sk8bcn
Joined: Apr 13, 2004
|
  Posted:
Jan 05, 2006 - 18:57 |
|
So finally I did understood it right! Biggie is a mathematician as me and I thought to myself:
"Their aren't discussing about the possibility that the RNG countebalance bad luck with a good stream when they didn't take time to make calculation about the probabilities such things occur, are they??? After reading Biggie it seems it's the case!!
ps: hyphotes means???? |
_________________ Join NL Raises from the Ashes |
|
Mezir
Joined: Aug 02, 2003
|
  Posted:
Jan 05, 2006 - 19:07 |
|
sk8bcn: Which is what I've been trying to point out all the time, that the good and bad streams occur with the same frequency as they would in true random, which I believe my test program has now shown to be true.
And I believe that by 'hyptes' BiggieB means the test that tests the hypothesis. |
_________________ Build a man a fire and he's warm for a day; set a man on fire and he's warm for the rest of his life. |
|
sk8bcn
Joined: Apr 13, 2004
|
  Posted:
Jan 05, 2006 - 22:58 |
|
then I am with you Mez' . Well not too much though, you know you have 2 major flaw DD3.5 and comming from Belgium |
_________________ Join NL Raises from the Ashes |
|
gallahbrains
Joined: Jan 06, 2006
|
  Posted:
Jan 06, 2006 - 00:36 |
|
Rynkky wrote: | Vicimus, you may want to recheck your proggie. 500% sounds quite high to me..
Quote: |
500000 psuedorandom numbers were just generated.
These numbers were organised into 10000 consecutive sets. Each sets contains 50 psuedorandom numbers.
A good stream averages above 3.9
A bad stream averages below 3.1
A good stream occurred 4.4% of the time.
A bad stream occurred 500% of the time. |
|
I pressed refresh about a thousand times and can't recreate that problem.
Also, as far as I worked it out a good team and bad stream should occur in real life about ~4.5% of the time and the program recreates that almost perfectly. |
|
|
Mezir
Joined: Aug 02, 2003
|
  Posted:
Jan 06, 2006 - 00:40 |
|
More on the same subject in this thread, containing more tests and their results. |
_________________ Build a man a fire and he's warm for a day; set a man on fire and he's warm for the rest of his life.
Last edited by Mezir on %b %06, %2006 - %13:%Jan; edited 2 times in total |
|
Rynkky
Joined: Aug 03, 2004
|
  Posted:
Jan 06, 2006 - 12:54 |
|
gallahbrains wrote: | Rynkky wrote: | Vicimus, you may want to recheck your proggie. 500% sounds quite high to me..
Quote: |
500000 psuedorandom numbers were just generated.
These numbers were organised into 10000 consecutive sets. Each sets contains 50 psuedorandom numbers.
A good stream averages above 3.9
A bad stream averages below 3.1
A good stream occurred 4.4% of the time.
A bad stream occurred 500% of the time. |
|
I pressed refresh about a thousand times and can't recreate that problem.
Also, as far as I worked it out a good team and bad stream should occur in real life about ~4.5% of the time and the program recreates that almost perfectly. |
Maybe he (she?) fixed it? Don't know .. I tried again and now I am getting correct-looking results. |
|
|
|
| |