I finally received my WP34S, cable, xtal and overlay. It took less than a week but it seemed much longer to me ;-) Haven't installed the xtal and capacitors yet, wanted to play "stock" for a while. Re-flashing it with ver 3 (full) went flawlessly (I still keep and old computer with an actual RS232 interface for this type of tasks). Overlay installation didn't go that well, I ruined it, but because I know myself I had ordered two sets and the second time went better, almost perfect. The key top labels was a very easy task (using tweezers of course). I'm surprised with all the efforts put into this project, this has been an excellent work, congratulations to the entire team.
My only problem so far is that the calculator is crashing (freezing altogether) from time to time when doing simple calculations (nothing fancy or elaborated), I wasn't able to find any pattern yet :-(, I realize this is still a beta code, I'd just like to know if someone else is experiencing it or just me. The only way I found to recover it is either a battery pull or pressing the reset button in the back, I've tried almost all the combinations of [on]+[any other key] with no avail.
Thank you in advance.
Ooops, you are right, I didn't flash the latest revision... my bad. Thanks!
I apologize for having put an unstable built into the release ZIP file. I should update it.
Oh, no problem, I should have checked first. It hasn't crashed so far since I updated the FW. :-)
Thank you!
Unfortunately, there were a couple of slightly less than stable versions. The current firmware seems pretty stable.
- Pauli
Compared to the 35s (BTW, I already put it away!) I can consider the WP34S totally bug free! I already tested the program that didn't work on the 35S (due to the jump backward in the program when using Solve) and this one passed the test!
Thank you!
The bugs in the 34S will be different to those in the 35s.
- Pauli
When I read that I have to put in a capacitor, I blanched. But I understand from you that it isn't required?
It's an option - you will need it for an exact real time clock. See TICKS, TIME, etc. in the manual.
No, you don't have to do the crystal modification. As far as I know the only features you can't use without the modification are the stopwatch and comunication to other 34s' or the PC.
Quote:
the only features you can't use without the modification are the stopwatch and comunication to other 34s' or the PC.
The latter isn't true. Please see pp. 108f of the manual for a procedure.
Even the stopwatch works without a crystal but it is not very accurate.
The released image with stopwatch support won't run without an installed crystal.
- Pauli
Quote:
The released image with stopwatch support won't run without an installed crystal.
- Pauli
You mean someone in the team destroyed part of my hard work? Oh my...
Fortunately, it is easy for a beginner to recompile a better image itsefl :-)
Edited: 25 Mar 2012, 4:00 a.m.
No, Marcus removed the non-crystal support from the stop watch images.
- Pauli
Quote:
No, Marcus removed the non-crystal support from the stop watch images.
- Pauli
I know, I was just joking.
It was just a convenience decision to have more than the 'official' build which does not include the stop watch. I thought it might be useful to have it but for space reasons, the non XTAL option had to go. ;-)
Could I not make a crystal-less stopwatch calibration mechanism similar to the serial port one? It would require a bit of code and a byte of flash space to store the calibration constant.
Quote:
Could I not make a crystal-less stopwatch calibration mechanism similar to the serial port one? It would require a bit of code and a byte of flash space to store the calibration constant.
I had planned to work on it and it still on my list. The problem was the lack of free flash space. But it is also possible to insert the calibration constant in the firmware by editing the .bin file.
If you want to implement a calibration function, feel free to do so and to contact me by email to talk about it.
Quote:
The latter isn't true. Please see pp. 108f of the manual for a procedure.
I can confirm this since I do it all the time. Until the release of the Qt-based emulator for Mac, I had no problem sending RAM contents back and forth between Windows-based emulator and the calc. Now with the Mac emulator I can easily do likewise--indeed, the interface is even simpler as one selects the serial port in Preferences, whereas under Windows one has to create a wp34s.ini file containing the name of the COM port.
Right now, the only glitches I encounter seem related to the Mac emulator's older firmware. For example, if I compile a state file using a wp34s.dat created by the Mac emulator as a "seed", the revised wp34s.dat file ceases to be readable by the Mac emulator, though it works fine in the Windows emulator with much newer firmware. I am looking forward to Pascal compiling and posting a newer version to Sourceforge.
This is the main reason to keep the classic Windows emulator alive. It should be always the same revision as the firmware. The Qt based emulators typically lag behind.
Quote:
This is the main reason to keep the classic Windows emulator alive. It should be always the same revision as the firmware. The Qt based emulators typically lag behind.
I can perfectly make an automatic build and probably even an automatic upload for the Qt emulators.
I ask you kindly to build more often, be it automatically or manually. The problem is that the release ZIP files should match each other's version. Maybe you upload the current builds somewhere else (just create a folder below the emulator directory for the most recent builds). I will keep you posted on any updates to the release binaries. I do them typically after a major bug had surfaced and has been fixed such as recently the bug in the solver causing a crash.
As I said, it is very easy for me checkout & build the 3 distributions for any svn version. Just let me know when you want me to.
So far, the Qt emulator had a beta status in my mind. Once it is really used, we should move to a build/distribute production model.
Edited: 30 Mar 2012, 4:43 a.m.