I stumbled across Arduino a few years ago when a class that I teach needed to upgrade from the ancient HandyBoard micros. The technician decided on Arduino and it's been great. Cheap, simple, open-source.
I've also added Arduino to my toolkit for test and measurement. I use to generally use PICs when a PC running custom software (usually Labview) wouldn't do it all. Arduino is cheap, simple and generally up to the challenge. Microchip just came out with an Arduino compatible board called chipKit, which runs in the Arduino development environment, is 32-bit, is much faster... and a few dollar cheaper. We've got one on order to check out.
What about Arduio for test and measurement? My company has been working with the idea of using Arduino for super cheap data acquisition for our clients. We've put a version of the code on our website, in case anyone else is interested in trying out super low-cost Arduino based data acquisition. We call it LArVa, because it originally linked Labview to Arduino, but it's becoming clear it can do more than just that:
I have used PIC's for years using PIC Basic Pro to develop my software. The latest project used a PIC16F887 to replace a control cabinet incorporating a PLC and a PC. The 18Cubic foot baox was replaced with the PIC and all in a 0.5 Cubic Foot enclosure. Not only did the system replace the biger one, but it also did more! Development time for the hardware and software was 3 months compared to 24 months for the larger system. The PIC's have a huge amount of available I/O and are easy to program. On Board Programming makes it simple to update and prototype a system, although you can pull the chip and pop it in an external programmer if needed.
I started with the Arduino. Lots of on-line support, programable on any OS. Reasonable prices, reasonable performance, compliant with most standards, lots of sensors. Free c-like developement environment (IDE). Very easy to prototype.
I am not so impressed with Microchip's products or IDE. The developement board I received had pin 1 marked *under* the connector. It took weeks to get a response as to where they had hidden pin 1. Developement boards are physically large and some support Daughter Boards for different MPUs, some don't. There are 3 different IDEs (although the new one in BETA should eventually program all chips) it is frustrating to download and install only to find out you picked the wrong IDE. IDE is C and project oriented and PC only (BETA one should be universal). Lots and lots of chips and chip famalies. The bottom line is that you will need to know a lot about what you need and a lot about what Microchip offers to get started. I found the website had plenty of information, but is not organized well for getting started.
Thanks for this information, Jon. I'm always on the lookout for simplified MCU solutions for mechanical engineers. Another option I came across last year is the Texas Instruments experimenter board for their low cost MSP430 line of MCUs. The kit is $4.30 + shipping. You need to know C++ and it gave me a chance to learn it via a .. For Dummies book. For those interested... http://processors.wiki.ti.com/index.php/MSP430_LaunchPad_(MSP-EXP430G2)
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.