Gerhard Fiedler wrote:
> But as I said (and Olin before), the same principle applies to small
> systems. It's independent of the absolute size of the system, and
> related to the relative size of the amount of data we're dealing with
> and the available resources.

I largely agree, but there is one distinction between small and big systems.
On big systems, I/O must be done for you by some operating system driver, or
often layers of drivers.  There can be considerable overhead just getting
into and out of the operating system, with the per byte overhead much
smaller once you're already in the inner loop.  For that reason, getting a
chunk of data from the OS at a time makes sense, even if the processing is
done sequentially eventually (which most of the time it is).

On small systems, there is no OS and getting to the routine that delivers a
byte is usually little overhead.  Now we can deal with I/O a byte at a time
up and down all the levels if that's how it's processed.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist