Correct 12C operation

Things that aren't right, that are in the queue to be fixed... along with workarounds.
Xeno
Posts: 55
Joined: Sat Oct 25, 2008 4:15 pm

Correct 12C operation

Post by Xeno » Sat Oct 25, 2008 5:31 pm

On your web page, you make a play of the fact (which I have confirmed) that some HP emulations out there give the wrong answer to the following problem:
n = 60
PMT = -90000
FV = 500000
Calculate i

One popular emulation currently gives the answer 1001.

The problem I have is that the 12E does not give any answer at all! It gives an error message instead.

This is particularly serious because another more general RPN calculator on the site, one third the price of the 12E, solves the problem correctly, as do my 17B physical calculator and my 12C ROM emulation on the Mac.

This makes standing on the claim of fidelity to a 12C's operation rather peculiar. Will you be fixing this?

Incidentally, I have entered the same calculation repeatedly on three different 12C-style calculators on each of two different iPhones so it is not just finger-trouble on my part.

Xeno
Posts: 55
Joined: Sat Oct 25, 2008 4:15 pm

Re: Correct 12C operation

Post by Xeno » Sat Oct 25, 2008 5:39 pm

In another thread, one of your users found that the following entry did not work:
PV = 100
FV = -200
i = 6
Calculate n

You suggested he reverse the signs of the first two entries, but I found it works for me either way around. That is good because that is the behaviour I would expect. So long as we respect the cashflow sign convention there should be no issue with which is labelled negative and which positive.

What I found interesting was the answer.
12E on iPhone = 11.90
well known 12C on iPhone = 12.00
nonpareil 12C ROM emulation on Mac = 12.00
HP 17B physical = 11.90

So, it seems a real 12C gives the answer to the required number of completed months but HP changed this behaviour in their more advanced 17B and that is the pattern you are following.

Which one do people favour here?

theDev
Posts: 240
Joined: Mon Aug 04, 2008 8:58 pm

Re: Correct 12C operation

Post by theDev » Sat Oct 25, 2008 6:09 pm

First... yes, it will be fixed. Funny how I harp on something and it bites me as well!

Actually, I usually run the problem with pmt positive and fv negative...

Let Roy and I find out what went foobar with reversing the signs--I think that it has to do with how the zero in PV is treated in this case, if you set the PV to -1 the solver has no problem. But it should deal with PV=0 so we'll get it fixed.

Thanks for point this out--

Kim

theDev
Posts: 240
Joined: Mon Aug 04, 2008 8:58 pm

Re: Correct 12C operation

Post by theDev » Sat Oct 25, 2008 6:16 pm

Xeno wrote:In another thread, one of your users found that the following entry did not work:
PV = 100
FV = -200
i = 6
Calculate n

You suggested he reverse the signs of the first two entries, but I found it works for me either way around. That is good because that is the behaviour I would expect. So long as we respect the cashflow sign convention there should be no issue with which is labelled negative and which positive.

What I found interesting was the answer.
12E on iPhone = 11.90
well known 12C on iPhone = 12.00
nonpareil 12C ROM emulation on Mac = 12.00
HP 17B physical = 11.90

So, it seems a real 12C gives the answer to the required number of completed months but HP changed this behaviour in their more advanced 17B and that is the pattern you are following.

Which one do people favour here?
Excellent observation... HP did indeed change the behavior (I did too, v1.x followed the 12C model and truncated the periods to integer) and Roy and I went back and forth on what to do. What we came down to is to include the decimal portion because it's just less confusing. But that's just us... I agree -- what do real people think?
(Roy's a math geek, I play with confusers... what do we know?)

Xeno
Posts: 55
Joined: Sat Oct 25, 2008 4:15 pm

Re: Correct 12C operation

Post by Xeno » Sat Oct 25, 2008 7:50 pm

Thanks for those responses.

I should have said what I liked on the second one. I agree that it should show actual time with decimal period if necessary, rather than rounding up o the next whole period. Rounding up is just misleading whereas it is easy for humans to say "11.9 = 12 payments", or to see 11.9 then put 12 into n and recalculate FV, PMT or i as appropriate to reflect what will really happen.

By-the-by, I have noticed that the nonpareil ROM emulation of the HP Voyager series (which includes the 12C) has been withdrawn for copyright reasons..... :shock:

Xeno
Posts: 55
Joined: Sat Oct 25, 2008 4:15 pm

Re: Correct 12C operation

Post by Xeno » Fri Nov 14, 2008 8:00 pm

In the updated v1.1.0, it appears you have silently fixed this problem. It now resolves the "differentiating problem" from your web site correctly. I have changed my App Store review, with a rating upgrade.

tsinvest
Posts: 8
Joined: Sat Nov 15, 2008 8:59 am

Re: Correct 12C operation

Post by tsinvest » Fri Nov 28, 2008 6:11 am

Hey Xeno, Crimson Research recently posted a FREE 12C that gets this calculation correct. Just thought I would let you know as you were interested in a free 12c before. I am still a little surprised that KIM is continuing to ask $24.95 for the 12E, especially with the number of bugs that keep popping up. Regards.
Xeno wrote:On your web page, you make a play of the fact (which I have confirmed) that some HP emulations out there give the wrong answer to the following problem:
n = 60
PMT = -90000
FV = 500000
Calculate i

One popular emulation currently gives the answer 1001.

The problem I have is that the 12E does not give any answer at all! It gives an error message instead.

This is particularly serious because another more general RPN calculator on the site, one third the price of the 12E, solves the problem correctly, as do my 17B physical calculator and my 12C ROM emulation on the Mac.

This makes standing on the claim of fidelity to a 12C's operation rather peculiar. Will you be fixing this?

Incidentally, I have entered the same calculation repeatedly on three different 12C-style calculators on each of two different iPhones so it is not just finger-trouble on my part.

Xeno
Posts: 55
Joined: Sat Oct 25, 2008 4:15 pm

Re: Correct 12C operation

Post by Xeno » Fri Nov 28, 2008 3:07 pm

I already have that and reviewed it on the Australian App Store in the following terms:
Another nonpareil?
Judging by the calculator's behaviour in a few tests I ran, this is another implementation of a virtual machine running HP ROM code. It has the 12-digit internal precision, slowness and other foibles of the original HP so if that is exactly what you are after, you are in luck.

Not that you can confirm or contradict any of this from the developer's web site or "support" site where so far as I could find this software is not even mentioned.

If your financial and aesthetic needs are satisfied by an original 12C, and many people's are, then this is cost free. If you rely upon higher precision and deep financial accuracy, then the costs may prove to be substantial.
I gave it three stars because I consider that fair for the capability, which is neither less nor more than an original 12C. If you are complaining about the price of the more capable Calc-12E, then why are you not complaining about the fact that Tomas Fors sells for the same price as Calc-12E the same product as Crimson Research is giving away free (albeit wholly unsupported in the latter case)? What about the less capable RLM 12C which is also sold for a "significant" price?

tsinvest, I am noticing that you do not respond directly to the substance of the replies I make to you, and that in this case you have repeated one of the first posts I made here, over a problem which was promptly fixed, and not dealing with any of the subsequent extensive discussion of that and related issues. Was your intention to bring that original bug to the fore again, as if it still existed? That would be behaviour amounting to trolling.

I have examined your seven posts here and every one of them is a complaint, and none involving original research on Calc-12E by yourself.

If you are really interested in Calc-12E, then why not make constructive posts about actual bugs and desirable features, posts you have never yet made, rather than always seeking invidious and futile comparisons? We see that you will not put your money down so why not do the job properly? Join a Ferrari forum and complain about their prices and unreliability compared with Toyotas.

If you are not interested, then why are you not ecstatic to announce that the Crimson Research 12C is clearly a better value proposition than Brock's machines and wander off into the sunset with 12C on your iPhone and two real 12Cs in your pockets?

