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

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
04/04/05 15:49
Read: times


 
#90976 - the value of verification
Responding to: ???'s previous message
Erik, I think, you work in C long enough to stop thinking in the low level way (no offence).
Nah, In addition to C I write assembler most weeks>

Its funny how your own statements come back, I said:
"A suggestion: after having built the table, use one of the coded routines to verify your table."

So, yes, I goofed, but I would have caught it by my practice before anything was released.

I take this opportunity, not to excuse my error, but to emphasize the value of using however slow, but functioning, coded routines to verify fast tables before releasing. A table is probably the easiest place to make a bug that will not be discovered by testing since no testing will get to all spots in the table.

Erik



List of 36 messages in thread
TopicAuthorDate
Reversing Algorithm            01/01/70 00:00      
   remembered this one            01/01/70 00:00      
      the fastest code in the west            01/01/70 00:00      
         Am I wrong            01/01/70 00:00      
            not as simple, but less table            01/01/70 00:00      
               lookup table            01/01/70 00:00      
                  nope            01/01/70 00:00      
                     swap, the nibble mirror            01/01/70 00:00      
                        nibble-wise?            01/01/70 00:00      
                     nope nope            01/01/70 00:00      
                        noipe, nope, nope            01/01/70 00:00      
                            noipe, nope, nope, nope            01/01/70 00:00      
                              Ok, Ok. Ok, OK            01/01/70 00:00      
                                 max.255 entries - or the tricks :-)            01/01/70 00:00      
                                    the value of verification            01/01/70 00:00      
                  Mirrored tracks            01/01/70 00:00      
                     Why ?            01/01/70 00:00      
                        Working fine            01/01/70 00:00      
                           why you post wrong code ?            01/01/70 00:00      
                              Yes. Code has bug - Peter            01/01/70 00:00      
                                 efficiency            01/01/70 00:00      
                                    Obvious Vs Smart code            01/01/70 00:00      
                                       sorry            01/01/70 00:00      
                                          No offence - only learning            01/01/70 00:00      
   One byte-reversing solution            01/01/70 00:00      
   Byte Bit Reversal            01/01/70 00:00      
      Re: Byte Bit Reversal            01/01/70 00:00      
   Bit Addressing solution!            01/01/70 00:00      
      Hardware?            01/01/70 00:00      
      2 byte of ram or 2 port?            01/01/70 00:00      
   Three ways:            01/01/70 00:00      
      Three (smart) ways            01/01/70 00:00      
      Thanks, Peter, but...            01/01/70 00:00      
         its faster            01/01/70 00:00      
   Topic has been discussed before            01/01/70 00:00      
   You could try a CPLD            01/01/70 00:00      

Back to Subject List