This is a very minor gripe: when pressing the rolldown button 3 times in a row fast (but not that fast, as fast as one would press it when deliberately pressing the same button 3 times), it sometimes doesn't respond to the third press.
I think the CPU should be fast enough to do the rolldown faster, not sure what causes the lag here, but it'd just feel slightly better if you didn't have to hold yourself back pressing buttons a bit slower to make sure they register.
Slight lag on the rolldown button
Re: Slight lag on the rolldown button
I'm not experiencing this problem on my DM32 unit. Perhaps it's a hardware problem with your particular keyboard.
DM15L, S/N 00548. DM42, SN: 00159. DM41X, SN: 00973. DM32, SN 00054.
Re: Slight lag on the rolldown button
Wow, Aardwolf is right... my rolldown key also misses a keystroke most of the time when it's pressed three times rapidly. HOWEVER, it's not just the rolldown key, it's ANY key, e.g. x<>y and even ENTER! The timing is crucial; all three keystrokes will be executed correctly if the keys are pressed either slowly enough or fast enough. But there's definitely a small range of timing which causes one keystroke to be missed.
Try this: clear the stack, then put 88888888 in X. Now press x<>y very rapidly exactly 3 times. If the 88888888 swaps position from where it started (as it should) then try again with slightly different timing (faster or slower). Eventually you'll find a keystroke speed at which the 88888888 will end up where it started (that is, swapping only twice instead of 3 times)... evidence of a missed keystroke.
My WILD GUESS is that the "keystroke preview" code is causing the problem, since that code only is executed when a key is held down for a very brief amount of time. If another keystroke happens between that moment and the preview finishing, I suspect that this keystroke will be missed. Just a guess, but I have a strong hunch about it.
Re: Slight lag on the rolldown button
Yes, I finally was able to reproduce this behavior, but only intermittently. At no time did I see the preview.Joe Horn wrote: ↑Sat Jun 10, 2023 9:13 pmTry this: clear the stack, then put 88888888 in X. Now press x<>y very rapidly exactly 3 times. If the 88888888 swaps position from where it started (as it should) then try again with slightly different timing (faster or slower). Eventually you'll find a keystroke speed at which the 88888888 will end up where it started (that is, swapping only twice instead of 3 times)... evidence of a missed keystroke.
DM15L, S/N 00548. DM42, SN: 00159. DM41X, SN: 00973. DM32, SN 00054.
Re: Slight lag on the rolldown button
I also think this might be related to code for handling the long press causing this as a bug side effect, but not certain of course.
I tried the latest firmware (DMCP5_flash_3.50t31_DM32-2.05.bin) and the issue is still present.
A way to reproduce this: put the values 3, 2, 1 and 0 on the stack in that order. Look at the 0. Press rolldown 3 times in rapid succession. You should see the 3 where the 0 was before if it rolled down 3 times, but sometimes it's only the 2 or the 1.
For reference: I have a super low powered solar-power only Casio fx-82 Solar II calculator. Just on solar power, I can enter a value like 2, and hit the x^2 button in as rapid succession as a human hand could possibly do (to be clear, with the DM32 the issue already occurs at less extreme speeds), it registers every single press, squaring the number every single time.
So the DM32 buttons are for sure laggy, and while careful use is fine, it is just at the border of where you may hit on this during normal usage at regular human button press speed.
The DM32 also has a much more powerful CPU than the Casio Solar, e.g. computing combinations of 99 with 55 goes immediately on the DM32 while it takes a second on the Casio. So the button lag is not a CPU speed issue.
I tried the latest firmware (DMCP5_flash_3.50t31_DM32-2.05.bin) and the issue is still present.
A way to reproduce this: put the values 3, 2, 1 and 0 on the stack in that order. Look at the 0. Press rolldown 3 times in rapid succession. You should see the 3 where the 0 was before if it rolled down 3 times, but sometimes it's only the 2 or the 1.
For reference: I have a super low powered solar-power only Casio fx-82 Solar II calculator. Just on solar power, I can enter a value like 2, and hit the x^2 button in as rapid succession as a human hand could possibly do (to be clear, with the DM32 the issue already occurs at less extreme speeds), it registers every single press, squaring the number every single time.
So the DM32 buttons are for sure laggy, and while careful use is fine, it is just at the border of where you may hit on this during normal usage at regular human button press speed.
The DM32 also has a much more powerful CPU than the Casio Solar, e.g. computing combinations of 99 with 55 goes immediately on the DM32 while it takes a second on the Casio. So the button lag is not a CPU speed issue.
Re: Slight lag on the rolldown button
Thanks for the feedback.
You must have super human reflexes, as I could not reproduce this.
In 20 tries, I got the proper result of 3.0.
I'd encourage other readers to also try and if this is a common problem for a reasonably large number of users, we can look at it.
Thanks
You must have super human reflexes, as I could not reproduce this.
In 20 tries, I got the proper result of 3.0.
I'd encourage other readers to also try and if this is a common problem for a reasonably large number of users, we can look at it.
Thanks
--bob p
DM42: β00071 & 00282, DM41X: β00071 & 00656, DM10L: 071/100
DM42: β00071 & 00282, DM41X: β00071 & 00656, DM10L: 071/100
Re: Slight lag on the rolldown button
I can confirm that I have the same rolldown issue.
Adding 3, 2,1 and 0 and performing the rolldown 10-12 times. About half the time I ended up with "2" instead of "3" in x-register.
(S/N 00045)
Adding 3, 2,1 and 0 and performing the rolldown 10-12 times. About half the time I ended up with "2" instead of "3" in x-register.
(S/N 00045)
Esben
DM42 SN: 00245, WP43 Pilot SN:00002, DM32 SN: 00045 (Listed in obtained order).
DM42 SN: 00245, WP43 Pilot SN:00002, DM32 SN: 00045 (Listed in obtained order).
Re: Slight lag on the rolldown button
I was able to reproduce it on the swap key - pressing 3 times rapidly but only getting two swaps. But as for whether it is actually a problem is a different question: I never press keys at that rate so I'm fairly certain that it's never going to be a problem for me.
Re: Slight lag on the rolldown button
http://www.emyamstein.ch/media/swissmic ... 30_cut.mp4
12 presses on R↓, 25 times in a row. I get one miss on the last round. Which equates to a little bit more than 1:300 presses (very far from 1:2... )
I know that keypad scanning is already quite a complex thing. I'm not sure, but I'd venture there's too much work for too little benefit in trying to implement anything faster than what we have.
EDIT: yes the name on the glass of my unit is wrong; I still have the prototype unit from when the hardwre wasn't completely ready...
12 presses on R↓, 25 times in a row. I get one miss on the last round. Which equates to a little bit more than 1:300 presses (very far from 1:2... )
I know that keypad scanning is already quite a complex thing. I'm not sure, but I'd venture there's too much work for too little benefit in trying to implement anything faster than what we have.
EDIT: yes the name on the glass of my unit is wrong; I still have the prototype unit from when the hardwre wasn't completely ready...