Hello Everyone,
It's late and I'm tired so I apologize about not replying individually to the bug reports, comments, and suggestions. The enthusiasm and encouragement is also much appreciated.
My poor head is not thinking perfectly straight tonight so I'll just try to weigh in on some of the discussion. I've found (and fixed) the bug Alain identified. I tried to push the text around on the form so none of it interferes with other objects. I've also moved the project from Delphi 4 up to Delphi 2006 (now a part of Borland Developer Studio) to see if that has any effect on the aesthetics. I want to see what else I can implement quickly before getting another version out, so it gets taken a little further instead of just a bug fix.
Bert has made the suggestions of creating a few mock microcontrollers to test out the various features. I support instead identifying a small handful of µCs that are representative. My suggestions are the 12F683, 16F628, 16F877, 18F452, 18F1330, and 18F67J50. Naturally these are the chips with which I have the most experience.
The sky is the limit for the application BUT I don't want to be working on this the rest of my life... (I'm happy to send the source to anyone on request.) There should be defined goals for the first "release" of the application. I would have been happy with setting the configuration fuses...8^) Plan ahead for features to go in subsequent versions. To paraphrase George Patton, "A good application today is better that a perfect application later."
In volunteer software projects "later" turns into "never" all too often so let's wind up with, at minimum, something useful even if it doesn't have everything.
I suggest putting the desired function list into two or three tiers--"Very Important," "Nice to Have," and "Would be Cool but..."
I am not the best programmer/electronics guy on the planet, probably not even in the top 50% of the (incredibly knowledgeable, helpful, gentle, and patient) contributors to this list. What I do have going for me is a broad base of general knowledge and the fact that I'm not really good at giving up on projects. With all of that said, any code, pseudo-code, logic, or even flow charts helping map out how to generalize any of these functions would make it go a lot faster, as the alternative is me plowing through a dozen or so databooks. I'm not averse to that, but it is slower.
And I gotta say, DT's ALLDIGITAL.pbp is a absolute work of art.
Best Regards (and apologies for rambling--again),
Paul
Bookmarks