m.


Sometimes good ideas need some time ;-) If you'd carefully read what was written here more than 2 months (!) ago, you'd find all the 'revolutionary news' of today already included.

Enjoy!

d:-)

Edited to correct an error in a foreign language.


Edited: 2 Feb 2013, 3:56 p.m. after one or more responses were posted


Quote:
If you'd carefully read what was written here more than 2 months (!) ago, you'd find all the 'revolutionary news' of today already included.

Well, I didn't state anywhere that my idea(s) were 'revolutionary' or 'new' either, and of course I can't remember everything that has been written long time ago anywhere.

But then I absolutely don't understand why the heck you've created all those troubles with polls over polls instead of simply making a modern, powerful calculator layout for the 21st century (and not an old-fashioned layout which we already have in hundreds of other existing calculators). ;-)

Franz


The customer is king - although he might need extra time. Just take a look at the design of November and compare. BTW, news spread fast along the big roads and reach small valleys far later (not to be taken too literally).

d;-)

Edited: 2 Feb 2013, 4:01 p.m.

I prefer A.

If screen real estate is lacking, you could have just one row of soft menus and change what's displayed when the user presses f or g. For example, pressing TRIG would bring up

SIN ASIN COS ACOS TAN ATAN
Then pressing [f] would change the menu to
SINH ASINH COSH ACOSH TANH ATANH
Pressing [f] again would revert to the previous menu.

This sounds better, but I would prefer to use the arrow keys to change the menu. That's how the 42s works.

My vote: B

I like B because you get default functions when no menu is displayed, "for free." I'm willing to risk the confusion that this may cause, but this would certainly be covered in the first few pages of any documentation. However, perhaps the keyboard design could address this by making those keys look different than the rows beneath.

I would suggest a discussion of what should be on the keys if option B is selected. I'm not sure that the selected functions on the top row and second row are what I would choose, but I'm just one opinion. (For examples: I'd prefer having an inverse log-base-y-of-x as the F of the y^x key, and at least two to four of the other keys are of very little value to me.)

Even with B, I think having the G-menu for LOG and TRIG would be useful, perhaps as G-shifted keys in the same location as the option A keys are.

I maintain my (atypical, uncommon, unusual) belief that double-shifted keys might be useful here. For example, one key could be SIN, ASIN, SINH, ASINH: SIN, f-SIN, g-SIN, g-f-SIN. Also: e^x, ln x, 10^x, log x: E^X, f-E^X, g-E^X, g-f-E^X. (Could also be done with the x^2 key to add cubes and cube roots.)

Perhaps a system flag could be created that would change whether the TRIG menu's unshifted keys are trigonometric or hyperbolic by default, as well. If you're using hyperbolics a lot, this could save a lot of keystrokes.

To address the possible confusion inherent in Option B, perhaps the "white band with upward-pointing chevron" that is above the top row of keys could be partially inscribed into the top row keys. For example, the top keys could have a white border on the left, top and right side of the physical key (perhaps only 1mm thick) to more clearly delineate that the key is dual-purpose.

Cheers,

Doug

Walter; If you end up using alternative "A", since you have empty space in the blue (shift g) line: could you include HMS+ and HMS- on two of the unused keys? If you still don't have the last two spots on that line filled at press time; have known two surveyors who assigned HMSx & HMS/. The last two are not very commonly used functions but few things are as sad as an unassigned softkey.
I'm going to assign the HMS+ & - to my user keyboard's second line, but i think they would go well on your trig menu for most everyone.


I agree with db... if A is what is chosen, then having all three layers of any menus full of any reasonably associated functions will be a good thing, IMO.


Well, it's a question whether the user is allowed to recharge (i.e. load the free positions in) predefined menus or not.

d:-)


I didn't even think of the possibility that we users could add to a system menu. That would be a useful capability.

Preliminary results so far:

19 votes for A,

10 votes for B.

d:-)


B-PEOPLE OF THE WORLD, UNITE! :-)


Did you choose the right medium for your call? We'll see ;-)

I vote for B

B!

Preliminary results so far:

19 votes for A,

12 votes for B.

d:-)


I've waited a long time before voting because I couldn't make up my mind. I'm still unsure, but since I grew up with the 48 and liked it...

A.


