Lpc2000

I agree 100%!
Baby steps...baby steps...

In my case, what I did was that Keil "Hello World" example with their 
polled UART device driver.

Then, I quickly dropped in my interrupt-driven driver.

On the hardware front...basically, the director and I deemed that we 
could 'take the risk' and just go straight to our own hardware...

schedules vs. risk and all...
Turns out that we made the correct decision, since I was able to get 
that "Hello World" example with our custom hardware fully operational 
in a couple of hours of receipt of the boards!

oh...
Actually, we had one cut that we had to do on the board. It turned out 
that our hardware engineer felt that we had no use for the JTAG port, 
so he disabled it! (this...after we expressly told him not to...sigh!) 
so....we wound up having to cut a trace on the JTAG header to the 
processor in order to get the JTAG operational on our board...(sheesh! 
he gave us the JTAG header/connector, and wound up disabling the JTAG 
port! go figure!)

Ken Wada

--- In lpc2000@yahoogroups.com, "rtstofer" <rstofer@p...> wrote:
>
> The process is known as 'stepwise refinement' and I'll credit it to 
> Niklaus Wirth because I first saw it described in his 
> book "Algorithms + Data Structures = Programs".
> 
> First you get a development board with some type of development tool 
> chain.  It doesn't much matter which one.
> 
> Next you run the simple examples.  For an embedded platform this is 
> usually a blinking LED.  For a computer platform this is "Hello 
> World!".  Both are sufficient to show that the tool chain and 
> development board are working.
> 
> Then you study all of the files that make up the example.  The 
> startup code, the main program, etc.  You look at every statement 
> and, using the device datasheet and board schematic, try to learn 
> exactly what it does.
> 
> In the embedded world there will be some startup code you just take 
> on faith because it will be a long time before you need to worry 
> about it.  But, things like setting up the IODIR register and then 
> using IOSET and IOCLR to turn things on and off are fundamental to 
> microcontrollers.
> 
> Now that the blinking LED is understood, the next step is to get the 
> serial port working and create the "Hello World!" example.  This may 
> be included with the software.  Nevertheless, it is the second step 
> and it is important because, while JTAG debugging is glamorous, 
> printf debugging is also useful.
> 
> Then it is time to move on to I2C or SPI.  Perhaps hang an MMC/SD 
> card on the bus and implement a file system.  Maybe hang an MP3 
> CODEC on the SPI bus and build a complete MP3 player.  That ought to 
> do it:  there is a little bit of everything in a project like that.
> 
> One step at a time...
>