Crafting math finaly solved!!!

Discussion in 'General Discussion' started by flytripper, Oct 1, 2019.

Thread Status:
Not open for further replies.
  1. flytripper

    flytripper Avatar

    Messages:
    16
    Likes Received:
    20
    Trophy Points:
    3
    OK here's the crafting facts , Prove me wrong and ill give you 50 artifacts!!!

    crafting tree for blacksmith, carpentry, tailor is basic skill teir 1 that determines the lvl of recipe you can learn, ok thats correct kinda (explain later)
    Next is MASTERWORK skill that determines success chance
    Next is MASTERWORK (gear type) effect lvl

    SO THE FACT IS ... nothing in the crafting tree works at all except the masterwork effect! at masterwork effect 100 you get new options, 120 the options are higher stats, and 140 ect,,ect.,,

    I have crafted as much or more then anyone and i keep notes and observe,, I level ALCHEMY Enchant success chance, the craft table says on Enchant chance 95%, 89%, 50% 19% , 10% .........and those percentages are not even used in success chance. We are led to beleive those percentage numbers mean something but they do not!!
    Its probly just a 50/50 chance in reality. 10 bows masterworked, 4 of them succeeded on 15% chance, half failed on 95% chance..

    SOTA MATH IS EASY crafting is a 50/50 chance for success, and every 20 lvls masterwork (effect) opens another teir of masterworks/enchantments...... MINOR to MAJOR

    So the 95% chance and all the other chances are bullshit, there 50% chance or something no matter what ur lvl is
     
    Grimbone, Jaesun, Rentier and 3 others like this.
  2. GMDavros

    GMDavros Avatar

    Messages:
    290
    Likes Received:
    387
    Trophy Points:
    28
    *gets bucket of popcorn and settles in*

    Fenris
     
    Jaesun, Girlsname, Pifester and 2 others like this.
  3. Rentier

    Rentier Avatar

    Messages:
    915
    Likes Received:
    1,303
    Trophy Points:
    105
    Location:
    Dragomir Mori
    hmmm...don't have so high a skill lvl and not so much materials - soo
    is there another bucket popcorn and a free seat? I am in then
     
  4. Lazlo

    Lazlo Avatar

    Messages:
    1,362
    Likes Received:
    2,958
    Trophy Points:
    113
    nah
     
  5. Elrond

    Elrond Avatar

    Messages:
    1,259
    Likes Received:
    3,970
    Trophy Points:
    125
    Gender:
    Male
    Location:
    Crafters Town
    Trying to squeeze in but all seats seem to be taken already :)

    Fly keep at it ... something good gotta come out....eventually.
     
  6. Lained

    Lained Avatar

    Messages:
    2,346
    Likes Received:
    4,028
    Trophy Points:
    165
    Location:
    Yeovil, England
    …sample size… …statistics… …something… …something…
     
  7. GMDavros

    GMDavros Avatar

    Messages:
    290
    Likes Received:
    387
    Trophy Points:
    28
    I'll certainly share my popcorn. ;-)
     
    Rentier likes this.
  8. Pifester

    Pifester Bug Hunter

    Messages:
    131
    Likes Received:
    270
    Trophy Points:
    18
    Gender:
    Male
    Location:
    Middle of No-Where Alaska
    sits back to watch
    my numbers are a little different from yours, 95%, 92%, 53%, 17%, 9%. And this is at 140 enchantment, +5 table, +5 Blessing, and a +9 ring. How are you getting -3% chance on 2nd/3rd effect and a higher chance on 4th/5th effect?
     
    Rentier likes this.
  9. Joe Zhudarak

    Joe Zhudarak Avatar

    Messages:
    347
    Likes Received:
    642
    Trophy Points:
    55
    Gender:
    Male
    In my opinion
    95% in real about 85%.
    86% - 75%
    50% - 40%
    15% - 15%
    9% - 0%


    For carapacian gear i have special roll chances

    95% in real about 50%.
    86% - 20%
    50% - 10%
    15% - 0%
    9% - 0%

    These numbers are based in my bad luck!
     
    Last edited: Oct 1, 2019
    Jaesun, K1000, Rentier and 3 others like this.
  10. Duke Gréagóir

    Duke Gréagóir Legend of the Hearth

    Messages:
    5,263
    Likes Received:
    11,109
    Trophy Points:
    165
    Location:
    Dara Brae
    Thank you for all of the hard work. :) I grabbed a seat too.
     
    Rentier likes this.
  11. kaeshiva

    kaeshiva Avatar

    Messages:
    2,863
    Likes Received:
    11,107
    Trophy Points:
    153
    Gender:
    Female
    This made me laugh - because its so true =/ lol
     
  12. Deathadder

    Deathadder Avatar

    Messages:
    110
    Likes Received:
    314
    Trophy Points:
    30
    Gender:
    Male
    I'll just sit in the isle on the floor and eat my popcorn...
     
    Rentier and FrostII like this.
  13. Pawz

    Pawz Avatar

    Messages:
    881
    Likes Received:
    1,409
    Trophy Points:
    93
     
    Lace, Deathadder, Pifester and 2 others like this.
  14. Chrystoph Reis

    Chrystoph Reis Avatar

    Messages:
    940
    Likes Received:
    1,872
    Trophy Points:
    105
    Gender:
    Male
    Seems about right *steals some popcorn and sits on floor*. Eww sticky!
     
    Deathadder, Rentier and Cordelayne like this.
  15. Gia2

    Gia2 Bug Hunter

    Messages:
    322
    Likes Received:
    625
    Trophy Points:
    55
    Gender:
    Male
    I bet 1000 cotos the distribution isnt normalized at all by any factor. Which means there are and there will ever be people who aint get the event even after thousands (litterally) of cycles
     
  16. Daxxe Diggler

    Daxxe Diggler Avatar

    Messages:
    2,704
    Likes Received:
    5,738
    Trophy Points:
    153
    Gender:
    Male
    Location:
    Virtue Oasis - Hidden Vale
    Percent chance (regardless of the %) .... the underlying problem with crafting!
     
    FrostII likes this.
  17. Bryce Pallaton

    Bryce Pallaton Avatar

    Messages:
    72
    Likes Received:
    61
    Trophy Points:
    8
    I am pretty sure the %'s matter. However, it doesn't work like people expect, and the % shown isn't what you'll get. I think the game rolls twice for every check, and only if both rolls pass do you get the result. Therefore, square every % to get the actual odds.

    Why does this happen? I don't know for sure, but I suspect it's like this:
    You craft. Your machine rolls for success. If it succeeds, your machine sends the result to the server (but if it fails, it just fails and only sends the destroyed item updates to the server). The server gets that request and says, "We can't trust the client, so I'll roll for this", and then sends back success or fail.

    This would match the odds I see in practice.
     
  18. Fenrus MacRath

    Fenrus MacRath Avatar

    Messages:
    314
    Likes Received:
    751
    Trophy Points:
    55
    Gender:
    Male
    Location:
    Out of the Blue and into the Black.
    YOU sir, may be a genius.
     
  19. GMDavros

    GMDavros Avatar

    Messages:
    290
    Likes Received:
    387
    Trophy Points:
    28
    Really poor design if that's what it is doing.
     
    Witcheypoo likes this.
  20. Bryce Pallaton

    Bryce Pallaton Avatar

    Messages:
    72
    Likes Received:
    61
    Trophy Points:
    8
    It's more common than you might think. It happens when one person is working on the client and the other is working on the server. Or one guy is doing both, but he's overworked and can't remember what he did yesterday and there is a deadline that's like NOW and no time to look at other code. So the work is done twice but when you run it, it seems to work and so it ships.

    Then some pUnk says, "I don't think the odds are right", and maybe a ticket gets written or a dev does a quick pass through the logs and says, "We have all the statistics, our distribution is uniform and frequency is indistinguishable from noise; RNG works." So the pUnk sighs, and says, "Yes, but what statistics are you capturing, and how do you capture them?" At which point the dev normally hangs up the phone.

    But it's an important question, because if you are that dev, and you look at the statistics, they will look right. Guaranteed unless you wrote your own generator - which you should not do, unless your specific thing is to write random number generators. Give me a pool of Diffie-Hellman Elliptic-Curve generated numbers which are then divvied out to players on a first-come, first-served basis and the timing guarantees a result that is, in fact, truly random.

    But all the statistics in the world won't tell you if you're rolling twice as much as necessary. Someone needs to look at the code and see what it is actually doing.
     
    Bedawyn likes this.
Thread Status:
Not open for further replies.