A HP 15CE keyboard handler bug?

[Martin Hepperle]

So far I was able to lock up my HP-15CE three times.
All cases seemed to be related to pressing "some" keys (in the heat of the battle, I could not reproduce the first instances when this happened).

However, I can reproduce the last case with my unit. Is this a general problem?

Key sequence:
1.) Hold 'f' and ENTER, press ON, release f and ENTER, then ON
=> The calculator switches ON and appears normal.

2.) Now try to switch it OFF by pressing the ON key
=> nothing happens. The ON/OFF key does nothing.

3.) Repeat the sequence (with the calculator still locked in ON mode):
Hold 'f' and ENTER, press ON, release f and ENTER, release ON
=> The calculator switches off.
After this sequence, the ON key seems to work normally as it should.

Also, pressing 'f' toggles the 'f' annunciator, but pressing it again does not toggle if back to unlit. Same for 'g'. Is this intended behavior?

The sequence can occur accidentally or when one tries the self test but confuses the 'f' and 'g' keys.
Initially, before discovering the 3.) workaround, I removed batteries or pressed the reset button for recovery, but this clears the memory.

Martin

[Maximilian Hohmann]

Hello!

I would say that this is a feature, not a bug: An "ON lock" that prevents you from accidentally switching off your calculator during the most critical phase of your lunar descent...

Regards
Max

[J-F Garnier]

Reproduced on the latest firmware 2023-11-06.
It doesn't happen on the previous firmware 2023-01-05.

Note that when in this locked ON state, the machine is in a high power consumption state, so don't stay in this state for too long.
I noticed that by removing one battery and monitoring the battery voltage that then drops from ~3.0V in idle state to ~2.6V in high power state.

A minor issue with easy workaround, but good to know it.

Quote:Also, pressing 'f' toggles the 'f' annunciator, but pressing it again does not toggle if back to unlit. Same for 'g'. Is this intended behavior?

This is the original 15C behavior, so I believe it's intended.

J-F

[brouhaha]

(02-02-2024 08:24 AM)Martin Hepperle Wrote:  Also, pressing 'f' toggles the 'f' annunciator, but pressing it again does not toggle if back to unlit. Same for 'g'. Is this intended behavior?

Yes. to cancel the f shift, press the backarrow key, whose f-shifted function is "clear prefix".

To cancel the g-shfit, press f and then backarrow.

HP established that mulltiple presses of the shift key don't toggle it at least as far back as the HP-80, in 1973. With calculators that didn't display the shift state(s), I suppose that made sense.

[Maximilian Hohmann]

Hello!

(02-02-2024 09:09 AM)brouhaha Wrote:  To cancel the g-shfit, press f and then backarrow.

Which is something that never made any sense at all. Since there can be only one active prefix at any time, only one Clear-Prefix function is required to clear it. Unless the developers had in mind to implement functions that needed f and g shift together.

Regards
Max

[Divasson]

Hi,

About this newly discovered mode:

- This happens only with the 100ms rebound The Calculator Store supplied firmware ("enhanced") with checksum C6C6h and dated 2023-11-06
- If you connect to a PC with the cable, you will see the calculator in usb as HPb10II+. This mode is intended to back up the calculator user memory - but we're still missing the PC side app.
- To go out of this mode without resetting, press f-ENTER and "on" again.

So, it is a feature, not a bug. To be updated shortly...

[Martin Hepperle]

(02-02-2024 09:46 AM)Divasson Wrote:  Hi,

About this newly discovered mode:

- This happens only with the 100ms rebound The Calculator Store supplied firmware ("enhanced") with checksum C6C6h and dated 2023-11-06
- If you connect to a PC with the cable, you will see the calculator in usb as HPb10II+. This mode is intended to back up the calculator user memory - but we're still missing the PC side app.
- To go out of this mode without resetting, press f-ENTER and "on" again.

So, it is a feature, not a bug. To be updated shortly...

I confirm, that this is one of "your" HP15CEs, flashed with the improved HP-16CE easter egg (unfortunately yet without the cable).

[brouhaha]

(02-02-2024 09:22 AM)Maximilian Hohmann Wrote:  

(02-02-2024 09:09 AM)brouhaha Wrote:  To cancel the g-shfit, press f and then backarrow.

Which is something that never made any sense at all. Since there can be only one active prefix at any time, only one Clear-Prefix function is required to clear it. Unless the developers had in mind to implement functions that needed f and g shift together.

No, they just didn't want to give up TWO possible key assignments to clear prefix. The key for clear prefix only does that for f shift. The g shift of that key is used for another function.

[Voldemar]

(02-02-2024 09:46 AM)Divasson Wrote:  So, it is a feature, not a bug.

Clear

[rjwalker]

To provide an update: just very happily taken possession of an HP-15C from TCS with firmware 0A0A 2024-06-03, and the "ON lock feature" does not seem to be present.

Richard.

[AnnoyedOne]

(09-01-2024 05:41 AM)rjwalker Wrote:  ...the "ON lock feature" does not seem to be present.

Perhaps because f-ENTER-ON isn't a bug but a feature.

USB mode for use with VoyagerSave. I've used it.

The "bug" was that the feature wasn't fully implemented in the earlier firmware. Now it is (with 0A0Ah).

https://www.hpmuseum.org/forum/thread-22078.html

A1

PS: My HP-12C+ has the same "bug". Hopefully Moravia will update its firmware too.

https://www.hpmuseum.org/forum/thread-22...#pid190217

[rjwalker]

Quote: USB mode for use with VoyagerSave. I've used it.

OH. Got it. Yes, (using the cable) I confirm that the keystroke sequence now does indeed put the calculator into USB HID mode.

Richard.

[AnnoyedOne]

(09-02-2024 07:38 AM)rjwalker Wrote:  ...I confirm that the keystroke sequence now does indeed put the calculator into USB HID mode.

IMO the memory backup/restore feature (with VoyagerSave) is more useful than it first seems. If your HP-15C ever has a "Pr Error" (for whatever reason) you can get its state back quickly.

The HP-20S has a neat feature where you can verify the checksum of a entered program. I don't know if other models do. With earlier HP calculators keying in a long program is both time consuming and error prone. With VoyagerSave it only has to be done once.

A1

PS: I think SwissMicros has software to read and list a program from a memory file. That'd be nice.

[CheshireChris]

(09-02-2024 02:40 PM)AnnoyedOne Wrote:  PS: I think SwissMicros has software to read and list a program from a memory file. That'd be nice.

I have an SM DM42, and that certainly allows you to export one or more programs as a ".raw" file to the calculator's USB-mountable flash storage space, and subsequently reload them from the file. You can also save and load the entire memory contents as a "state" file allowing you to make a snapshot of your current calculator environment. I don't know if the Swiss Micros version of the 15c has similar capabilities.

[Thomas Radtke]

(09-02-2024 02:40 PM)AnnoyedOne Wrote:  The HP-20S has a neat feature where you can verify the checksum of a entered program. I don't know if other models do.

The 32SII and 33S can, maybe the 32S, too. The 35s has a non-working checksum.

[rprosperi]

(09-04-2024 04:43 PM)Thomas Radtke Wrote:  

(09-02-2024 02:40 PM)AnnoyedOne Wrote:  The HP-20S has a neat feature where you can verify the checksum of a entered program. I don't know if other models do.

The 32SII and 33S can, maybe the 32S, too. The 35s has a non-working checksum.

My favorite HP Calculator bug. The feature added to insure entry of a program is correct, fails. Sometimes. <sigh>

[Martin Hepperle]

I wanted to try out the VoyagerSave Program under Windows 7.

After plugging in, the 15CE was detected as a HID device, but not with a human readable name (HP10b).

However VoyagerSave does not recognize the calculator properly. Could this be a compatibility problem with Windows 7?

VoyagerSave reports
Error: calc size too big: 61440
Error: no calculator found

My calculator firmware shows the checksum C6C6h.

It becomes somewhat confusing, which firmware is which and which would work with VoyagerSave.

So far I found:

• Original firmware: checksum 9090h
• first firmware with DEC fix (and 80 ms key debouncing?): AF90h
• firmware with DEC fix (and ??? ms key debouncing?): C6C6h (in my calculator)
• firmware with 100 ms key debouncing: 0A0Ah
• HP_IAR_USB_120ms.bin (8-bit Checksum 0A) (? same as above?)
  

The last file seems to be the latest, recommended firmware?

   
(oops: the red arrow should be moved down by one line)

[AnnoyedOne]

(09-10-2024 09:29 AM)Martin Hepperle Wrote:  Could this be a compatibility problem with Windows 7?

It works fine on my Win7 SP1 x64 box. Note that VoyagerSave is a 64-bit program. As far as I know there is no 32-bit version available.

(09-10-2024 09:29 AM)Martin Hepperle Wrote:  My calculator firmware shows the checksum C6C6h.

You need to load the very latest dated 2024-06-03 checksum 0A0Ah.

https://www.hpmuseum.org/forum/thread-22078.html

A1

[Martin Hepperle]

O.K. ... after flashing HP_IAR_USB_120ms.bin my calculator shows

Che--0A0Ah
2024-06-03
C 2008 hP

(not sure why the Copyright is still dated 2008, are there any legal implications?)

Anyway, with this firmware the VoyagerSave program works.

VoyagerSave.exe -s HP15CE.calc
2048 bytes were saved to HP15CE.calc

Thank you for the help,

Martin