I liked my HP-48GX, my third HP calculator, and still do -- especially because logs and trigs came back to the keyboard, where they used to be on my very first HP calculator (HP-15C). On my previous one, an HP-28S, these were located in menus. I used my HP-28S again last week, only to remember how cumbersome this feature is.

Results:

20 votes for A,
12 votes for B.

Thanks to everybody participating. This was one of the most vivid polls I remember having read here - and gave a faint idea of religious wars around 1600 ;-) Anyway, we'll continue with a neutral top row of keys.

There will soon be an updated file UI.pdf showing up on the website of this project. I recommend to read those 12 pages carefully for at least two reasons:

  1. Show me my errors I certainly made.
  2. Understand the concept of the 43S. Some things changed a bit based on recent discussions. And in documentation, new information overwrites old - so only the most recent file is valid (dated 2013-02-04 now).
Enjoy! And don't forget asking questions ...

d:-)

Said file is uploaded already - you find it here.

Edited: 4 Feb 2013, 9:19 a.m.


Quote:
I recommend to read those 12 pages carefully for at least two reasons:

  1. Show me my errors I certainly made.
  2. Understand the concept of the 43S. Some things changed a bit based on recent discussions. And in documentation, new information overwrites old - so only the most recent file is valid (dated 2013-02-04 now).

Well, I won't claim it to be an error, but is the little "x" between the "i" and the numeric value as depicted below for the display of a rectangular form complex number set in stone?

My preference would be to do without, like this:


Also, the above indicates that there will be RECT and POLAR display modes. Is there any chance that complex values could simply stay in and be displayed in whatever form they were entered (as I depicted in my mocked up 43s display here)? The following sentences might then be added to the above:

->REC will convert a polar form complex number to rectangular form complex number, and have no effect on a rectangular form. ->POL will convert a rectangular form complex number to polar form, and have no effect on a polar form complex value. If scalar values occupy stack X and stack Y, ->REC will ->POL will operate on the two values and return two values to stack X and stack Y according to the current angular mode. The CPLEX menu shall contain functions to assemble and disassemble complex values from or to scalar values in stack X and Y.

And before you ask, the display format of results of functions operating on two complex values would be selectable as follows:

Always rectangular

Always polar

Match stack X value’s display form

Display of complex results from functions in which only one argument is complex would match the display format of the complex input.

Complex results arising from operations involving one or more real numbers would be selectable as follows:

Always rectangular

Always polar

Or they could just always be rectangular if you prefer.


Hi Jeff,

Quote:
Well, I won't claim it to be an error, but is the little "x" between the "i" and the numeric value as depicted below for the display of a rectangular form complex number set in stone?

Nothing's carved in stone yet. You can have a multiplication dot as well (since mathematically it's a multiplication anyway).
Quote:
Also, the above indicates that there will be RECT and POLAR display modes. Is there any chance that complex values could simply stay in and be displayed in whatever form they were entered (as I depicted in my mocked up 43s display here)?

Sounds like an extra status bit for each complex number, meaning a factor 2 less precision. I assume we can live with that.
Quote:
The following sentences might then be added to the above:

->REC will convert a polar form complex number to rectangular form complex number, and have no effect on a rectangular form. ->POL will convert a rectangular form complex number to polar form, and have no effect on a polar form complex value.


Of course :-)
Quote:
If scalar values occupy stack X and stack Y, ->REC will ->POL will operate on the two values and return two values to stack X and stack Y according to the current angular mode.
I'd like to do that with a dedicated command instead.
Quote:
The CPLEX menu shall contain functions to assemble and disassemble complex values from or to scalar values in stack X and Y.

I concur.
Quote:
And before you ask, the display format of results of functions operating on two complex values would be selectable as follows:

Always rectangular

Always polar

Match stack X value’s display form

...

My understanding of POLAR and RECT modes is that these determine the format of the result.

But that's just my opinion - and we didn't discuss that topic in any detail yet. So don't hold your breath - it will take more than just a while.

d:-)

Quote:
Also, the above indicates that there will be RECT and POLAR display modes. Is there any chance that complex values could simply stay in and be displayed in whatever form they were entered (as I depicted in my mocked up 43s display here)?

What format is the result of adding or multiplying a rectangular complex number with a polar one?


- Pauli


I would be happy with the following choices for the display format of the result of your example or any other functions operating on two complex values:

Always rectangular

Always polar

Match stack X value’s display form

Or just forget the last one and have the first two set by the RECT and POLAR modes, as they will have no other uses if complex values stay in their entered display format.


