15c LE sqr and sqrt problems



#63

Anyone having issues with their HP 15c LE, regarding the square and square root keys? I'm getting different (incorrect) results on the 15c, compared to my 33s, using the same key strokes.

Example on the 15c:

25 sqrt results in 25.00

25 g x^2 results in 300.00 (33s and pc emulator say 625)

Any thoughts on this? I have an email into HP support also.

Thanks,
Dave


#64

I don't have a problem on my 15C LE. Sounds like a faulty chip. Did you do the checksum self test #2 ?


#65

How do you know if the check sum (test #2) works? I don't get an error message, just "ChE--A3A3h", then "2009-07-02", and "C 2008 hP". Second two screens are the copyright info...not sure what screen 1 is.

Also, is the self test (X and ON keys) almost instant? manual says 23-25 seconds of the word "running", but all I see is it flash by nearly instantly, going to all segments lit.

Thanks,
Dave


#66

Wierd. Everyone else gets checksum ChE--FFFFh and the date should be something like 2011-04-15. I wonder if the chip was flashed with the wrong ROM like for the 12c+.

#67

That is the wrong check sum and date for the 15C LE. It sounds like the data for the 12C AE. Maybe your 15C LE got flashed with the wrong ROM. That would also explain your other odd results. I believe that the upper left key on the 12C is N, when you enter a number and press N, nothing happens to the display. I believe that g, N is "x12", multiplying the keyed in value by 12.


#68

Quote:
That is the wrong check sum and date for the 15C LE. It sounds like the data for the 12C AE. Maybe your 15C LE got flashed with the wrong ROM.

Don't send it back. You have a rare collectible!
#69

Do you have USER turned on? Pressing those 5 keys will run programs A through E if you do, and you can get some rather unexpected results if you aren't paying attention. If it's on, press USER (f RCL) again to turn it off. I'm getting correct results on my machine.


#70

Not if you type a shift key first. So 25 g x^2 will still result in 625. Also 25 f A will reverse and give 5.

#71

I get the correct answers.

#72

Are you sure you are not in complex mode? (Is there a little "C" annunciator in the lower right corner of the display?) If so, press g, 5, 8. (g, CF, 8)

Sorry if you are well aware of complex mode, but inadvertantly being in complex mode and getting unexpected results is a common problem expressed here.

If this is not your problem, I'd try resetting your calculator via the reset hole under the battery cover, or with the calculator off, press and hold the minus key, press ON, then release both. Either action will clear memory and reset the calculator to its default state.


#73

It won't matter, unless there is a non-zero value in the y stack register. See above, sounds like it's a firmware problem.


#74

I don't believe that is his problem, see my other post (I think he got the 12C AE firmware.) But I will point out that the stack y value value does not matter with the 15C in complex mode for sqrt or x^2, that is the 32S, 32sii, 33s and 34s paradigm. (I'm sure you knew this :-) ) With a bit of calculation, I found that sqrt(25.0000 + i 1224.744872) is 25.0000 + i24.4949, which of course would just show 25.00 in the display. Pretty unlikely that he would have accidentally gotten that complex value entered.

Edited: 18 Sept 2011, 1:58 p.m.

#75

I found what the problem is!!!

You have a 12c+ with the skin of a 15c :-(

Those are the results if you make the same operation with the same key in a 12c

25 [n] = 25

25 [g] [12x] = 300

So the wrong firmware

Miguel

Edited: 16 Sept 2011, 10:30 p.m.


#76

Sure sounds like it's the firmware...good grief.

Thanks guys, I'll be talking to HP soon.

-Dave


#77

I'd be interested to hear about your experience with HP customer support. I've never dealt with them before.

#78

I'd say it's 100% that you have the 12c AE firmware, everything you noted is consistent with that. Assuming that this came straight from the factory this is sign that quality control has gone to pot there.

Is TW following this thread?


#79

He probably follows and reads just about everything but cannot comment all of the time.

#80

Maybe it's not a "bug", just an "undocumented feature", which is why it's a Limited Edition.

I'll keep you posted, once I get hold of HP.

#81

If this turns out to be the only one like it, it is indeed a "rare collectible"! ;-P


#82

Not really, since anyone with a cable, access to an old copy of old 12C+ firmware and SAMBA can crank out a ton of these and there would be no way to know that they weren't factory defects.


I'm starting up my production line now ..... :)


#83

Hi KAtie,

ona more serious note: the 15c firmwatre has never been released before...so to update any of these machines will essentially result in HP publishing that? Or am I confused...for instance tat thread smetime bsk with close-up photos f 15c boards, somehow figuring our the ROM from the physical board?? some of this stuff os just magic.


Edited: 16 Sept 2011, 11:38 p.m.


#84

Quote:
..for instance tat thread smetime bsk with close-up photos f 15c boards, somehow figuring our the ROM from the physical board?? some of this stuff os just magic.

Looks like your cat was walking on the keyboard while you were typing this. :)


You're correct though, the 15c firmware has never been released and HP would have to make that available to anyone that wants to upgrade to a bug-fixed version -- if that ever happens. The only alternatives would be for HP to set up some place to send calculators to for the upgrade OR some group of people here decide to write their own firmware for the 15C and make that freely available.


#85

One question: The firmware appears not to be protected (might be a wrong assumtion). So, if a 15C LE goes bad, one could take a 12C+, reflash it, and give keys and bezel of the 15C LE to it? Would that work?


#86

I think that you're asking if there is small hardware difference in the 12C+/AE vs the 15C LE that the firmware looks at and will only run on the proper hardware. If so, I suspect not since we have one reported case where the factory appears to have sent out a 15C LE with 12C+ firmware. Probably the check would be there either way but maybe not.

Until HP releases a fixed firmware for the 15C LE this is a moot point since only beta testers would have a copy of the firmware anyway and as far as I know there is no way to extract the firmware from an existing 15C LE and save a copy of it.

-Katie

#87

Quote:
Looks like your cat was walking on the keyboard while you were typing this. :)

You've read The Silent Meow?
#88

This would be extremely worrying if it is true.
Seems like a major lapse in quality control and a somewhat bizarre production run.

At least it is fixable by reflashing, or one would hope so.


- Pauli

#89

So, I wonder if any 12C+ AE calcs got the 15C LE firmware?


#90

Since there are presumably a lot more of the former than the latter, it's likely we'd never hear about it if they did.


#91

I thought that the 12C+ AE was also a pretty limited run. Of course if you throw in the regular 12C+...

#92

What's your unit number?


#93

Quote:
What's your unit number?

#02187


#94

What doesn't make sense is why would a machine be flashed differently mid production run. I have units 1081, 1200, 1598 and 2970 and they all calculated 25^2 correctly. Your unit is in between my units.

Edited: 17 Sept 2011, 9:47 a.m.


#95

I would think the boards are all flashed and tested before heading to assembly. If a 12c board accidentally fell into the 15c bin it would simply be picked up and fitted into a 15c case. It would just be random as to which (un)lucky person got the wrong board.


#96

That makes sense... so could he take a 12c on the side, make calculations using the same keys of his (un)lucky 15c and see if they give the same answers...?

Edited: 17 Sept 2011, 10:15 a.m.

#97

It could have been just one board got mixed up, but what are the odds that just one got from a tray or bin of 12C+ boards to a 15C tray or bin? I think it would be very unlikely. More likely would be that one whole bin or tray would be confused. Of course we don't know exactly how they do things, but I bet there are more like this.


#98

After WWII we sent Drs. Deming and Juran to Japan to teach them about quality control. Is there anyone we can send to China?


John


#99

Quote:
After WWII we sent Drs. Deming and Juran to Japan to teach them about quality control. Is there anyone we can send to China?

The Japanese.

That didn't work too well in 1937.

given the back labels are likely shuffled semi-randomly (put on in no particular order) it is no surprise that the unit's limited edition number appears between closely numbered ok units.

my own manufacturing experience is with firmware being loaded right at the end of the production line just before units are boxed up for shipping (CB430 winCE boxes loaded with o/s and machine control software). in that way completed units can be stockpiled in the factory and the absolute latest firmware added just before they go out the door. in the case of HP calculators, this would mean final testing and loading firmware would be combined into a single step on a single jig.

so it could be a whole run of calculators have had the wrong firmware loaded. i'd certainly like to see photos of the unit in question to verify:

- back label markings (12c or 15c)

- keyboard escutcheon (12c or 15c)

- key labelling (12c or 15c)

- badge next to LCD (12c or 15c)

including two different firmwares, that gives 32 possible assembly combinations. if the unit in question has mismatched physical components (like 12c key labelling and 15c escutcheon) then it COULD be a rare collectable unit indeed.

btw, i must ask anyone at HP who is listening: how many times in the past has HP shipped a calculator with the firmware or PCBs or ROMs for a different model fitted? i'm picking this instance is probably the absolute FIRST...

Edited: 17 Sept 2011, 11:16 a.m.


Here are a couple shots of the front, and back:


After a few emails with HP tech support, which I mentioned the suspicion of a firmware issue, I got this reply:


Quote:
Please try running the following self test for the calculator:
-Turn calculator off;
-Press and hold key ON;
-Press and hold key [ divide ];
-Release ON;
-Release [divide ];
-Press all the keys. Starting with key [ n ] and one by one towards the right until [ divide]. Go to the next row, staring with key [ yx ] until [ x ]. (Note: that key [ Enter ] must be pressed twice, when doing the third and fourth row.)

If after pressing all the keys the result on the display is "12", then this is using 12C firmware.
Please let us know the result you get after completing this test.


Guess what mine displays? "12"

Shall I go buy lottery tickets tonight?


No news here. Your firmware date and checksum already confirmed that you have 12C firmware in ROM. Please let us know what HP tells you to do with the calculator and if you follow their suggestion.

-Katie

p.s. Where did you get this calculator from, Buy.com?

Edited: 17 Sept 2011, 4:45 p.m.


I will keep everyone posted on what happens next.

Latest email says they will review the situation on monday, since calculator specific tech support is only a m-f operation.

Yes, buy.com

-Dave


Please do. I hope you get a correct replacement (no charge I shipping). I got one of my two from buy.com and it worked correctly. What a bummer!


HP tech support has advised me to return the unit to my place of purchase, in this case buy.com.

I'm a bit surprised, considering the unusual issue, that they didn't want it returned directly to them.

At least buy.com provides a return label, I just hope it doesn't take forever, and that they don't run out in the mean time.

-Dave


I think if I were in your position, I'd simply buy another unit directly from the HP store, and have Buy.com issue a refund. Now that you can get one from HP for about $85 with a discount coupon, you probably won't lose much in the transaction.

Edit: Never mind. I just checked the HP site and they're already out of stock. Back to price gouging on TAS I guess.

Edited: 19 Sept 2011, 10:15 a.m.

Instead of returning it to buy.com, you should contact Tim per his post above.


I will do that right now, thanks!!

But, you don't have an "n" key and the first key on the left side of the second row is "SST", not "y^x", so how did you do this ? ;)

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.


It has 2 batteries in it, and contacts in both battery pockets.

Wow, this was probably only made a few units after mine (CNA1310KBJ, LE #02195) if I'm correct about the last part of the serial number being a sequential production code. So at least maybe this means that not an entire production run got misprogrammed.


Maybe they ran the 15C calculators first, then switched the machine over to flashing the 12C firmware before all the 15C units were through the line somehow.

I first thought its odd that factory's EOL (end of the line) test did not catch this. But further thinking, this makes sense. The line was originally set up to produce 12C only therefore all they needed to test at the final inspection was if the machine fires up. For the same reason, firmware was probably flashed at the PCB assy level, or even at the chip level.

When they brought in variable (hp-15C) they probably did not bother to update the quality process...


