WP43 News
Re: 43S News
Thanks, HPMike, for reporting. IIRC we had a similar problem once before. I put it on our issue list and come back to you when I've got an answer. Thanks in advance for your patience.
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
Re: 43S News
Hi,
I have just realized that CLX doesn't always clear (make it zero) the x register:
Try 8 ENTER 9 CLX, and you will see the previous 8 instead of a 0. Here CLX behaves as an UNDO of the previous input.
Try 5 ENTER 8 + .You will have 13 in the X register. Then press 5, and the 13 goes to Y. Then CLX and the 13 returns to X. Here CLX behaves as DROP.
This seems unconsistent to me. Tried with my 41, and the behaviour of CLX is always consistent. It puts a 0 in X without disturbing the stack.
BTW, could anybody point out any advantages of CLX over DROP in a 4 line display machine? Aften many (many) years using CLX, I was very happy with the 50G having DROP instead of CLX as the default way to clean X (and the stack if pushed enough times).
Regards,
Manuel.
I have just realized that CLX doesn't always clear (make it zero) the x register:
Try 8 ENTER 9 CLX, and you will see the previous 8 instead of a 0. Here CLX behaves as an UNDO of the previous input.
Try 5 ENTER 8 + .You will have 13 in the X register. Then press 5, and the 13 goes to Y. Then CLX and the 13 returns to X. Here CLX behaves as DROP.
This seems unconsistent to me. Tried with my 41, and the behaviour of CLX is always consistent. It puts a 0 in X without disturbing the stack.
BTW, could anybody point out any advantages of CLX over DROP in a 4 line display machine? Aften many (many) years using CLX, I was very happy with the 50G having DROP instead of CLX as the default way to clean X (and the stack if pushed enough times).
Regards,
Manuel.
Re: 43S News
Thanks for asking. May I ask back? Did you use CLR CLX or [<-] ?
And about CLX (the real one) vs. DROP: CLX is a command with maximum tradition; if you delete it, you get many new friends. DROP together with automatic stack lift is (IMO) the cleaner solution, but YMMV. So we implemented both.
- 8 ENTER 9 [<-] returning 8 in X and Y looks logical to me. You just deleted an open entry.
- 5 ENTER 8 + 5 [<-] returns 13 which is ok since you deleted an open entry again.
- 8 ENTER 9 CLX returns 0. in X as you have expected.
- 5 ENTER 8 + 5 CLX returns 0. in X as you have expected.
And about CLX (the real one) vs. DROP: CLX is a command with maximum tradition; if you delete it, you get many new friends. DROP together with automatic stack lift is (IMO) the cleaner solution, but YMMV. So we implemented both.
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
Re: 43S News
Hi Walter,
Thank you for your reply.
I can't see a complaint in my post at all. Maybe too much radioactivity there? I was pointing out what I (still) consider an inconsistency.
I was effectively using the back arrow. From the Owner's Manual, Pag 319, I understand (obviously I am wrong) that key performs the function of backspace until the last digit of a number to delete, and then it calls the CLX function. This is the behaviour I also was expecting as it is the one I've experienced in every RPN machine I've used since hp added the backarrow feature back in 1979. Call me nostalgic if you want, but that worked flawlessly.
Considering that a number being introduced is a "temporary information displayed" just as a warning message, is an error under my view, as it forces you to use the real CLX from the menu (3 keystrokes to clear X). This could be acceptable in a calculator with a separate entry line (a la RPL), but not very practical nor consistent in one in which you enter data directly in X.
Of course this is only my opinion.
BTW, I also think DROP together with automatic stack lift is the cleaner solution, but it is 2 keystrokes away, and it can't be assigned to the backspace key behaviour AFAIK.
Anyway, thank you for this interesting proyect.
Mit freundlichen Grüßen.
Manuel.
Thank you for your reply.
I can't see a complaint in my post at all. Maybe too much radioactivity there? I was pointing out what I (still) consider an inconsistency.
I was effectively using the back arrow. From the Owner's Manual, Pag 319, I understand (obviously I am wrong) that key performs the function of backspace until the last digit of a number to delete, and then it calls the CLX function. This is the behaviour I also was expecting as it is the one I've experienced in every RPN machine I've used since hp added the backarrow feature back in 1979. Call me nostalgic if you want, but that worked flawlessly.
Considering that a number being introduced is a "temporary information displayed" just as a warning message, is an error under my view, as it forces you to use the real CLX from the menu (3 keystrokes to clear X). This could be acceptable in a calculator with a separate entry line (a la RPL), but not very practical nor consistent in one in which you enter data directly in X.
Of course this is only my opinion.
BTW, I also think DROP together with automatic stack lift is the cleaner solution, but it is 2 keystrokes away, and it can't be assigned to the backspace key behaviour AFAIK.
Anyway, thank you for this interesting proyect.
Mit freundlichen Grüßen.
Manuel.
Re: 43S News
Hi Manuel,
Seems I've to explain a bit:
Seems I've to explain a bit:
- If somebody writes that a command (e.g.CLX) doesn't work as expected then this is a complaint about this command IMO. A complaint is nothing bad - to the contrary it's good since it allows for clarification or correction.
. - You point to p. 319. I've copied the respective section plus the corresponding instruction on p. 316:
.
What is written there? If you press [<-] and there is an open input then it deletes the last character entered. If none is left, it cancels pending command like EXIT and returns. Finished. (It would only call CLX if none of the upper 5 conditions in the table is fulfilled.) This is exactly what you have observed. - Now we can discuss whether this is the correct functionality. Then we have to take care of all the conditions users may press [<-] ...
. - The designers of the Voyagers knew the problem: .
- In the HP-42S you need 3 keystrokes for calling CLX.
. - I don't see any reason why you can't assign DROP to [<-] but maybe I've overlooked something.
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
-
- Posts: 1102
- Joined: Tue May 02, 2017 5:48 pm
- Location: Netherlands
- Contact:
Re: 43S News
Same here for part 1 of your statement. Thanks for part 2 - will think about it (cf. points 2 and 3 of my post).Thomas Okken wrote: ↑Thu Mar 24, 2022 9:13 pmIn RUN mode, ← executes CLX if you're not in number entry mode, and also if you are in number entry mode and deleting the last character.
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
Optical bug in EQN
One observation / optical issue I stumbled upon in the current version (0.22.3) of the windows simulator (may have been present before, though):
When entering a new / editing an existing equation in EQN-mode, the arithmetic operator for multiplication displayed changes from a dot (∙) to the cross (x) as soon as the equation becomes "too long".
Being "too long", apparently, requires
1) the equation to be labelled, i. e. it has to have a name separated from the actual equation by a colon (:) and
2) pushing that colon out of the displayed range during the input of the equation.
However, this only seems to have an effect on the visual representation of the operator, not the actual execution of the solver or anything connected to it, so it can be considered merely cosmetic.
When entering a new / editing an existing equation in EQN-mode, the arithmetic operator for multiplication displayed changes from a dot (∙) to the cross (x) as soon as the equation becomes "too long".
Being "too long", apparently, requires
1) the equation to be labelled, i. e. it has to have a name separated from the actual equation by a colon (:) and
2) pushing that colon out of the displayed range during the input of the equation.
However, this only seems to have an effect on the visual representation of the operator, not the actual execution of the solver or anything connected to it, so it can be considered merely cosmetic.
Re: 43S News
Just verified. 8 points for you!
EDIT: Will be solved with next release (0.22.4).
EDIT: Will be solved with next release (0.22.4).
Last edited by Walter on Sat Mar 26, 2022 8:12 am, edited 1 time in total.
WP43 SN00000, 34S, and 31S for obvious reasons; HP-35, 45, ..., 35S, 15CE, DM16L S/N# 00093, DM42β SN:00041
More problems with calculator losing contents of memory
Today I fired up the emulator, and everything that I entered yesterday was gone. There was no error message to indicate that this should have happened. Well, almost everything. There are two 3x3 matrices A and B that are now null (all zeros), which was the original configuration until I added data, so maybe that was saved and all subsequent changes were not saved. The backup.bin file has yesterday's time stamp from when I first started working, so maybe it's not saving the status like it should when the emulator is exited. Is there some way to manually execute an update to the backup.bin file or otherwise save the calculator's memory ?
DM15L, S/N 00548. DM42, SN: 00159. DM41X, SN: 00973. DM32, SN 00054.