OK, I take your last offer.

Since we're at that topic: how about the angular modes? These can't be sticky or we get a mess on display IMO.

d:-)


Angular modes tie into units I suspect.

- Pauli

I see no problem with having the angular mode setting rule the display. E.g., a value displayed as 5 / 36.8699 in degrees mode would be converted to 5 / 0.6435 if the mode was changed to RAD. If you need to enter 5 / 36.8699° and the calculator is in RAD mode, you will have to first change to DEG mode, unless there can be a simple way to tag the ° symbol.


I don't see any problem either :-) So why shall there be one with complex numbers?? If you need to enter 5 / 36.8699° and the calculator is in RECT mode, you can easily do so using the theta symbol - and with the completion of your entry the number will be converted to its rectangular representation.

d:-)


I sense that we have a failure to communicate. I assume that I am not clearly expressing my concepts.

If my intention is to enter 5 / 36.8699°, then I wish it to be displayed in that format, regardless of any POLAR or RECT mode setting. I wish to be able to enter 5 / 36.8699°, press ENTER, then key in 3 i 4, press x<>y and see 3.0000 i 4.0000 in stack level Y and
5 / 36.8699° in stack level X. I do not wish for the calculator to convert a complex value entered in polar representation into the rectangular representation. I would prefer that the calculator not do any conversions unless I ask it to do so. I like to think that I know better than the calculator what I am doing and the best representation for complex values for that purpose. RECT and POLAR modes would only determine the display format for the results of functions operating on two complex values that happen to be expressed in different formats. For matching formats, results would be presented in whatever format the entered values were expressed in.

Now, if I desire to enter 5 / 36.8699°, but the calculator is in RAD mode, I can indeed key in 5 then press the theta symbol to enter the angle, but unless I can tag that angular value with the degree symbol, the calculator will not know that the 36.8699 value I just keyed in is meant to be in degrees, and so will treat it as being in radians. So I have to either convert 36.8699 to 0.6435 before keying it in, or, change the angular mode to DEG, enter my value, then switch back to RAD mode, at which time the displayed value would change from 5 / 36.8699° to 5 / 0.6435 r*. If I can tag the 36.8699 with the degree symbol, then I have no objection to the calculator converting it to radians upon entry, as I presumably set the calculator in radians mode, after all.

I hope the above is clear, if not, I will keep trying as long as you are interested.

* - I use the lower case “r” here for radians, but I realize that a pi symbol is planned to be used.

Regrettably I haven't read this message before, and I missed the 4.2.13 13:13 UTC deadline (it's not 13.13 here yet, but I'm in UTC-3). FWIW, option A is much preferred by me.
Best regards


Possibly Related Threads...
Thread Author Replies Views Last Post
  Trivial news of the 43S (no Prime) Walter B 117 2,618 11-22-2013, 03:26 AM
Last Post: Raymond Del Tondo
  HP Prime: How do I insert/delete a column/row in a spreadsheet ? Michael de Estrada 3 239 11-20-2013, 10:01 PM
Last Post: Michael de Estrada
  What exactly is the WP 43S Harold A Climer 5 260 11-16-2013, 03:58 PM
Last Post: Jonathan Cameron
  HP Prime 2013 8 13. Rev:5106 bluesun08 2 154 09-28-2013, 11:05 AM
Last Post: debrouxl
  Back to School/Work/Other Calculator Poll Eddie W. Shore 63 2,173 09-12-2013, 06:03 AM
Last Post: bhtooefr
  Another wishlist for 43s Andrew Nikitin 3 195 07-14-2013, 09:25 PM
Last Post: Andrew Nikitin
  Some random wish list items for the 43s Marcel Samek 18 617 07-11-2013, 05:35 PM
Last Post: Paul Dale
  News from the 43S Walter B 78 1,394 07-08-2013, 10:33 AM
Last Post: Victor Koechli
  2-battery 12c keyboard row quit Neil Hamilton (Ottawa) 1 118 06-23-2013, 10:46 AM
Last Post: Kimberly Thompson
  Poll: SandMath and 41Z Overlays anyone? Ángel Martin 20 640 05-20-2013, 04:13 PM
Last Post: aurelio

Forum Jump: