[comp.arch] RISC Microcontrollers

rick@pcrat.UUCP (Rick Richardson) (03/22/88)

Having spent the better part of my career building imbedded systems
around all sorts of braindamaged microprocessors and listening to
hardware types bitch that they can't increase the clock for fear
of failing EMI, I'd like to know if anyone is working on a RISC
microcontroller.

Given the around 20,000 dhrystone performance of SPARC and MIPS at
16 Mhz clock, I feel confident that:

	1) 1/8 of this clock would beat the pants off of the uP's
	   we typically use today by a factor of two.
	2) I could convince the hardware types to use a 16 bit
	   bus (instead of 8) given that the performance is
	   there and the EMI isn't.  I doubt I could convince
	   them to move up to 32 bits, however.
	3) I could also (probably) convince HW that they'd
	   still be saving money even if we need twice as much
	   program ROM to accomodate the RISC philosophy.

Is anybody working on a RISC micrcontroller with about 2000
dhrystones @ highest frequency == 4 Mhz, on a 8/16 bit bus,
with a state-of-the-art C compiler and a cost not to exceed
$10 in 100K quantities?????  Oh, don't forget the ICE for it;
prefer that the ICE take COFF format load modules.  The ICE
should cost no more than $15K.  Give us 9. We'd pay up to 25K for the
cross C compiler binary license on a super-mini.  Don't forget
to make sure the chip is compatible with popular peripheral
chips (both MUX'ed and non-MUX'ed addressing!!!).

-- 
		Rick Richardson, President, PC Research, Inc.

(201) 542-3734 (voice, nights)   OR     (201) 834-1378 (voice, days)
uunet!pcrat!rick (UUCP)			rick%pcrat.uucp@uunet.uu.net (INTERNET)