Lpc2000

Hi Rob,

Since I'm still awake at this time of the night I'll answer quikly. The 
costs are only relative, so the bit-banged I/O is comparatively "free" I 
guess as compared to "$$$". A bit-banged transmit routine is about as 
primitive as you can get and most embedded programmers can whip up one 
in a very short space of time (gimme 15 mins or less). Compare that with 
interfacing a UART and getting it running right, I can absolutely 
guarantee that the bit-banged transmit is the cleanest, most simple, and 
cheapest option available.

Hell, over short distances you can even run the transmit output directly 
at logic levels into a PC without anything more than maybe a 
current-limit resistor. I've never seen this not work or cause problems, 
and since this is simply for debugging then there are no compliance 
issues to worry about.

my2cents worth of I/O
Peter Jakacki


Robert Adsett wrote:

>At 09:30 AM 10/19/04 +1000, you wrote:
>  
>
>>EXTRA SERIAL CHANNEL SELECTION GUIDE
>>------------------------------------------
>>EXTERNAL UART + CODE + INTERFACE    = $$$
>>EXTERNAL SPI UART + CODE            = $$$
>>EXTERNAL MICRO + CODE               = $
>>BIT-BANGED I/O + CODE               = free
>>    
>>
>
>That does make the assumption that pins and code are free and that the CPU 
>isn't already occupied with other time critical functions. ;).
>
>I don't know about the original poster but I place some value on my 
>time.  For a strictly debug it may well be cheaper to simple hang some HW 
>off a header to SPI or I2C even if the chips are more expensive than a 16550.
>
>Robert
>  
>