▼
Posts: 180
Threads: 23
Joined: Apr 2012
Hi all (*really* long time, no post!):
I just received a WP-34s that I purchased from hpcalc.org. From the start, I have noticed problems with it not registering some of my keystrokes. It does not seem to be limited to one or two *keys*, but rather, it *looks* like the keypad-reading loop in the FW is missing my keystrokes. But I didn't see anyone else complaining about this issue, which would lead me to believe it is *not* a FW issue.
I sure hope that I can get this resolved, because the WP-34s is the *incredible*! It would hands-down be my every-day, go-to calculator, if not for this key-press problem; it just isn't practical to have to visually verify that every keystroke registers.
Any ideas, anyone? Did I just get a lemon underlying HP-30b? Or is this a known FW issue?
Much thanks,
Bruce.
▼
Posts: 180
Threads: 23
Joined: Apr 2012
Sorry; I forgot to mention the FW level: 2769.
Thanks,
Bruce.
Posts: 875
Threads: 37
Joined: Jul 2005
Bruce,
There have been similar reports and there has been some discussion of this issue. I believe Marcus has looked into keypress handling, but don't know if it was ever fully resolved if this is a hardware or firmware issue. Perhaps Marcus will comment.
Posts: 3,283
Threads: 104
Joined: Jul 2005
I'm the one who wrote the keyboard scan. It should be reliable and I don't have any problems with it. There are complaints from some people which I cannot verify here. I can recommend two things:
- Try your device with the original 30b firmware: Do you have similar problems?
- Try a different device with the 34S firmware: Do you have the same trouble?
A possible reason for keyboard problems is a misplaced overlay obstructing the path of the keys. The 30b is especially picky because the keys rotate and need some headroom directly above the key. If the sheet is glued a little too far 'south' the keys cannot move freely. A sharp blade can help here.
If you are a very fast typer it may help to reactive a special speed-up version of the firmware for you which increases the processor clock when keystrokes start filling the buffer. I hope you have the cable so you can re-flash your device.
▼
Posts: 180
Threads: 23
Joined: Apr 2012
Marcus:
Thank you for the detailed response.
Unfortunately, I don't have a cable. I just e-mailed Gene. I hope that he still has some available. If not, are there schematics such that I could make one?
I have access to both another WP-34s (due to my greedy/excited response resulting in the purchase of two), and I also bought a spare 30b from Amazon, so I will definitely perform some experiments.
I don't believe it is a problem of keystrokes filling the buffer, because it can happen on the first or second key-press. It seems to me that it is either flaky hardware, or a FW architecture that can miss a quick keystroke. I'm getting the impression that the FW is not at fault---otherwise many people would be complaining about what I'm seeing.
I know that you have probably heard this many times before, but I want you to hear it again: To you, Pauli, Walter, Eric, and all the others who have invested serious time into making the WP-34s a reality, I owe a *HUGE* debt of gratitude! I haven't followed the MoHPC forum for 10 years---because of the sad decline in HP calculators---but I am now greatly excited about the WP-34s and the other custom-HW / custom-FW offerings! I've worked with enough embedded-FW people to realize the effort involved. I'm heading over to the donation page on Sourceforge right now. You guys have really made a mark!
▼
Posts: 3,283
Threads: 104
Joined: Jul 2005
You missed much of the fun watching the (positive) development of the HP calculator division in the last few years. Without open minded people at HP, namely Cyrille & Tim, giving away enough information abut the 20b/30b line of calculators, the WP 34S would have remained an academic exercise of W & P.
I appreciate your warm words. I'm still hoping the keyboard issues can be sorted out. Are you a 'quick typer' or do you experience the problems even on the first keystroke? I'm asking because I want to make sure the firmware is behaving properly. The debounce algorithm is aimed at debouncing the 'up' state of a key. The reasoning behind it is that the rotate & click mechanism provides a tactile feedback which demands immediate reaction if the key closes the first time. To avoid bouncing, the time a key needs to be up is increased (to cycles of 10ms).
Edited: 16 Apr 2012, 4:54 p.m.
▼
Posts: 3,229
Threads: 42
Joined: Jul 2006
Quote:
Without open minded people at HP, namely Cyrille & Tim, giving away enough information abut the 20b/30b line of calculators, the WP 34S would have remained an academic exercise of W & P.
We wouldn't have started if the development kit wasn't available.
- Pauli
▼
Posts: 180
Threads: 23
Joined: Apr 2012
Pauli:
Everyone involved has done a terrific job; there should be many proud engineers at HP and Agilent!
Thanks for all your hard work,
Bruce.
Posts: 3,283
Threads: 104
Joined: Jul 2005
Pauli, forgot your early designs of a scientific calculator? The history dates back before the advent of the 20b SDK.
▼
Posts: 3,229
Threads: 42
Joined: Jul 2006
Yes, the scientific 4-banger. I'd love to see it in production :-) However I knew it wasn't going anywhere at the time.
Once the 20b dev kit came out, I figured things might go somewhere so I contacted Walter. The rest is history.
- Pauli
Posts: 4,587
Threads: 105
Joined: Jul 2005
And there were some other nice designs I'd also still love to see in production. Well, we will be working on the successors of them once.
Posts: 180
Threads: 23
Joined: Apr 2012
Marcus:
The missed keystroke can happen on the first keystroke. I will test-out a 30b and another 34s, and let you know what I find.
Thanks for all your help,
Bruce.
▼
Posts: 3,283
Threads: 104
Joined: Jul 2005
I fear you have a less then perfect hardware. :-(
|
