Page 2 of 4 FirstFirst 1234 LastLast
Results 16 to 30 of 50

Sharing some anvil info

This is a discussion on Sharing some anvil info within the Game Questions & Tips forums, part of the Knight Online (ko4life.com) category; Originally Posted by Codex some background: i have never understood people who try to burn shitloads of items before upgrading ...
Page: 2


  1. #16
    Senior Member KingHOBO's Avatar
    Join Date
    Oct 2009
    Posts
    118

    Default

    Quote Originally Posted by Codex View Post
    some background: i have never understood people who try to burn shitloads of items before upgrading something important, but as so many ppl do it, i thought why not check it out, if its true, maybe it can be used to make some extra money off the anvil.

    in theory, any toss of a coin is not affected by the ones who preceeded it. for instance, if i toss a coin 100 times, and the first 99 times i get heads, the 100th time i toss the coin still has 50/50 chance of heads/tails. now the anvil is no ordinary coin toss, but the same principles of statistics still apply (including the above law of indepedency between two random occurences).

    to test this, i have made alot of high class upgrades which i kept track of to figure out if there is any dependency between one upgrade and the next. here are my findings:

    success rate of upgrading high class items from +2 to +3 with high class upgrade scroll is: 60% (60 out of 100 items went from +2 to +3 and 40 burned).

    success rate after a failed upgrade of high class items from +2 to +3 with high class upgrade scroll is: 65% - 26 out of 40 items went from +2 to +3.


    success rate after a successful upgrade of high class items from +2 to +3 with high class upgrade scroll is: 60% (stays the same) - 36 out of 60 went from +2 to +3.

    i made about 100 upgrades overall so to make this data as accurate as possible.
    the upgrades were made on cw2 when no one else used the anvil.

    this data goes to show that there is little dependency between one upgrade and the next (although there was a slightly increased chance of a successful upgrade after a failed one).

    i would love to see more ppl post here success rates of other upgrades (like +4 to +5 with bus, or +6 to +7 with trina), so everyone could benefit and manage risk taking with the anvil wisely. please post success rates that you tested and not what you think the success rates are.

    also if someone still thinks that you should always burn shitloads of items before upgrading something important, i would love to hear why you think that, and if it ever worked for you.
    I think your on to the general idea but I'm no statician.

    Take a dice with 6 sides, it's a 50% chance of rolling an even number on any give try. Every roll will give you a 50% chance but, what's the chance of rolling 3 odds in a row or 4,5, or higher? So what would be the chance of rolling 100 dice and getting all odds? Much lower than 50%.

    Another thing is the anvil does go in cold and hot streaks.

    Good Luck testing out your upgrading.

  2. #17
    Haters gonna hate Senior Member MushtMan's Avatar
    Join Date
    Jan 2010
    Location
    Can't live if you're dead, just sayin...
    Posts
    1,853

    Default

    imo like someone already said it's just a random number generator, maybe even based on the clock (it's KO you know...), if you fall in the number range you succeed, if you don't you fail.

  3. #18
    Unlimited
    Guest

    Default

    Although your numbers maybe correct, you also have to realize that there is a hidden feature in most RNG codes called a "Gold Sink" -> Gold sink - Wikipedia, the free encyclopedia

    Every MMO has a gold sink feature in which certain features of the game, such as the anvil in KO, is used to make you "throw" your money away, but it is a necessity to do so to further better your game play. The way the anvil is a gold sink is that it helps lower inflation mainly in part that, say you buy a Shard for 10 million, plus the BUS to upgrade it to +7. On the +6 to +7 upgrade, it fails, thus you lost about 30~ million gold.

    My point of this is that although these numbers maybe accurate with a certain percentage of error, there will always be the gold sink code that will say "Too many +8s at the moment - Fail upgrade." So yes, maybe at times you upgrade +7-+8 with 60% success rate on High Class items and end up getting 6/10 +8 Shards but it's almost stupid to think that the next time you upgrade another 10 +7 Shards that you'll get 6 to +8, it's just a statistical improbability.

  4. #19
    HumanStupidityIsInfinite Senior Member xtommyleongx's Avatar
    Join Date
    Jun 2007
    Posts
    284

    Default

    if the anvil could be figured out... say hello to world peace

  5. #20
    Senior Member
    Join Date
    Mar 2006
    Posts
    144

    Default

    unlimited, you do have a point there,
    the anvil does act as an item sink, the question is if it corresponds to the amount of items in the server and adjusts its success rates accordingly, or are the success rates fixed, and the item sinking is constant regardless of how many items there are in the server.
    lets say if there are too many shards+8 on the server and too little raptors+8 on the server, is the upgrade chance of shards from 7 to 8 different from those of raptors from 7 to 8?

  6. #21
    Haveyouseenmybeartibbers? Senior Member JDizzle's Avatar
    Join Date
    Nov 2006
    Location
    michigander
    Posts
    2,083

    Default

    well i made a +9 today anvil, so suck it!

  7. #22
    Devil's Advocate Senior Member Zealous's Avatar
    Join Date
    Dec 2007
    Location
    New Zealand
    Posts
    1,223

    Default

    Theres some super simple reasoning behind peoples superstitions, and an even more simple explaination as to why it doesn't work:

    Flip a coin once, 0.5 (1/2) chance of heads

    Flip a coin twice, 05 (1/2) chance of heads

    Flip a coin thrice, 0.5 (1/2) chance of heads

    Following determinism, we are told each individual flip has the same 0.5 chance, right? Getting two heads in a row would then be 0.5 x 0.5 = 0.25 (1/4); three ina row would be 0.25 x 0.5 = 0.125 (1/8); four would be 0.125 x 0.5 = 0.0625 (1/16)

    So theoritically if you burn three items, the chance of the next item being a burn would be 6.25% (or 93.5% chance of success).

    Alright heres the part where it doesn't work, are you ready?

    Cause probability theory is only theory, and the anvil's a tempermental bitch.

    No matter how much smaller the theoretical probability of a burn may become after preceeding burns, in any situation with more than 1 outcome, the possibility for a burn occuring is still 1, rather 100%.

  8. #23
    Senior Member Jammin's Avatar
    Join Date
    Feb 2009
    Posts
    889

    Default

    To the original poster: a topic was made some time ago by someone who made a pretty tought to understand mathematical statement where he explained that burning things did increase the odds of your upgrading succeding.

    I dont remember exactly how it was but it made a lot of sense. Perhaps someone could find that topic. Cant be fucked to search since i dont even know where to start with :P but im 100% sure said topic exists.

    EDIT: it was something in the line of what Zealous said. But yes, much like he said, anvil is evil.

  9. #24
    Banned Senior Member
    Join Date
    Mar 2006
    Location
    Israel
    Posts
    2,943

    Default

    Well , if we assume the results are independent of eachother , then zealous is completely wrong.
    If we assume they are dependent , then he's right.
    It all comes down to whether they are independent ( previous upgrade does not influence the current one ) or dependent ( previous upgrade does influence the current one )
    There is no correct math behind this anvil because we don't know the most basic stuff about it
    With a much larger experiement similar to codex' we'd be able to ponder upon the results better , and maybe conclude whether it's dependent or independent.

    P.S : Zealous is wrong eitherway because he fails to distinguish the chance to do something x times in a row from the chance to do something one time after a known sequence of prior results.

  10. #25
    Almighty Moderator LambOfGod's Avatar
    Join Date
    Jul 2009
    Location
    Peru.
    Posts
    1,507

    Default

    Quote Originally Posted by Jammin View Post
    To the original poster: a topic was made some time ago by someone who made a pretty tought to understand mathematical statement where he explained that burning things did increase the odds of your upgrading succeding.

    I dont remember exactly how it was but it made a lot of sense. Perhaps someone could find that topic. Cant be fucked to search since i dont even know where to start with :P but im 100% sure said topic exists.

    EDIT: it was something in the line of what Zealous said. But yes, much like he said, anvil is evil.
    man you're getting lazy lately >.>.

    Go +9 How Many %?


    ps: checa mi pm en msn ptm.

  11. #26
    Banned Senior Member
    Join Date
    Mar 2006
    Location
    Israel
    Posts
    2,943

    Default

    Quote Originally Posted by PendexO View Post
    man you're getting lazy lately >.>.

    Go +9 How Many %?


    ps: checa mi pm en msn ptm.
    The guy in that topic that took probability course in university obviously knows nothing about the true meaning of what he learnt.

  12. #27
    Senior Member Jammin's Avatar
    Join Date
    Feb 2009
    Posts
    889

    Default

    Quote Originally Posted by PendexO View Post
    man you're getting lazy lately >.>.

    Go +9 How Many %?


    ps: checa mi pm en msn ptm.
    What you talking about ? Ive allways been lazy

  13. #28
    Tits or destiny? Senior Member Dark3lf's Avatar
    Join Date
    Jul 2006
    Location
    Fınland oke?
    Posts
    2,976

    Default

    Let's just put it this way. No matter how many times you flip a coin there's always a 50% chance for it to be heads and 50& chance for it to be tails. No matter how many items you upgrade you can never know the outcome.

  14. #29
    Senior Member
    Join Date
    Jun 2006
    Posts
    253

    Default

    "So theoritically if you burn three items, the chance of the next item being a burn would be 6.25% (or 93.5% chance of success)"

    false

  15. #30
    Senior Member
    Join Date
    Apr 2008
    Posts
    119

    Default

    Quote Originally Posted by Codex View Post
    unlimited, you do have a point there,
    the anvil does act as an item sink, the question is if it corresponds to the amount of items in the server and adjusts its success rates accordingly, or are the success rates fixed, and the item sinking is constant regardless of how many items there are in the server.
    lets say if there are too many shards+8 on the server and too little raptors+8 on the server, is the upgrade chance of shards from 7 to 8 different from those of raptors from 7 to 8?
    knowing Mgame the way i know, after looking in several source codes of private servers, im almost sure that even today, the odds of the upgrades are not connected to the ammount of +8s, since when you upgrade it doesn't search through the database of the server counting for the amnount of items that already exist. its all random.

Page 2 of 4 FirstFirst 1234 LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •