Another subtle HP-15C CE difference

[naddy]

I just noticed another tiny difference between the CE and the original 15C:

When you run the 1L self-test on the CE to turn on all display segments, this includes the low battery indicator. By comparison, the corresponding end of the ON/× test on the original Voyagers does not turn on the battery indicator.

If this has been mentioned before, it's not mentioned in Gene's pinned summary.

[lvt]

The screen on the original Voyager might not have the battery indicator icon.

[naddy]

Yes, it does.

[lvt]

(10-20-2024 02:28 AM)naddy Wrote:  Yes, it does.

What model precisely?

My gold HP 12C has the same screen as the 15C and it doesn't have the battery icon.

[ThomasF]

(10-20-2024 09:15 AM)lvt Wrote:  

(10-20-2024 02:28 AM)naddy Wrote:  Yes, it does.

What model precisely?

My gold HP 12C has the same screen as the 15C and it doesn't have the battery icon.

All original Voyager models had a low battery warning, but it is not a normal icon, i.e. it is not part of the display test shown after running the selftest.

From the HP15C manual (page 260 in the English Owner's Handbook, page 105 in the 10C Owners Handbook):
   

Check this picture: HP15C with low bat warning

Cheers,
Thomas

[rprosperi]

(10-19-2024 07:20 PM)naddy Wrote:  I just noticed another tiny difference between the CE and the original 15C:

When you run the 1L self-test on the CE to turn on all display segments, this includes the low battery indicator. By comparison, the corresponding end of the ON/× test on the original Voyagers does not turn on the battery indicator.

If this has been mentioned before, it's not mentioned in Gene's pinned summary.

Nice catch, and interesting observation!

The original classic Voyager test sequence is run by the original ROM code, turning on various combinations of LCD segments (or all at once) but never the low-power segment, likely to not mislead the user about battery conditions, but possibly that segment was controlled independently. I always thought the original [ON][/] sequence (lighting various segments for each key) weak as one can't tell what might be missing, whereas the [ON] [X] sequence lights them all, so easy to see if something is missing.

The 15c CE test sequence on the other hand is done by the emulation layer code, turning on all segments independently of the original ROM code, and this code simply ignores the old sequence, as can be seen by it's different approach of turning all on and then turning them off via pressed keys, which is more useful IMHO.

More than likely, the author of this new code never even noticed this subtlety.

[J-F Garnier]

May I add a not-so-subtle difference, not yet reported, that I recently noticed while testing something else?

On the genuine HP-15C, GSB [label] (or [A]..[E] in user mode) displays the target step while the [label] key is held down.
On the 15c LE/CE, the program execution starts immediately.

Well, no dramatic consequences, I agree.

J-F

[naddy]

(10-20-2024 09:53 AM)J-F Garnier Wrote:  On the genuine HP-15C, GSB [label] (or [A]..[E] in user mode) displays the target step while the [label] key is held down.

Same for f A..E in non-user mode.

[naddy]

(10-20-2024 09:53 AM)J-F Garnier Wrote:  On the genuine HP-15C, GSB [label] (or [A]..[E] in user mode) displays the target step while the [label] key is held down.
On the 15c LE/CE, the program execution starts immediately.

I can't decide whether that's a bug or a feature. If you call label, the next step is always going to be LBL label. Showing it is redundant. The exception would be GSB I, although people hardly use that outside a program. However, the CE does show the target step for SOLVE label and ∫ label, where it is equally redundant.

[J-F Garnier]

(10-21-2024 01:40 PM)naddy Wrote:  I can't decide whether that's a bug or a feature. If you call label, the next step is always going to be LBL label. Showing it is redundant. The exception would be GSB I, although people hardly use that outside a program.

The difference very likely comes from a minor bug in the emulation layer. I can bet it was not intentional.
And it remained unnoticed since the 2012 HP 15c Limited Edition ! (unless I'm wrong)

Now for the usefulness of the original design, I believe it comes from the same logic that makes the target step displayed when R/S , SST or BST are held down.

J-F

[naddy]

(10-21-2024 02:49 PM)J-F Garnier Wrote:  Now for the usefulness of the original design, I believe it comes from the same logic that makes the target step displayed when R/S , SST or BST are held down.

Yes, it was intended for those and the other functions incidentally feed into the same code path.

[J-F Garnier]

(10-21-2024 01:40 PM)naddy Wrote:  I can't decide whether that's a bug or a feature. If you call label, the next step is always going to be LBL label. Showing it is redundant. The exception would be GSB I, although people hardly use that outside a program. However, the CE does show the target step for SOLVE label and ∫ label, where it is equally redundant.

I can think of a possible usage. Image you have multiple labels of the same name:

Code:

001 LBL A
002 2
003 *
004 R/S
005 LBL A
006 2
007 /
008 R/S

First GSB A will show step 001, next one will show step 005.

OK, it's an ad-hoc justification :-)

J-F

[brouhaha]

(10-20-2024 09:43 AM)rprosperi Wrote:  The original classic Voyager test sequence is run by the original ROM code, turning on various combinations of LCD segments (or all at once) but never the low-power segment, likely to not mislead the user about battery conditions, but possibly that segment was controlled independently.

Yes, on all of the Voyager models that used custom HP chips, rather than commercial microcontrollers (SunPlus/GeneralPlus or Atmel/Microchip), the low battery indicator was controlled exclusively by hardware in the display driver, and the ROM code had no way to control it.

Quote:The 15c CE test sequence on the other hand is done by the emulation layer code, turning on all segments independently of the original ROM code, and this code simply ignores the old sequence, as can be seen by it's different approach of turning all on and then turning them off via pressed keys, which is more useful IMHO.

In the microcontroller-based Voyagers, with the low battery segment controlled by microcontroller ROM code, it was necessary for the display self-test to include the low battery segment, as there would be no other practical way to test it in production.

Quote:More than likely, the author of this new code never even noticed this subtlety.

That's possible, but even if the author recognized it as a difference, it still would be a necessary change.

[lvt]

(10-20-2024 09:37 AM)ThomasF Wrote:  

(10-20-2024 09:15 AM)lvt Wrote:  What model precisely?

My gold HP 12C has the same screen as the 15C and it doesn't have the battery icon.

All original Voyager models had a low battery warning, but it is not a normal icon, i.e. it is not part of the display test shown after running the selftest.

From the HP15C manual (page 260 in the English Owner's Handbook, page 105 in the 10C Owners Handbook):


Check this picture: HP15C with low bat warning

Cheers,
Thomas

Thanks for the info, the asterisk is there indeed.