Chief Delphi - Development enviroments orther than MPLAB

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)

- Programming (http://www.chiefdelphi.com/forums/forumdisplay.php?f=51)

- - Development enviroments orther than MPLAB (http://www.chiefdelphi.com/forums/showthread.php?t=26586)

Re: Development enviroments orther than MPLAB

I voted for MPLAB, but I just did MSC++ .NET, and I'm really happy. :D Now that's an IDE.

Re: Development enviroments orther than MPLAB

Quote:

Originally Posted by Venkatesh

Vim? Fwah.

I used vi (not vim) and make. Not GNU make either. Microsoft NMAKE.

If I had access to a Mac, I would use Subethaedit, but I don't. If somebody were to port Subethaedit to Windows, I would regard them in the ranks of Cerf.

Microsoft?!?!? :ahh:

As my old signature used to say, "I'm not anti-Microsoft, I'm just pro-quality." :)

Yeah, MSC++ .NET is great (surprisingly, a good Microsoft product). The only reason I didn't use it is MPLAB allowed quick access to the compiler, while MSC++ .NET isn't custimizable for that sort of thing. Next year I'm going to find something better for our controls team to use. :)

Re: Development enviroments orther than MPLAB

GVim/Vim + Cygwin + IDE for compilation. Don't have access to a linux computer other than at home for programming, and too lazy to figure out how to use MCC's command line.

Re: Development enviroments orther than MPLAB

Compared with the general rabble, Microsoft almost caters to developers.

Re: Development enviroments orther than MPLAB

Its too bad that only 3 voted for linux...Most of my team is affraid of it, including the other programmer, so im stuck using windows at the competitions...

Re: Development enviroments orther than MPLAB

I use ActiveState's Komodo a lot, otherwise Eclipse is worth a good look. Both are available under Linux or Windows. KDevelop under Linux is also worth a good look.

A lot of bots didn't work right off the bat at the early competitions, because people didn't follow the latest IFI/FIRST news about software updates and weird initial states upon RC startup.

As for NMAKE, it's piece of $@#$@#$@#$@#. Gnu MAKE is available for Windows as well. Though I think it's high time that somebody come out with a better MAKE one of these days.

Re: Development enviroments orther than MPLAB

We decided early on to scrap the MPLAB IDE and use cygwin & make instead. After a month or so I finally convinced the other mentors to use VIM. Now we're actually teaching the students how to use it.

An invaluable development tool I found a few years ago is cscope (http://cscope.sourceforge.net/). It parses all the source & header files in your build and creates a 'database' that allows you to quickly jump to a function definition, find all calls of a function, find all references to a symbol, etc. Even better you can integrate it with VIM so all you have to do is put your curser on a function call, hit <ctl>] and you're now at the function definition; hit g<ctl>\ and you have a list of all occurrances of the symbol under the cursor. I'd probably be half as productive at work if I didn't have VIM & cscope. Give it a try. If you need help send me a PM or start a new thread.

Re: Development enviroments orther than MPLAB

I got our code working the day of shipping this year (goes to show how having access to the robot is a good thing eh). Last year however, we never really had the time to test, so we got to our regonal, and didn't move for hte first 2 practice matchs. Seems auto code didn't work right, go figure. :p

Re: Development enviroments orther than MPLAB

Cool! I like that. MS has been doing it for awhile, but it's nice to now someone else is.

Re: Development enviroments orther than MPLAB

Quote:

Originally Posted by CrashZero

:ahh: Thats nothing all our robot did when we shipped it was spin in a circle. cuz the stupid build team put one of the fricking moters in backwards. I think that when we get there i can download a code that works but we will literally be testing it on our bot for the first time then. If anything goes wrong i am sure that all you programers out there know that it is me who will be blamed for screwing it up, even though if the builders whould have put it in right it would not have been a problem. :ahh:

You think that that's nothing. We shipped six pieces of metal with the rest of the requirements. And a few motors. :)

Re: Development enviroments orther than MPLAB

Ok I admit that is worse then what I had, but it is kinda funny; you must admit :D
But your team didn't know how well the code worked when you shipped yours, they knew that mine just made it spin around and around which was definatly not a confidence booster. But I am hopeing to have it compile :ahh: Today. This code that I have now better work or you may not see me at any compitions cuz my team will have killed me. :ahh:

Re: Development enviroments orther than MPLAB

Quote:

Originally Posted by CrashZero

Our code has exactly one place in the code that knows how to run the drive motors.

Code:

// Set the left and right drive motors to the specified speed values.

//

// The speed values must be in the range [-127, 127].

// For both motors:

//    0 is stop

//    postive values move that wheel forward

//    negative values move that wheel backward

//

void SetDriveMotors(int speedLeft, int speedRight)

{

        pwmWheelL = pwm01 + speedLeft;

        pwmWheelR = pwm02 + speedRight;

}

This simple encapsulation has many advantages.

This chunk of code is the only place that has to keep track of "bigger than 127 goes forward, less is backward, proportional to the difference" or the other way around if either motor is wired the other way.

It's much easier to read "SetDriveMotors(10, -10)" than "pwm01 = 137; pwm02 = 117;" and know that it's asking for a slow right turn.

If one of the motors gets wired backwards, it's trivial to change the +/- for that motor in one well-defined place. (We also put polarized connectors on the motors, so that motors can't get reversed for anything less than a motor replacement.) It's also trivial to move the drive motor to a different PWM if need be.

The software team gets so little time with the working robot, it doesn't pay to spend time and energy blaming the build/electronics team for something being backwards. If you keep the knowledge localized in the code, it's trivial to adjust, compile and download. This is especially handy if you just finished a big repair two minutes before the next match and there's no time to crawl back in and reverse a couple of wires.

Re: Development enviroments orther than MPLAB

Quote:

Originally Posted by Texan

I'm not anti-Microsoft, I'm just pro-quality

Hehehehe that's great. I used to use MPLAB, but it started to annoy the crap out of me so I switched to Crimson and now only go into MPLAB for compiling.

Re: Development enviroments orther than MPLAB

Quote:

Originally Posted by K0r34nB0y

Hehehehe that's great. I used to use MPLAB, but it started to annoy the crap out of me so I switched to Crimson and now only go into MPLAB for compiling.

don't even do that - mplab shows you the command lines it uses - just copy them into a batchfile.

Re: Development enviroments orther than MPLAB

eh... I'm lazy... BTW Does anyone know what the heck is up with this error in mplink?
"Error - could not find file 'clib.lib'" I keep getting that when I try to compile on my own computer. I checked and clib.lib is right where it's supposed to be. Does it matter that I'm using the demo of c18?