[rc5] Re: Supercomputers & Bovine, Q&A

root marcus at dfwmm.net
Mon Aug 4 20:38:15 EDT 1997


bennett_t1 at popmail.firn.edu wrote:
> 
> True, RISC chips are faster than CISC, and therefore better, In most
> respects.

Ah, this last little phrase saved you. Instead of flames, I post an
explanation of one of "those respects" were RISC can often fail.
RISC has a lot against it. The only thing for it is simplicity and
thus lower cost per performance for the chip itself.

However, RISC means more instructions fetched to do a certain task.
That means bigger programs that occupy more memory, more disk space,
more cache and ultimately more power consumption because the poor
chip has to drive all of those external address and data lines more
often.

When building embedded systems the total cost of production is often
the decisive factor in a design's success or failure. That often means
that you cannot afford to have a big 256x32 ROM array because it costs
too much for the chips, the board space, and the power supply.

General purpose computers like PC's or workstations are not under
nearly the cost pressure that small embedded systems are. The cost
of an extra $2 for motherboard space, $30 for cache or ROMs and
some more for DRAM will not kill a product. In a box that is supposed
to sell for just a couple of hundred bucks or less such costs can
be fatal.
----
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.



More information about the rc5 mailing list