Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
Richard Erlacher
04/02/15 19:18
Modified:
  04/02/15 19:19

Read: 829 times
Denver, Co
USA


 
#190458 - Maybe I don't understand their claim ...
Responding to: Erik Malund's previous message
It would seem to me, based on what they say, that it just has to run at 12 MHz in order to achieve that score. There are some advantages to operating at only 12 MHz, including, perhaps, I/O, which might be pretty hard to use with external hardware if its physical cycle >20x as fast as the nominally less than 1 MHz cycle rate at which the original was intended to run.

There are numerous opportunities through which the cycle rate can remain fairly low while the instruction execution rate is sped up due to out-of-order execution, simultaneous execution, etc, but I don't see that large an increase in performance.

I do wonder what the real facts are.

RE

List of 11 messages in thread
TopicAuthorDate
"New 8051 IP core is 29X faster than original"      Andy Neil      04/01/15 01:18      
   if this is not gobbelygook, then      Erik Malund      04/01/15 07:01      
      That's benchmark scores for you      Andy Neil      04/01/15 10:35      
         'creativity'      Jim Granville      04/07/15 16:35      
            26.85x      Jan Waclawek      04/13/15 12:39      
      Maybe I don't understand their claim ...       Richard Erlacher      04/02/15 19:18      
   How about the power?      Oliver Sedlacek      04/17/15 04:44      
   From DCD's DQ8051 product page..      Valentin Angelovski      04/19/15 06:51      
      VAX MIPS      Andy Neil      05/14/15 13:18      
         MIPS      Michael Karas      05/18/15 14:18      
            order of magnitude      Andy Neil      05/20/15 09:31      

Back to Subject List