I hope that the attitude I have discussed in the latter paragraphs does not truly reflect your views and am happy to engage directly in private discussion if you wish to use PMs.

regards
David

tsinvest
Posts: 8
Joined: Sat Nov 15, 2008 8:59 am

Re: Correct 12C operation

Post by tsinvest » Sat Nov 29, 2008 12:56 pm

Sorry to have offended you. In one of your prior post you asked where you could get a free 12C for the iTouch - I only intended to relay this information to you. I guess, as you say in your post, I have been too vocal about the bugs that seem to keep popping up in the 12E. You are also correct in that I haven't purchased the program yet and that I am still commenting on it. It is here that we differ. I don't need to purchase it to know that it has had calculation bugs - this is partially the reason I haven't purchased it. I prefer not to pay $24.95 to be a "beta tester" for a program that was supposed to be more accurate than all the rest. There are features of the program that I like, specifically the display of the registers that are in use. It actually amazes me that you and I are having this dicussion about a small twenty five dollar program, nevertheless, when the developer makes a claim as powerful as they did here regarding accuracy, and that they even have the original hp programmer of the 12C working on the project, I would not have expected the bugs that surfaced to have occurred.

This is interesting - the free 12C gave accurate answers and the 12E gave inaccurate answers before it was fixed. Do you think you have "higher precision and deep financial accuracy" because the developers told you so? Maybe I will try to sell you a Toyota and tell you it is really a Ferrari - just disregard what facts may lead you to think otherwise.

All sparring aside, I am glad you are happy with your 12E, and, for the time being I think that I will use the free 12C and the "RCL" button to see my registers.

Thanks for your post.

My best, Tom

Xeno wrote:I already have that and reviewed it on the Australian App Store in the following terms:
Another nonpareil?
Judging by the calculator's behaviour in a few tests I ran, this is another implementation of a virtual machine running HP ROM code. It has the 12-digit internal precision, slowness and other foibles of the original HP so if that is exactly what you are after, you are in luck.

Not that you can confirm or contradict any of this from the developer's web site or "support" site where so far as I could find this software is not even mentioned.

If your financial and aesthetic needs are satisfied by an original 12C, and many people's are, then this is cost free. If you rely upon higher precision and deep financial accuracy, then the costs may prove to be substantial.


I gave it three stars because I consider that fair for the capability, which is neither less nor more than an original 12C. If you are complaining about the price of the more capable Calc-12E, then why are you not complaining about the fact that Tomas Fors sells for the same price as Calc-12E the same product as Crimson Research is giving away free (albeit wholly unsupported in the latter case)? What about the less capable RLM 12C which is also sold for a "significant" price?

tsinvest, I am noticing that you do not respond directly to the substance of the replies I make to you, and that in this case you have repeated one of the first posts I made here, over a problem which was promptly fixed, and not dealing with any of the subsequent extensive discussion of that and related issues. Was your intention to bring that original bug to the fore again, as if it still existed? That would be behaviour amounting to trolling.

I have examined your seven posts here and every one of them is a complaint, and none involving original research on Calc-12E by yourself.

If you are really interested in Calc-12E, then why not make constructive posts about actual bugs and desirable features, posts you have never yet made, rather than always seeking invidious and futile comparisons? We see that you will not put your money down so why not do the job properly? Join a Ferrari forum and complain about their prices and unreliability compared with Toyotas.

If you are not interested, then why are you not ecstatic to announce that the Crimson Research 12C is clearly a better value proposition than Brock's machines and wander off into the sunset with 12C on your iPhone and two real 12Cs in your pockets?

I hope that the attitude I have discussed in the latter paragraphs does not truly reflect your views and am happy to engage directly in private discussion if you wish to use PMs.

regards
David

Xeno
Posts: 55
Joined: Sat Oct 25, 2008 4:15 pm

Re: Correct 12C operation

Post by Xeno » Sat Nov 29, 2008 1:30 pm

I read and remember pretty well so it is a waste of space to keep quoting entire posts.

The context in which I asked about a free 12C was that you asserted there was one during your first complaint whereas there was not, although RLM had previously made an introductory free offer on his "12C Lite" (and I obtained it at the time). There is now a free and unsupported one which I observed on its release, and I am as uninterested in it as I was at the time that I explained I was already aware of all of the 12C models on the App Store.

Improved accuracy and capability of the 12E over the 12C is already proven and explained to you. Bugs are found by users, in all software, and the more so when they perceive that it has additional capability worth exploration.

The current known bug list comprises one accidentally introduced in the last update, disabling the Lst x key, which is being fixed again in the next. Is that the bug which deters you?

The only other topics under discussion are improvements proposed, to which your contributions have been....?

If you are serious, explain to us what specifically will lead you to reject the free 12C in favour of Calc-12E. If it is price, name your highest bid.

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest