Hi Everybody,

Just a short message really in response to something Jay wrote in one of my other subject folders (yet again Jay, you've said something which has sent my brain into overtime ;-) ).

It seems to me that the main reasons people use microcontrollers in their designs are, to keep component count down, to keep complexity down and to keep cost down. This obviously doesn't come as a great supprise seeing as that was more or less the aim when microcontrollers were first invented.

What I've been thinking about is, is it bad design practice to miss-use a microcontroller if it is to the benifit of the design? You would think that the answer would be yes, but from what I can tell, people seem hung up on only using the microcontroller for exactly what it was intended.

The usual kind of applications chips like the 8052 are used in are things like remote-controls, experimental robots etc... generally, things that are small, but provide a relativley large amount of functionality. Obviously in these types of situation it is necessary to keep component count down.

Say you wanted to use an 8052 in some sort of multi-purpose co-processing system (which for those who care, is one of the things I specialise in), if you're working with some large computing system, it obviously isn't essential to minimize everything in sight, nor is it so important to worry about things like current consumption (not to the extent that you would have to in a battery-powered application anyway).

A complicated co-processing board such as the one's I'm working on right now, will in the end have quite a high component count so some say it would be better to use a microprocessor instead of a microcontroller, in fact to do so wouldn't increase the component count much further.

What people tend to forget is that using a microcontroller in what really is a microprocessor's application will give you plenty more benefits than disadvantages, don't forget all those on-chip peripherals you can still use (UARTs, Timers etc..), you wouldn't get those inside you're common-as-muck everyday microprocessor.

Lets look what a standard 8052 could give you:

* Built-in System Clock Osc.
* RAM (even if it's only the 128/256 register bytes)
* 2 Timer/Counters (at least 40 extra pins added to your design if you had to use an external timer chip)
* 1 UARTS (up to 40 extra pins if you were to use two decent UARTs like the 6402 and even that wouldn't be as good as the UART inside the 8052)
* At least 8 spare I/O pins (even more for a lot of applications)
* Power/Reset Managment (present on most modern day 8052s)
* High degree of functionality in a small amount of space
* EPROM/EEPROM/FLASH 8052s give internal program memory, and sometimes extra EEPROM space for data storage.
* Reasonable amount of address decoding inside chip (think of all the glue-chips you would need to control all the peripherals if they were external!

Right, I've listed plenty of advantages to using a microcontroller, only one of them mentions the space taken up by the chip. Should we only think that microcontrollers are used to make designs small and power efficent? Do we have to forget about all the other advantages a wonderful device such as the 8052 can bring?

I know this is a bit of an unusual one for this forum, I know we're only really suppose to use it for technical questions, but I though this might make a nice change. Besides, I would love to know what people really think about what a good use for a microcontroller is, and what a bad use is. Do we all have the same opinion?

If you've got this far, thanks for taking the time to read my endless babbling!! Bye for now,

Matt.