>Call me a skeptic, I'll be surprised if these chips are anything but
>moderately successful. They don't do anything that much better than the
>PIC, AVR or MSP430 family -- and those chip families all have a solid
>infrastructure complete with low-cost 3rd party tools available.

I'm generally in favour of SIC (Single Instruction Computer) architectures,
I designed one myself, together with several AMUs (Arithmetic Movement
Units).

But, about the MAXQ, they didn't make any mention to the MHz they think
they'll reach (but they claim to have made accurate simulations of power
consumption, thus they could have provided some MHz figure as well).

Moreover, using the PIC16 as comparation is unfair, to say the least.
The MAXQ is probably superior even to the PIC18, but making such an
unfair comparison abusing the PIC16 doesn't tell me many good things
about the "honesty" of the whole presentation, and thus I tend to become
skeptical also about the rest.

For what we know, the chips may have such a complex logic that, not being
pipelined, the max clock speed may very well be low, vanishing every
advantages in terms of cycles/instruction.

It's like in the times of Z80 vs 6502.. many 6502 supporters (I was
among them) said that the 6502 was superior because of the lower average
cycle count per instruction. But the Z80 was available at more MHz than
the 6502..
I didn't understand, at that time, that by using the same fabrication
process two designs may have different clock speeds. Evidently the Z80
gained back somewhere else (not really pipelined, but..) what it lost
in the cycle count.
Cycles per instruction say nothing, if one doesn't say also the max
clock speed. I can make a CPU that calculates a logarithm function in
just one clock cycle.. wow, but with a max clock speed of 100 KHz. So
that should not be "wow", but "bad design". Having as little logic as
possible between each clock cycle is what makes a good design (that's
how I like to say it in my mind).

You said it right.. that presentation has been written by some marketing
guy.. those are truly skilled when it comes to convince us not to buy
something.

Anyway, let's wait and see something more concrete.

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu