Page 3 of 5 FirstFirst 12345 LastLast
Results 41 to 60 of 88

  Click here to go to the first Dev post in this thread.   Thread: What do you think?

  1. #41
    Senior Member Kaziscate's Avatar
    Join Date
    Jul 2014
    Posts
    1,089
    Thanks Thanks Given 
    181
    Thanks Thanks Received 
    505
    Thanked in
    222 Posts

    Default

    Quote Originally Posted by Fyrce View Post
    This is obviously a lie. What moron has done 500k runs? Thats right, therefore lie.
    It takes 500k runs for a 100% chance.

  2. #42
    Senior Member Kaziscate's Avatar
    Join Date
    Jul 2014
    Posts
    1,089
    Thanks Thanks Given 
    181
    Thanks Thanks Received 
    505
    Thanked in
    222 Posts

    Default

    Correct me if I am wrong,but with the drop rate for the helm and belt being 0.0002% from elite Cryostar, it would be 0.00024% (with egg shell ring only),0.00026% (with lepre only),0.0003%(with 25% luck elixir only or have egg ring and lepre equipped),0.00032%(with 30% luck elixir only),0.00034% (25% luck elixir+egg ring[35% in total]), 0.00036% (25% luck elixir+lepre or have 30% luck elixir+egg ring[40% in total]),0.00038% (30% Luck elixir+lepre [45% in total]),0.0004% (25% luck+lepre+egg ring[50% luck]), and 0.00042% (30% luck+lepre+egg ring[55% in total]) Also, i am basing this off the new reroll system where 1 reroll counts as 2 runs


    Sent from my iPhone using Tapatalk

  3. #43
    Member
    Join Date
    Apr 2014
    Location
    Air Force
    Posts
    53
    Thanks Thanks Given 
    19
    Thanks Thanks Received 
    3
    Thanked in
    2 Posts

    Default

    I ate cake around 3 times and twice pancakes while i was reading, i think i never read so much O.o

  4. #44
    Member
    Join Date
    Apr 2014
    Location
    Air Force
    Posts
    53
    Thanks Thanks Given 
    19
    Thanks Thanks Received 
    3
    Thanked in
    2 Posts

    Default

    I ate 3 times cake and twice pancakes while i read, i think i never read this much O.o

  5. #45
    Senior Member
    Join Date
    Aug 2013
    Posts
    2,276
    Thanks Thanks Given 
    650
    Thanks Thanks Received 
    582
    Thanked in
    328 Posts

    Default

    Quote Originally Posted by Kaziscate View Post
    Correct me if I am wrong,but with the drop rate for the helm and belt being 0.0002% from elite Cryostar, it would be 0.00024% (with egg shell ring only),0.00026% (with lepre only),0.0003%(with 25% luck elixir only or have egg ring and lepre equipped),0.00032%(with 30% luck elixir only),0.00034% (25% luck elixir+egg ring[35% in total]), 0.00036% (25% luck elixir+lepre or have 30% luck elixir+egg ring[40% in total]),0.00038% (30% Luck elixir+lepre [45% in total]),0.0004% (25% luck+lepre+egg ring[50% luck]), and 0.00042% (30% luck+lepre+egg ring[55% in total]) Also, i am basing this off the new reroll system where 1 reroll counts as 2 runs


    Sent from my iPhone using Tapatalk
    it's not that high, the end percentage of getting yellow loot would be:

    [percentage of directly looting yellow] + [percentage of directly looting legendary] * [total loot reroll chance] * [percentage of directly looting yellow]

    if we insert numbers into it and let's say percentage of directly looting legendary is as high as 50%, and we use maximum reroll buffs we can get, it will be

    0.0002 + 0.5 * 0.55 * 0.0002 = it's 0.000255, not 0.0004ish :P
    always milking, no service
    AL IGN: Extrapayah/Extraparah/Extrajelek
    Support Shared Pet Stable Please!

  6. The Following User Says Thank You to extrapayah For This Useful Post:


  7. #46
    Senior Member Fredystern's Avatar
    Join Date
    Apr 2014
    Posts
    802
    Thanks Thanks Given 
    75
    Thanks Thanks Received 
    66
    Thanked in
    59 Posts

    Default

    Unbelievable

    A simple thread summon lots of math master on forum *-*

  8. The Following 2 Users Say Thank You to Fredystern For This Useful Post:


  9. #47
    Senior Member Fredystern's Avatar
    Join Date
    Apr 2014
    Posts
    802
    Thanks Thanks Given 
    75
    Thanks Thanks Received 
    66
    Thanked in
    59 Posts

    Default

    Quote Originally Posted by Melthyz View Post
    I agree with you, luck elixir is a total waste at underhul. I'm not sure if a lep makes a difference. I just put it on cuz it seems like the right thing to do.
    Nope, luck elixir was very useful in underhul, if you reroll in underhul after killing boss, its like you kill the boss twice so you got the second chance to set (only works if got legendary - arcane) in other hand if you got common - epic its work like in other map common become rare, rare - epic, epic - legend vroom explained it

  10. #48
    Senior Member
    Join Date
    May 2016
    Posts
    1,235
    Thanks Thanks Given 
    5
    Thanks Thanks Received 
    84
    Thanked in
    45 Posts

    Default

    Quote Originally Posted by Fredystern View Post
    Nope, luck elixir was very useful in underhul, if you reroll in underhul after killing boss, its like you kill the boss twice so you got the second chance to set (only works if got legendary - arcane) in other hand if you got common - epic its work like in other map common become rare, rare - epic, epic - legend vroom explained it
    Umm... and wat have u looted with luck elixir on antignome pieces? A mold perhaps? I haven't looted anything from underhul with or w/o luck elixir on..

    Sent from my SM-G928V using Tapatalk

  11. #49
    Senior Member Fredystern's Avatar
    Join Date
    Apr 2014
    Posts
    802
    Thanks Thanks Given 
    75
    Thanks Thanks Received 
    66
    Thanked in
    59 Posts

    Default

    Quote Originally Posted by Melthyz View Post
    Umm... and wat have u looted with luck elixir on antignome pieces? A mold perhaps? I haven't looted anything from underhul with or w/o luck elixir on..

    Sent from my SM-G928V using Tapatalk
    Nothing i didnt lvl up my character to 56 since i stay in 47 for pvp, i only run in underhul when my guildmate need help =$

  12. #50
    Senior Member
    Join Date
    May 2016
    Posts
    1,235
    Thanks Thanks Given 
    5
    Thanks Thanks Received 
    84
    Thanked in
    45 Posts

    Default

    Well until u run those maps a few hundred times n come up empty handed u wouldn't know wat we're referring to.

    Sent from my SM-G928V using Tapatalk

  13. #51
    New Member
    Join Date
    Jun 2016
    Posts
    5
    Thanks Thanks Given 
    1
    Thanks Thanks Received 
    1
    Thanked in
    1 Post

    Default

    Quote Originally Posted by Ireliaa View Post
    Btw its not %100 on 500k run with 0.0002 rate its just expected run count to get. You can get sooner or later than 500k run but most likely close to that number
    Bold emphasis mine: Careful with your wording there. 500k is *not* the most likely (which is of course 1, the mode of the distribution. IOW, you are *most likely* to succeed on the first attempt compared to succeeding on any other specific attempt).

    500k is the *mean* of the distribution. Nothing more, nothing less. It is ~1.44 X the median, or said another way, half of the players would see success before 500000/1.44 ~346575 attempts.

  14. #52
    New Member
    Join Date
    Jun 2016
    Posts
    5
    Thanks Thanks Given 
    1
    Thanks Thanks Received 
    1
    Thanked in
    1 Post

    Default

    Quote Originally Posted by Ireliaa View Post
    althought as you said the chance begins before 500k as half the chunk of bellcurve is there the biggest probability is the median
    but if you say total number of players who got earlier whi,ch is the area under curve yes you are right but also same amount got it more than 500k run on the right side

    my assumption is based on this actually just like coin flip


    put it simple its same chance to get it 346575 and 653425
    No, it's not.

    You are ~1.85 X more likely get you first success on trial 346575 versus trial 653425.

    The distribution of trials to first success is most certainly *not* a "bellcurve", it is a geometically decaying distribution (hence the name).

    The mean is greater than the median because of the *weight* of tails (number of trials), *not* their probabilities, and the "biggest probability" is most certainly not the median, it's the mode, which as I said is 1.

  15. #53
    Senior Member
    Join Date
    May 2016
    Posts
    1,235
    Thanks Thanks Given 
    5
    Thanks Thanks Received 
    84
    Thanked in
    45 Posts

    Default

    Any rocket scientists here that would like to make an input? Holy smokes, if it drops it drops, if it doesn't well maybe next time. There you go I just figured out the formula... You're Welcome!!!

  16. #54
    New Member
    Join Date
    Jun 2016
    Posts
    5
    Thanks Thanks Given 
    1
    Thanks Thanks Received 
    1
    Thanked in
    1 Post

    Default

    Quote Originally Posted by Ireliaa View Post
    ...then enlighten me if you make distrubution for all cases total runs lets say 1m total run over months how many pieces are we gonna see in game?
    say it like prob of 1 piece is:2 pieces is: 3 piece is:
    That would of course be the binomial distribution. Here's the PMF for 0 to 4 seen by trials:

    Name:  testing geometric distribution hypothesis fit - e.g. pity timer in games.png
Views: 237
Size:  19.1 KB

    Note that these are not a "bellcurve" - the probability is much too low for the # of trials under consideration to be using the normal approximation.

  17. #55
    Member Sorayai's Avatar
    Join Date
    May 2016
    Posts
    192
    Thanks Thanks Given 
    42
    Thanks Thanks Received 
    29
    Thanked in
    22 Posts

    Default

    Is this mathematic competition or just
    "What do u think"?
    Lol

  18. #56
    Senior Member
    Join Date
    May 2016
    Posts
    1,235
    Thanks Thanks Given 
    5
    Thanks Thanks Received 
    84
    Thanked in
    45 Posts

    Default

    Sorry, what was this thread about? I'm assuming rocket science or something of that nature...

  19. #57
    Senior Member stricker20000's Avatar
    Join Date
    Jun 2011
    Posts
    771
    Thanks Thanks Given 
    420
    Thanks Thanks Received 
    199
    Thanked in
    118 Posts

    Default

    Quote Originally Posted by Kaziscate View Post
    Correct me if I am wrong,but with the drop rate for the helm and belt being 0.0002% from elite Cryostar, it would be 0.00024% (with egg shell ring only),0.00026% (with lepre only),0.0003%(with 25% luck elixir only or have egg ring and lepre equipped),0.00032%(with 30% luck elixir only),0.00034% (25% luck elixir+egg ring[35% in total]), 0.00036% (25% luck elixir+lepre or have 30% luck elixir+egg ring[40% in total]),0.00038% (30% Luck elixir+lepre [45% in total]),0.0004% (25% luck+lepre+egg ring[50% luck]), and 0.00042% (30% luck+lepre+egg ring[55% in total]) Also, i am basing this off the new reroll system where 1 reroll counts as 2 runs


    Sent from my iPhone using Tapatalk
    I guess you can't say that because it depends on legendary drop rate. I calculated the chance for set item from boss if legendary drop rate was 1/30. Im not sure how you calculated that but it seems wrong to me. If I noticed it correctly you calculated those numbers based on a pink drop every run.

    So pretty optimistic numbers :P
    We need party-option for raid! :P

  20. #58
    Senior Member stricker20000's Avatar
    Join Date
    Jun 2011
    Posts
    771
    Thanks Thanks Given 
    420
    Thanks Thanks Received 
    199
    Thanked in
    118 Posts

    Default

    Quote Originally Posted by p-value View Post
    No, it's not.

    You are ~1.85 X more likely get you first success on trial 346575 versus trial 653425.

    The distribution of trials to first success is most certainly *not* a "bellcurve", it is a geometically decaying distribution (hence the name).

    The mean is greater than the median because of the *weight* of tails (number of trials), *not* their probabilities, and the "biggest probability" is most certainly not the median, it's the mode, which as I said is 1.
    What Ireliaa meant is that the probability regarding each try specifically is always the same.
    What p-value meant is that the probability to loot the set item after 500000 runs is higher than looting it after 700000 runs if you consider the total number of runs you have already done.

    I know what p-value is referring to but I can not reproduce the "course" of the probability in your diagram, e.g. maximum for 1 set item is at 500k trials.
    We need party-option for raid! :P

  21.   Click here to go to the next Dev post in this thread.   #59
    Spacetime Studios Dev Carapace's Avatar
    Join Date
    Apr 2012
    Posts
    2,137
    Thanks Thanks Given 
    139
    Thanks Thanks Received 
    2,208
    Thanked in
    875 Posts

    Default

    super fascinating thread.

    Just to inject a bit of how we utilize this kind of stuff, the original poster is doing an example from the point of view of a single person farming a single item for themselves. Unfortunately in an MMO the reality is that while you may want the item for yourself, there are thousands of others that also want this item. So if you have say 10,000 people with those odds farming for certain things then you start to see that it's not about an individual farming an item, but the entire server farming for an item. This is where the largest disconnect is when it comes to loot distribution, because you have to think in terms of the entire player base and not the individual experience. If the odds were in favor of every person farming for themselves in a reasonable amount of time no one would be farming anymore because everyone is sick of getting the drop that isn't worth anything. In the end it all goes back to basic economics, and supply and demand.

    Just food for thought
    | | | |

    Need help? Please visit our Support Website at http://support.spacetimestudios.com

  22. The Following 4 Users Say Thank You to Carapace For This Useful Post:


  23. #60
    Senior Member stricker20000's Avatar
    Join Date
    Jun 2011
    Posts
    771
    Thanks Thanks Given 
    420
    Thanks Thanks Received 
    199
    Thanked in
    118 Posts

    Default

    Quote Originally Posted by p-value View Post
    Bold emphasis mine: Careful with your wording there. 500k is *not* the most likely (which is of course 1, the mode of the distribution. IOW, you are *most likely* to succeed on the first attempt compared to succeeding on any other specific attempt).

    500k is the *mean* of the distribution. Nothing more, nothing less. It is ~1.44 X the median, or said another way, half of the players would see success before 500000/1.44 ~346575 attempts.
    346575 attempts lol. Thats about 20 years of farming if one run takes 5min and 4 hours of pure farming every day. I doubt anyone plays old game standard like this in 20years. Nothing against your numbers, they seem to be right. It's just ridiculous thinking about it as a individual player/farmer.
    We need party-option for raid! :P

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
  •