RNG

Discussion in 'Crafting & Gathering' started by Aldo, Jul 7, 2019.

Thread Status:
Not open for further replies.
  1. Turk Key

    Turk Key Avatar

    Messages:
    2,561
    Likes Received:
    4,012
    Trophy Points:
    153
    Gender:
    Male
    Coming from a science and economic background, I have been taking most arguments against RNG with a grain of salt. I have no problem with the use of a random number generator in game design. There will always be those who get consecutive "losses" on the roll who claim foul, blaming the system not working when really it is. Up until now I have just dismissed such arguments out of hand. Recently however I have come to the realization that even if the RNG generator is working properly, the way it is used can lead to non-random outcomes.

    I am not good at communicating my ideas effectively so bear with me....My example in fact has nothing to do with RNG. Remember awhile back when in the interests of addressing lag issues, Port decided that the health buffs were causing lag because they were updating one's health every tic? They decided to adjust the mechanic to only deliver the buff once every 5 tics. Over time the buff was the same, but the delivery was broken into fewer calculations. The argument was that because total health buff was the same over time that the solution essentially gave no change to game play. It turned out that was definitely not the case because when a player is close to death, waiting 5 seconds (tic) for the next accumulated buff was just too long and the player died more often than before the change. Game play was definitely changed, but the argument that the buff was not nerfed was never effectively challenged. We operate to this day with the change that "makes no difference".

    Now getting to the RNG. I was fishing last week and caught two sailfish in consecutive casts. Yes, this is possible within an RNG application, but extremely improbable. I got to thinking. Perhaps our RNG app is running perfectly, but the way we apply it changes the outcome. For example (and I am not saying that this is a fact), if they are using a similar mechanic to the health issue above to reduce computations per tic when an RNG is generated, one could catch two sailfish in a row because the RNG was not used on both casts. I have become a skeptic now and believe that defending the RNG is a bit more complicated than proving that random numbers are generated on every roll. I don't doubt they are, but the application of that data and the timing of the "generation" have now made me dubious as to the outcome.
     
    Ataniiq likes this.
  2. Echondas

    Echondas Bug Hunter Bug Moderator

    Messages:
    3,783
    Likes Received:
    4,001
    Trophy Points:
    165
    Gender:
    Male
    Location:
    NY
    This frustrated me also and still does sometimes - perhaps this can be be re-examined or improved upon now that we are on a newer version of Unity - maybe @Chris can comment
     
    Aldo and Ataniiq like this.
  3. Turk Key

    Turk Key Avatar

    Messages:
    2,561
    Likes Received:
    4,012
    Trophy Points:
    153
    Gender:
    Male
    So over the last 30 days there were 45041 fish caught. Of that, 22 were sailfish. The odds of catching a sailfish on a single cast are 1 in 2047. This means the odds of catching two sailfish in a row are 1 in 4,190,209. The chances for me to catch two sailfish in a row are thus essentially 0. I have to assume that we do not have all the information to explain such absurd real world results. I feel the RNG is working properly, but that other factors are affecting the results of the instance as per the OP.
     
  4. Synergy Blaize

    Synergy Blaize Avatar

    Messages:
    603
    Likes Received:
    1,532
    Trophy Points:
    105
    Gender:
    Female
    Location:
    Australia- The Land Down Under
    Think yourself extremely lucky that you caught TWO Sailfish on consecutive casts!
    I go out fishing for an hour or so, most days when I am on & to ME...
    The amount of Crafting XP I have put into my skills (I have GM'd all Fishing skills & some are 120+) , "seem" worthless in catching "rare" fish!!!

    I went out with a friend who hasn't GM'd any fishing skills & they caught a "rare" Trophy fish
    & the Best I caught, were bait... Sardines & Herrings.
    (This is not including the Common Trophy fish I caught, which are essentially useless)

    Fishing NEEDS to be looked at...
    SKILLS need to be be more relevant in catching RARE Trophy Fish, rather than RNG.

    If Fishing is all simply about RNG... ???
    Then I have wasted a whole lot of My Time & crafting XP in increasing my Fishing skills!!!!
     
