1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  2. Greetings Guest!!

    In order to combat SPAM on the forums, all users are required to have a minimum of 2 posts before they can submit links in any post or thread.

    Dismiss Notice

Some insight on how broken the random # generator is.

Discussion in 'UO Siege Perilous' started by Speedy Orkit, Sep 30, 2008.

  1. Speedy Orkit

    Speedy Orkit Grand Inquisitor
    Stratics Veteran

    Joined:
    May 16, 2008
    Messages:
    4,132
    Likes Received:
    9
    Did two tests last night, since I was going to make these items on test center anyway:

    Cross bows: 400 made
    100% chance to make
    55% chance for exceptional

    131 exceptional, 269 non excp

    Ecru rings: 200 attempted
    50% chance to make
    5% chance for exceptional

    67 made
    7 excp

    Honestly, I've noticed this stuff before, but now that I've actually tried this, it really bugs me.
     
  2. Petra Fyde

    Petra Fyde Peerless Chatterbox
    Stratics Veteran Alumni Stratics Legend

    Joined:
    Jan 5, 2001
    Messages:
    30,884
    Likes Received:
    5,165
    sent it in on a feedback form? If not, do. Jeremy actually reads those.
     
  3. T'Challa

    T'Challa Certifiable
    Stratics Veteran

    Joined:
    May 12, 2008
    Messages:
    1,969
    Likes Received:
    15
    You will never see "true" results of a RNG until you start doing batches of hundreds of thousands...

    Where is Trep when you need him? The guy is a friggin mathematical genius...he could explain it better than I ever could.
     
  4. Lorddog

    Lorddog Crazed Zealot
    Stratics Veteran Stratics Legend

    Joined:
    Oct 25, 2004
    Messages:
    3,115
    Likes Received:
    298
    Cross bows: 400 made
    100% chance to make (100%)
    55% chance for exceptional (49%)

    131 exceptional, 269 non excp

    Ecru rings: 200 attempted
    50% chance to make (34%)
    5% chance for exceptional (10%)

    67 made
    7 excp

    I put in () the percentages you actually had. it doesnt seem outragouly off but it is signigicantly off.
     
  5. Goron

    Goron Guest


    yeah, 400 tries, 200 tries; those are mere hiccups when it comes to probability.
    I just used Minitab to generate 9 sets of '200 excru ring attempts'.
    I got 7,4,4,1,5,6,12,9,11 percent exceptional(s), in other words, 14,8,8,2,10,12,24,18,and 22 exceptional rings in each set. averaging out it was 6.5% exceptional- I'd say that's pretty close to 5%... I bet if I were to run 100 sample runs, I'd be even closer. But, you will notice one of my sets had an 11% exceptional rate- that could be your ONE try... maybe the next 200 would have yielded 1%...
     
  6. Scuzzlebutt

    Scuzzlebutt Babbling Loonie
    Stratics Veteran

    Joined:
    Mar 9, 2007
    Messages:
    2,215
    Likes Received:
    4
    If you flip a coin 1000 times and it comes up heads 999 times, it still only has a 50% chance of being tales on the 1000th toss. You only change percentages when you add or remove other possible outcomes.Thats all im going to say about probabilities.
     
  7. Goron

    Goron Guest

    Pessimist... only 50% chance? I say it has a whopping 50% chance!
     
  8. GoodGuy

    GoodGuy Slightly Crazed
    Stratics Veteran

    Joined:
    May 30, 2008
    Messages:
    1,357
    Likes Received:
    1
    Right, speedy is assuming that the game is gonna make sure you get those percentages everytime. I mean flip a coin a thousand times, you may get 60% heads, does that mean a coin has a 60% chance of being heads? No its still 50% everytime you flip it.

    you need to test at least 10,000, thats around when the error rate for typical probability will be low enough to make a decently accurate judgement.