I think the pre-fab one is a write once affair except for the advanced users who know how to dissect it and put a new program on the chip. It would be nice if the scangauge let you do that, but I think they are too worried about pirating to make the hex file available for public download. The installation might seem less intimidating, like installing a tachometer vs building then installing a tach. It would get into more cars that way. But I'm willing to agree that it is still a tad early. If I can get a batch of PCBs sorted out, that would be a good start.
Let me go point by point with Bens comments, I think we are on the same page here.
1. Gracias, I hope it helped your record tank (very nice numbers there). I assume you were happy with Yoshi's initial vx figure?
2. BLUE LCD sucks, yup. I added a link to this guy in post 1. I've got a few, was hoping to order a few PCBs to try them out.
NHD-0216K1Z-FL-YBW
3. That has been on the todo list for a while. I'll bump it up a notch. I need to get off timer(0) and on to timer(2) with my own ISR first.
4. The screens just kind of happened
I don't know if there will be enough program space left once the "important" updates get made to have a fully user definable screen. And with this little real estate it will always be a compromise. It isn't what you were asking for, but one *could* add their own custom screen function to the screen list when they program it, or you can daisy chain mpguinos as much as you want with a simple "y" adapter:
5. I know, mentioned above. I am thinking about it a little bit, but not very deeply yet.
6. Tank or current resets? Current will reset after 7 minutes of inactivity. If tank is resetting occasionally when you use the starter then you need a better power connection to your battery, maybe run your power leads straight to the battery terminals and see if that helps.
PS: The injector high time and vss count is accumulated in an instance of Trip named "tmpTrip". To fix number three, (once I have my own timer ISR), we need to keep track of the time between vss pulses and compute the speed from there rather than from the number of pulses recieved in the last 1/2 second. I need to debounce the vss pulses properly first though so the metro doesn't need such a hacked vssPulsesPerMile figure (again dependant on the fancy schmancy timer ISR). Why don't I just quit messing around and do that?