    Ataniiq, Vaiden Luro, Aldo and 2 others like this.
  5. Anpu

    Anpu Avatar

    Messages:
    7,944
    Likes Received:
    9,015
    Trophy Points:
    153
    Location:
    Hemut
    I do have GM Fishing, but I only have Lava Fishing at 40. I just don’t see any reason at all to ever level it any further.....
     
    Aldo and FrostII like this.
  6. Turk Key

    Turk Key Avatar

    Messages:
    2,561
    Likes Received:
    4,012
    Trophy Points:
    153
    Gender:
    Male
    This makes sense. I think the first roll is to see if your cast caught something based on the % shown when you cast. The second roll is to determine what you caught. So not raising skill further only affects how far you can cast. I doubt it makes a whit of difference as to what you catch. At the minimum, if the RNG roll for IF you caught something is positive, the second roll (my opinion) is to determine what it is. It is this second roll I suspect is being "optimized" in some way that affects the outcome. I have never understood how casting further could make any difference as to what you catch as in fact it doesn't affect the shown percentages. If you are at GM and in a fresh fishing area your percentage success is 71% whether you cast close to shore or at the max distance.
     
    Jaesun likes this.
  7. Gravidy

    Gravidy Avatar

    Messages:
    257
    Likes Received:
    613
    Trophy Points:
    28
    Gender:
    Male
    Location:
    San Francisco, CA
    This is the problem when designing-by-spreadsheet. The numbers appear to be *correct*, but players don't play on spreadsheets, they play in the game. If the game designers used game-play to influence their design decisions instead of their spreadsheets, we'd likely have a much different (and better) game.

    In your particular case above, the spreadsheet sums up the healing from all the "tics." However, in actual game-play when you're dead, any remaining tics no longer count, but the spreadsheet fails to take that into account.
     
    Aldo and FrostII like this.
  8. FrostII

    FrostII Bug Hunter

    Messages:
    5,884
    Likes Received:
    11,033
    Trophy Points:
    153
    Gender:
    Male
    Location:
    Pacific Northwest
    Not so.
    Your %success depends on the water "type" for any given skill level.
     
    Aldo likes this.
  9. Turk Key

    Turk Key Avatar

    Messages:
    2,561
    Likes Received:
    4,012
    Trophy Points:
    153
    Gender:
    Male
    What I meant was that if you walk up to the shore, in this case salt water, and cast, you will get the max (71%) whether you cast close to shore or at maximum distance. The % may vary as to water type.
     
    FrostII likes this.
  10. FrostII

    FrostII Bug Hunter

    Messages:
    5,884
    Likes Received:
    11,033
    Trophy Points:
    153
    Gender:
    Male
    Location:
    Pacific Northwest
    Correct, the %max is dependent on both water type and skill level.
     
    Aldo likes this.
  11. Arkah EMPstrike

    Arkah EMPstrike Avatar

    Messages:
    4,542
    Likes Received:
    8,100
    Trophy Points:
    153
    I agree with you. A long time ago @Violation Clauth caught on to this and found that he seemed to be able to break bad luck "streaks" by waiting them out(once he had a single failure, he would wait a minuite or so before trying again). You will see many recurring instances of people failing several times in a row with 99% success chance on faster tasks like Harvesting and Refining. The outcomes are noticeably "streaky" more often than you would expect to see in this game.
     
    Feldon Grimshaw and Aldo like this.
  12. Arlin

    Arlin Avatar

    Messages:
    379
    Likes Received:
    603
    Trophy Points:
    43
    I used to maintain a spreadsheet of masterwork results, and I can pretty much guarantee that over time the RNG is correct, but I'm also pretty sure there's some kind of issue with "streaks" also.
     
  13. Turk Key

    Turk Key Avatar

    Messages:
    2,561
    Likes Received:
    4,012
    Trophy Points:
    153
    Gender:
    Male
    Yep. I don't think the RNG is rolled on every craft attempt, fishing attempt, combat swing and etc. I am collecting data on fishing catch percent based on what shows on the screen. It will take a long time though to collect enough data to have confidence in the results.
     
    Jaesun likes this.
Thread Status:
Not open for further replies.