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

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
04/08/07 05:27
Read: times


 
#136803 - 805x, but modified
Responding to: ???'s previous message
I don't like a few features of the 805x, but don't generally complain. However, I don't like the relationship between the system clock and the serial channels, timers, etc. I also prefer that instructions take no more than one clock cycle to execute, irrespective of their content, an would expect that those instructions that can be executed concurrently or out of sequence be executed that way in order to save time.

I'm sticking to the same basic instruction set, particularly in object form, as the tools are pretty good and pretty avaiable. While it might take a year or two to whittle up and verify a new core that's done to my liking, along with the modified peripherals that I'd prefer, I have to find time to contemplate, design, and implement the features and modifications. Who knows ... I may not live long enough to see it realized.

It's still pretty preliminary so far. I've examined some of the enhancements, but they'll be no more difficult than simply implementing a single-clock core.

RE


List of 26 messages in thread
TopicAuthorDate
Freebie 8051 cores            01/01/70 00:00      
   They do work to a limited extent, but...            01/01/70 00:00      
      there's a range ...            01/01/70 00:00      
         What architecture?            01/01/70 00:00      
            805x, but modified            01/01/70 00:00      
               are you reinventing the wheel?            01/01/70 00:00      
                  Yes, but it's worth the trouble            01/01/70 00:00      
                     wrong place to read            01/01/70 00:00      
                        I only looked at their site once ...            01/01/70 00:00      
                     SiLabs pricing?            01/01/70 00:00      
                        That's what they had on their website            01/01/70 00:00      
   RTFM?            01/01/70 00:00      
      Mostly, it's absence of "TFM"            01/01/70 00:00      
   OP Update            01/01/70 00:00      
   I hate to assume, but            01/01/70 00:00      
      "free" ... I doubt it ...            01/01/70 00:00      
         OK a recollection from a seminar            01/01/70 00:00      
            Maybe from this convention?            01/01/70 00:00      
   is FPGA a suitable "substrate" for a processor?            01/01/70 00:00      
      it's not as easy as the marketeers say ...            01/01/70 00:00      
         FPGA Software            01/01/70 00:00      
      FPGAs work fine for processors for us            01/01/70 00:00      
      FPGA-based processors            01/01/70 00:00      
         Shouldn't it do better than 50 MHz?            01/01/70 00:00      
            50 MHz is pretty fast            01/01/70 00:00      
            My most recent design work            01/01/70 00:00      

Back to Subject List