Ironbound

Announcement

Collapse
No announcement yet.

The RNG is broken

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • The RNG is broken

    I just played an arena game using the cruelty crusader. Only crit 20% of the time and squeaked out a win. So I tried again thinking I had to do better. ZERO critical hits the entire game. WTF?!?

  • #2
    Woow that aint cool, i mean, im the king of shitty RNG, but you are a god dude.But stil ...we have to check it.

    Comment


    • #3
      Long series of unwanted numbers say absolutely nothing about the correct operation of an RNG. In fact their existence is one way to tell RNG output apart from simulated RNG output that humans produce when asked to write down or say a string of what they feel are "randomly chosen numbers".

      A lot of game players (and gamblers too) want something other than a true RNG, but they just don't know it. They think what they want is a "better" RNG or an RNG that "is not broken". But the real description of what they want is a so-called RNG where if it rolled really low for a while, it will be more likely to roll in the middle of the range or higher later on Or vice versa. But that would not actually be a random series of numbers! That is exactly what has been known as The Gambler's Fallacy for hundreds of years. Not understanding it clearly has led countless gamblers and gamers into frustration and loss.

      Comment


      • #4
        A pseudo-random RNG would "feel" better but lead to a broken game. Just calculate what kit does the most damage over 8, 9, and 10 turns, and boom, you're done.

        Comment


        • #5
          Agreed.

          Comment


          • #6
            I'm not expert in statistics, is this correct?
            With 20% chance to crit each time, you have a 4,4% to never crit in 14 rolls (I expected it to be rarer!).
            With 30% it would be 0,67%
            With 40% it would be 0,08%
            With 50% it would be 0,006%
            Last edited by phaolo; 10-15-2018, 06:24 AM.

            Comment


            • #7
              Originally posted by phaolo View Post
              I'm not expert in statistics, is this correct?
              With 20% chance to crit each time, you have a 4,4% to never crit in 14 rolls.
              With 30% it would be 0,34%
              With 50% it would be 0,006%
              Correct for 20% chance crit each time -- that should mean 4.4% likelihood of going 14 rolls in a row without a crit.
              For 30% crit chance you must have missed something, it works out to 0.68%.
              Your figures for 50% crit chance were also correct.

              Comment


              • #8
                Originally posted by Fnord View Post
                Correct for 20% chance crit each time -- that should mean 4.4% likelihood of going 14 rolls in a row without a crit.
                For 30% crit chance you must have missed something, it works out to 0.68%.
                Your figures for 50% crit chance were also correct.
                Thanks, even if you replied during my edits :P
                Last edited by phaolo; 10-15-2018, 06:33 AM.

                Comment


                • #9
                  When a game that had been out for only a month or two produces a result that is literally one in a million it is reasonable to ask if the RNG is working. And it is also reasonable to complain when you are the victim of said one in a million event.

                  Comment


                  • #10
                    Ah, but that is the "Retrospective" form of the Gambler's Fallacy. That's where, when something very unlikely occurs, we say "wow, it should have been a long time before that ever had any significant chance of coming up." Turns out that's just not true!

                    I feel ya, though -- I think I have had numerous battles in Ironbound where I didn't crit once, or maybe crit one single time. That's painful!

                    Comment


                    • #11
                      Originally posted by BAgate View Post
                      When a game that had been out for only a month or two produces a result that is literally one in a million it is reasonable to ask if the RNG is working. And it is also reasonable to complain when you are the victim of said one in a million event.
                      I agree that the RNG feels a bit wonky, but that's not a "one in a million event", it's ~4 games every 100! (surprisingly, I thought it was super rare too).
                      You should aim for 30% crit minimum.
                      Last edited by phaolo; 10-15-2018, 03:42 PM.

                      Comment


                      • #12
                        I have chatted with Gerula about this. However long ago, they did something like a million simulations with the game engine and analyzed the result, giving them confidence in the game's RNG. It's just as normal to see the unlikely results as it is the likely results. If this were a long chain of numbers, it's just a matter of what glimpse of that chain we get in our playing "window". I find that if I play a high attack build, like my Voodoo Needle/Beast Fang assassin, which usually gets ~30 attacks in a match, I see very typical numbers. Sometimes, I get better and sometimes worse, but usually the 50% or so that I'd expect for a crit rate. If I do something that might get 8 or 9 attacks, or less, all bets are off. Might be normal or abnormal...

                        I admit, that I have found it excruciating to get massacred by an opponent's outrageous RNG, but that's life with a fistful of dice.

                        Comment


                        • #13
                          Nothing in this game is normal lol

                          Comment


                          • #14
                            Pseudo RNG works well in games like DOTA 2, which are cybersport discipline. So, I gotta, there is nothing wrong to include that mechanic in ironbound. I wish I would play as much **** RNG matches as godlike RNG, but it goes normal or bad and if feels sad

                            Comment


                            • #15
                              Play enough and you see all results. Lots of normal, some bad, some great. And it's easy to get a distorted picture. Get pissed off and quit in round 5, then your crit may be low but might have evened out by turn 10. And two hands have different crit rates, and the second hand probably gets 1-2 attacks less. So figuring out "I should have gotten X crits" is not as straight-forward as it seems, especially once you start modifying those values with items.

                              I just look at it like a dice roll game. I played a lot of warhammer fantasy tabletop, and a bit of DnD, so fistfuls of uncooperative dice aren't uncommon. But the feeling of seeing all those 6's turn up when you've dumped 20 dice on the table is a lot of fun, too.

                              Comment

                              Working...
                              X