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

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
08/05/05 16:30
Read: times


 
#98887 - Wow!
Responding to: ???'s previous message
Jan Waclawek said:
I have prepared a "tester", which compares a result from shifting an arbitrary input value with a reference routine, and, more importantly, counts the instruction cycles of the "tested" shift routine, finds their minimum and maximum and adds them up to a sum (which then divided by 31 gives the average duration).


Wow, Jan, nice effort! While not that terribly important, I thought I might mention that I'm finding that the test program comes up with the correct best, worst, and total cycle count--but it seems to be mistakenly identifying the values of the accumulator for the best and worst case. Not sure why, or maybe I just goofed something up over here.

Also, for what it's worth, I'm seriously considering building an assembler and mini-simulator for the website. Obviously this would allow the website to assemble submissions at the moment they are submitted and, like your test suite, verify that the routine comes up with the right answer and do a performance profile for each possible set of inputs. While this probably isn't the most productive use of my time, the idea has sort of wedged itself into my brain and I think I'm going to have to do it just because it'd be fun. :)

Regards,
Craig Steiner


List of 17 messages in thread
TopicAuthorDate
Code Challenge operational            01/01/70 00:00      
   Questions for current challenge ?            01/01/70 00:00      
      Instruction Cycles?            01/01/70 00:00      
         best case/worst case            01/01/70 00:00      
            how about 30 instructions            01/01/70 00:00      
               30 instructions?            01/01/70 00:00      
                  i did count inside loops            01/01/70 00:00      
            Changed last night            01/01/70 00:00      
      Tester/Counter            01/01/70 00:00      
         Wow!            01/01/70 00:00      
   The 3 better solutions            01/01/70 00:00      
   Definition of BEST            01/01/70 00:00      
      Definition of best            01/01/70 00:00      
         Good idea            01/01/70 00:00      
   pre tags?            01/01/70 00:00      
      Yes            01/01/70 00:00      
         A highlight please            01/01/70 00:00      

Back to Subject List