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

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
03/07/07 12:38
Read: times


 		 #134456 - let's calculate a bit
Responding to: ???'s previous message
OK so add up 4 cycles for the interrupt and return... Or, let's be more generous and calculate 20 cycles per "unnecessary" interrupt.

The maximum baudrate a vanilla (+ alike) '51 can handle in 9-bit mode is one character per 176 instruction cycles. In anything but the most extreme protocols, there is at least a byte or two of payload after the address, one or two bytes' time delay to turn around, and a couple of bytes of response, then some more delay for turning around the bus. Let's assume 6 bytes' time altogether, that's around 1000 cycles.

So, the "unnecessary" interrupt took up 2% of your precious processing time... I don't think this is really worth even mentioning...

On the other hand, those 50 gates or so, costing $0.00000001 per chip, make up such a nice line in the Features list, "Enhanced UART with automatic address recognition"... :-)

JW


List of 59 messages in thread
TopicAuthorDate
is AT89C51 can support Auto Address Rec: feature ?            01/01/70 00:00      
   IF they aren't there, it doesn't do it!            01/01/70 00:00      
      why the automatic address recognition?            01/01/70 00:00      
         Every little helps!            01/01/70 00:00      
            let's calculate a bit            01/01/70 00:00      
               an often forgotten fact            01/01/70 00:00      
                  certainly            01/01/70 00:00      
                     let me try            01/01/70 00:00      
                        this sounds like....            01/01/70 00:00      
                           As I said, it is not easy to find an example thst            01/01/70 00:00      
                              if it\'s not easy to find an example...            01/01/70 00:00      
      WHO told you that you HAVE to use a derivative t            01/01/70 00:00      
         Well ofcourse i am not going to design 100000            01/01/70 00:00      
            I will help you, but not write your code            01/01/70 00:00      
               Well thanks Erik            01/01/70 00:00      
               well sir if i write a code then could you check it            01/01/70 00:00      
                  sure            01/01/70 00:00      
                     well sir its this ...            01/01/70 00:00      
                        you can't do this            01/01/70 00:00      
                           well instead of Rx            01/01/70 00:00      
                        On Comments            01/01/70 00:00      
                           So Sorry for my Mistake            01/01/70 00:00      
                              stop accusing yourself            01/01/70 00:00      
                                 Well sir its Honor for me :-)            01/01/70 00:00      
                           The Right one here goes            01/01/70 00:00      
                              I think you made the wrong change            01/01/70 00:00      
                              this will (almost) work if you only have one data            01/01/70 00:00      
                                 Well sir Eric i think you are Talking about it....            01/01/70 00:00      
                                    not yet            01/01/70 00:00      
                                    Ooooh, another '51 pitfall...            01/01/70 00:00      
                                       Hows That ?            01/01/70 00:00      
                                          still not good...            01/01/70 00:00      
                                    I just noticed            01/01/70 00:00      
                                       as I said            01/01/70 00:00      
                                          Ah i am not in Hurry....            01/01/70 00:00      
                                             still not good II.            01/01/70 00:00      
                                                I have did it... you can see it i have Edited it .            01/01/70 00:00      
                                             ok, look at line 4            01/01/70 00:00      
                                                Simple,like if Address not match then GO_ON            01/01/70 00:00      
                                                   WHAT is in 'b'?            01/01/70 00:00      
                                                      I have finally got it.....            01/01/70 00:00      
                                                         now add the PUSH/POP PSW...            01/01/70 00:00      
                                                            Well i think that Pushing PSW is not neccesary ..            01/01/70 00:00      
                                                               better this than be sorry....            01/01/70 00:00      
                                                               it is            01/01/70 00:00      
                                                                  Sir you are right that carry is effected...            01/01/70 00:00      
                                                                     avoid common pitfalls            01/01/70 00:00      
                                                                     famous last words            01/01/70 00:00      
                                                                        Well sir i will never forget the carry is changed.            01/01/70 00:00      
                                                                           RTFM            01/01/70 00:00      
                                                                        Well sir How to save the Carry ??            01/01/70 00:00      
                                                                           see http://www.8052.com/forum/read.phtml?id=134583            01/01/70 00:00      
                                                                              Well sir i will see this later...            01/01/70 00:00      
                                                                                 so, that is the reason for the hurry            01/01/70 00:00      
         No compulsion!            01/01/70 00:00      
   Oops its Erik not Eric (Proc: For Multiple Data)            01/01/70 00:00      
      stop hurrying            01/01/70 00:00      
   It just hit me - you can't            01/01/70 00:00      
   Well sir i have give up ... means no computer...            01/01/70 00:00      

Back to Subject List