Goodbye

[Scampy]

You are saying its not practical to program an Arduino supported chip and place it into a custom designed PBC as you would a PIC?
If so is it common for Arduino projects to be made up of jumpered development modules to make up the final project?

To a degree yes. The ATMEGA 2560 is an 8 bit micro running at 16mhz, with 256 KB of program space, but its in a 100 pin TQFP package which makes programming and soldering to a custom hobby PCB more of a challenge compared to an 8 bit PIC which is in a DIP or SOIC package.

I've just ported a project that was originally developed around an 18F based PIC in DIP package. I developed a custom PCB with all the supporting components on it and then would program the PIC via my EasyPIC5 and then place that into the board. Porting this to the Ardiumo based platform the custom PCB had the majority of the supporting components on it, but the board was designed to plug directly into the header pins on the Mega2560 board, like a large custom shield (as can be seen here testing the TFT screen). The mega was a cloan purchased for £11 from Amazon, and cost less than the individual ATMEGA 2560 chip on its own.

[Normnet]

Got it...thanks

[Normnet]

For 3 simpler $25 PIC 8 bit plug in modules see Microchip Direct and do a search for PIM and check development tools then 8 bit. Probably not a lot of 8 bit due to they are available in DIP packages for easier prototyping.

[queenidog]

Thanks for all your replies. I wasn't going to come back but I was curious if anyone was agreeing with me...and some did.

I just bought 6 Arduino UNO boards for $12 (CAD) each, all taxes and shipping included. (They were shipped from Canada, but no doubt made in China). These are genuine UNO's made to the specification demanded by Arduino. As such it has a boot loader installed on it and it's just plug and play. I made an ultrasonic scanner using the inexpensive HC-SR04 and an UNO. How easy was it? In under 15 minutes I had one working, using the built in libraries and example code. then I started tweaking it with my limited knowledge of C++ at the time. How would this have been done with a PIC development system and PBP? I likely would have to write the code (with Arduino libraries it was very simple, like 5 lines of code), I would have to set the fuses and all that other overhead PICs require that Arduino takes care of. Once running, I would then have to make a board to have it standalone and I would have to provide an on-board power supply or converter. With the Arduino I downloaded the short program through USB, took the unit (with little breadboard on top) to another location, provided 12 volts from a Walwart and voila! There it goes. I added an I2C display later, again something that took very little time despite not using I2C before.

My current project uses 7-16 bit I2C expanders lighting 110 RGB LEDs. Each LED is addressable. Using the UNO there are only 5 wires interfacing to my project: 3 PWM DIO (so I can fade any light using PWM) for the RGB, and 2 wires for I2C, clock and data. I made a low power "model" that the UNO was quite capable of driving. I'm up to 16 pages of code to display the various patterns and have TONS of space left.

I've been using Arduino for over a year and have not had to update anything. It is very stable. When there is an update to a library, it's usually for more functionality (eg the Wire library for I2C and the Ultrasonic library for the SR04) than for bug fixes.

Having said that, the Arduino IDE is a bunch of crap! Little functionality. No debugger, I can't even PRINT my code. For that I have to cut and paste into Notepad++. The compiler is 10x SLOWER than the PBP compiler. THIS is why I was hoping I could use PBP, to eliminate those disadvantages. I'm sure Dave Taylor (MELabs) could have come up with something.

I've got 3 robots using Arduinos, and my desk has 3 Arduinos connected to 3 MORE projects. No more waiting for PCB to be made and discover the mistakes I made making them. Create something with Arduino and use the board in final stage.

The photo I attached shows a breadboard of the I2C interfaces for my display. The green board is one I had made. Yes, I had to make a board, but not the controller, seen at bottom right (Arduino UNO).

[Scampy]

I tried Amtel Studio 7 which is a very good alternative to the basic Arduino IDE, and offers a lot of extra tools.

As I've mentioned elsewhere, I too made the switch because you can get up and running on the arduino platform very quickly and have something running using existing examples and library files. Once you have those examples running, it's normally an east task to modify the example to use it to perform the functions your want. As mentioned I wanted to use a TFT screen for the next version of my thermostat project that originated back in 2009 and was based on an 16F877 and 4 x 20 LCD (and I couldn't have done that without DT and Hentiks help). It would have been a lot of hard work and beyond my expertise to write the code in PB or ASM in order to get it to work, yet on the arduino it was displaying text in around 20 minutes of hooking the screen up and loading the example code. I then played with the commands to make the screen more relevant to the project, and then added an example code for reading an 18B20 sensor and displaying the temperature on the screen... by the end of the evening I had the time and temperature displayed on the screen along with a few additional variables... It makes it so easy to develop projects, but as mentioned, it's base around using these boards (uno, nano, mega etc) at the heart and stacking plug in modules (shields) into them. Where as programming PICs often results in a single custom designed PCB that the PIC plugs into.

It's really a shame that PBP didn't keep up with the pace and support more upto date hardware with standalone include files....

[queenidog]

You and I are on the same wavelength. I remember your name from previous posts I made when I was a PIC user.
I think MikroElektronika would be the way to go. They have the best development system ever made (I had one), with support for their Basic, C++, etc. They also have "modules" though not as extensive as Arduino chaps. I didn't like the Mikro because the Basic was too structured, yet I have to adhere to that structure with my Arduino C++ !! Who knew? Despite that, Arduino is still the better "system" for quick development. I too added LCDs to my repertoire and use the same codes in most of my software for troubleshooting. Also, since even the lowly UNO has dedicated lines for I2C (clock and data), I have now migrated to I2C capable LCDs. Only 4 wires (inc power and ground) required to the LCD now instead of that ribbon cable mess we are used to.

[khoog]

We use Arduino as a tool to quickly check out sensors and other components. It's great for making blinky lights, but it's terrible for learning embedded systems programming. All you're doing is running programs and libraries that someone else wrote, no one knows how they work, and it's almost impossible to track down all the nested routines and libraries that are called for even the simplest tasks.

We don't use PB very much anymore, but we invested in PB3 to help support the product and stay up to date. When we found that none of our existing programs would compile with it (getting error messages that said "unspecified error" didn't help) we decided it wasn't worth spending the time to get it all running. Our PIC development is now all done in C, and we dropped back to 6.1 for the quick test routines.