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

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
09/10/07 13:53
Read: times


 		 #144308 - uniformity is not the only requirement on a pRNG..
Responding to: ???'s previous message
For example 0,1,2,3,4,...,65535,65536,0,1... - this IS uniform but barely would be fit for a pRNG ... :-)

I _don't know_ if the corrected LFSR is a good pRNG or not (in fact, I don't even know if a LFSR is a good pRNG or not, even if it is widely used as such). "Good" here means, passed the Diehard tests or similar test set. I even don't know what does "passed Diehard" exactly mean - DIehard spits out a result which has to be interpreted - I am just relying on what others say on the topic.

JW




List of 25 messages in thread
TopicAuthorDate
HOWTO generate a uniformly-distributed random nu            01/01/70 00:00      
   how much code/data/idata/xdata space can you...            01/01/70 00:00      
      I can wait till evening            01/01/70 00:00      
      LFSR            01/01/70 00:00      
         OK but            01/01/70 00:00      
            there is no such thing            01/01/70 00:00      
               event triggered Timer reading            01/01/70 00:00      
                  no, well            01/01/70 00:00      
            LFSRs generally suffer from the "bad seed" problem            01/01/70 00:00      
               Maximal-Length LFSR?            01/01/70 00:00      
                  uniformity is not the only requirement on a pRNG..            01/01/70 00:00      
   most rng's            01/01/70 00:00      
      hummm            01/01/70 00:00      
         pseudorandom is NOT random            01/01/70 00:00      
            A Dilbert            01/01/70 00:00      
   Easy way to have it ???            01/01/70 00:00      
      block cipher, in chaining mode...            01/01/70 00:00      
      on uniformity of pseudo- or real-random            01/01/70 00:00      
         what's xtea prng RANGE ?            01/01/70 00:00      
            its not a dedicated pRNG algorithm....            01/01/70 00:00      
               random float ??            01/01/70 00:00      
                  tell us what is the application            01/01/70 00:00      
                     The application is...            01/01/70 00:00      
                        take any 32 bits...            01/01/70 00:00      
                           OK            01/01/70 00:00      

Back to Subject List