Edited: 17 Sept 2011, 9:24 p.m. after one or more responses were posted


Along those lines, they could have had a 12C+ PCB lying around, already flashed. I'm trying to get a grip on how this could be a one-off. This might be one plausible explanation.


There is one more thing puzzles me. HP 15C LE has two (2) 2032 batteries vs HP12C+ (Platinum)has only one (1) 2032 battery. How did they get a 12C board into a 15C case with two batteries? So is it running on a single battery, and there are no contacts for the second?


The black and gold 12c+ has two batteries.

"12C+" is an informal name for the current production 12C. It is an emulation of the original 12C rom on a modern processor. The 12C Platinum is totally different hardware and software. (I am not a hardware or software expert, the preceding is based on my more or less layperson perspective regarding what I have read here over the years.)


Correct. It's confusing, because HP never documented the changeover from 12C to 12C+. The extra large battery cover is the sign that it's a 12C+ and not an 12C. AFAIK the 12C 25th AE is the same as the original 12C and the 12C 30th AE is the same as the 12C+. The 12CP is its own animal. You are paying an extra $10 for a fancy gift box. Ironically, it appears that HP flashed the 30th AE ROM with an older buggy version of the code than the regular 12C.

Edited: 19 Sept 2011, 9:21 p.m.


Quote:
AFAIK the 12C 25th AE is the same as the original 12C...

Michael, it,s actually a 12c Platinum 25th Anniversary Edition, which is a dual RPN/ALG version that was a re-design of the all-silver 12c Platinum. i.e it is a little faster than the original 12c, and has the greater number of program steps, uses 1 cell.

As noted by another poster the 12C+ is not the same beast as the 12C Platinum. All 12C+'s (AFAIK) have 2 x 2032 batteries in parallel. You change one at a time to prevent data loss. Unlike the original 12C and 15C with 3 batteries, the removal of both batteries *WILL* result in memory loss no matter how quickly you do it--at least I certainly have never managed it.

That said, the newer 12C Platinums also have 2 x 2032 batteries in parallel. Mine certainly does. These are still labeled as "New Design" on the HP web site.

Cheers,

-Marwan


Quote:
That said, the newer 12C Platinums also have 2 x 2032 batteries in parallel. Mine certainly does. These are still labeled as "New Design" on the HP web site.

That's probably because they are comparing it to the the "old" design as the all-silver Platinum without the parenthesis. I think the 2 cell Platinum adaption was a silent roll-out just to stay in-line with the 12C+ for production purposes, which makes sense from a production point of view (why have two different moulds if you don't need to?).

Possibly Related Threads...
Thread Author Replies Views Last Post
  Possible bug with sqrt function in the HP prime Michael de Estrada 6 828 11-15-2013, 12:49 PM
Last Post: Michael de Estrada
  OT - A lucky find - Busicom Handy-LE LE-120A Cristian Arezzini 2 575 09-26-2013, 04:43 AM
Last Post: Cristian Arezzini
  Low power warning for HP-15C LE and batteries Nick_S 1 514 09-16-2013, 09:34 AM
Last Post: Hans Brueggemann
  HP 15C-LE replacement still available? Borja 16 1,803 08-22-2013, 11:16 AM
Last Post: Michael de Estrada
  JTAG on HP-12C and HP-15C LE Ingo 5 948 07-01-2013, 06:37 PM
Last Post: Paul Dale
  HP 15C LE extremums Richard Berler 29 3,168 05-20-2013, 03:26 PM
Last Post: Dieter
  New 15C LE bug? Paul Dale 3 706 02-05-2013, 09:27 PM
Last Post: Mike Morrow
  HP 15C LE, Program Display Format Control Uli 4 818 01-20-2013, 01:22 AM
Last Post: Ethan Conner
  HP-15C LE, wow! New owner euphoria. Sasu Mattila 16 1,656 09-07-2012, 06:55 PM
Last Post: lars Bergström
  HP 15c LE bugs Alexander Oestert 11 1,134 09-06-2012, 10:29 AM
Last Post: Peter Murphy (Livermore)

Forum Jump: