MISC not for embedded?

["Berwyn Hoyt" <blhoyt@xxxxxxxxxx>]

I agree with all of this.  My company uses embedded processors 
extensively in monitoring applications.  We never need speed.  
Cost & on-chip peripherals (lack of other costs) are usually the 
biggest factors.  Chuck's P8 and P32 ideas look really good for 
price, but I doubt they'll have the on-chip peripherals.  (And I doubt 
they'll be for mass-sale for a LOOONG time.  Correct me if I'm 
wrong.)

> What embeddable processors need is:
> - Many on-chip peripherals
> - On board (small) RAM en (big) [[[E]E]P]ROM.
> - Low cost (so high production volume)
> - Good support from the manufacturer
> 
> Speed is often not essential.
> 
> Not really the playing field for the MISC it seems. The MISC needs
> to find a niche between simple embedded applications and real computers.
> Seems that it will have to compete with the ARM, i960, H32 (?) etc.
> 
> What clients in this field often want is that there is progress in the
> development process of the processor. When I buy an n Mips processor
> today, can I buy an 2n MHz processor next year to satisfy the growing
> requirements of my own clients and my software etc. (Or when the processor
> turns out to be too slow after my software development cycle of several
> months.)


Berwyn Hoyt: BLHoyt@xtra.co.nz
1 Betts Ave
Masterton
New Zealand

This tagline is randomly selected:
Thought for the day:
    Communist (n): one who has given up all hope
    of becoming a Capitalist.