Magic Find and its efficiency: A statistical insight

  • #43
    Superb work, my assumptions were confirmed. I hope the effort paid off for you in form of some nice drops ;)
  • #44
    Just perfect solid work as expected :-)
    Thank you for your insight!!
  • #45
    Really nice work.
    I thought about some analysis like this, too. But I could not manage it.


    What i'm wondering about: why magic and white drop chance is decreased?
    It seems so that we can never eliminate white drops before magic ones. The decrease in magic and white drops in your analysis is equivalent (percentual).

    So that we can never really get it to zero.

    And like someone before mentioned: when legendary drop chance decreases rare drop chance?

    Excuse me for my bad english, but it isn't my native language.
  • #46

    Superb work, my assumptions were confirmed. I hope the effort paid off for you in form of some nice drops ;)


    Thank you!
    As a matter of fact I only found a single rare item with a proper stat roll. The four legendaries I found were low on iLvl except one which proved to roll out shit stats. I guess lady lack wasn't smiling to me ;)
    I made around 8 mil selling all the stuff I got, but most of the cash I made ended up used again as I am trying to max my MF gear without losing all my dps.


    Just perfect solid work as expected :-)
    Thank you for your insight!!


    And thank you; I hope that programming skills of yours come to further use in whatever you find applications for (for those of you wondering what I am talking about, this guy made a smaller analysis himself).


    What i'm wondering about: why magic and white drop chance is decreased?
    It seems so that we can never eliminate white drops before magic ones. The decrease in magic and white drops in your analysis is equivalent (percentual).

    So that we can never really get it to zero.

    And like someone before mentioned: when legendary drop chance decreases rare drop chance?

    Excuse me for my bad english, but it isn't my native language.


    1) Magic and white drop is decreased because the rare drops are "stealing" their %'s - so to speak. It seems to be difficult to eliminate white drops completely, but I would like to get to a theoretical point of which it would be possible.
    Keep in mind that the research is based on gobos. While their base dropchances should be somewhat similar to other creatures dropping phat loot (elite packs and bosses), it is not certain what the base dropchances are and it is highly possible they are not sharing the same drop table. If the datapoint at 0MF gets increased in data size it may help to serve as a hint.

    2) We will most likely never get to a point where we can see legendary drops stealing the %'s from rare drops simply due to the huuuge amount of data it would require and even for 10 people farming 24/7 they'd may not make it before a patch could possibly change things.

    3) Don't worry about the language; if we've misunderstood your question, just let us know. English is not my native either.

    Edit: typo
  • #47
    +1

    Excellent post! Thank you for doing the research. I've been searching for answers for the last week or so and always got mixed results. "Just use NV or 100 mf is plenty". It's nice to see some of the mis-conceptions cleared up, or at least as good as it can be with out blizzard coming out and saying "These are the numbers". Again, thank you!
  • #48
    OK, not willing to gain any fame etc.
    Just contributing.
    25 full runs of Inferno Whimsyshire with 5 stacks of NV and MF gear (263% /w stacks).

    iLvL # %
    63 75 12% \
    62 125 20% 44% total
    61 75 12% /


    And now a little question. I've heard somewhere, that pony land is considered to be a part of Act III-IV on Inferno difficulty. So the drop rates for that location have to be somewhat equal to the named Acts.
    Now, today, there was a patch increasing drop rates, which if I remember well was as follows:

    iLvl 63: 16.3% \
    iLvl 62: 21.7% 65.1% total
    iLvl 61: 27.1% /

    I don't see them to be very similar... And that is 263%MF!
  • #49
    I think an interesting thing to notice here, is that MF increases in value the more MF you have :) (unless I've misinterpreted the item quality graph, in which case my excuse is it's late and I'm tired :D).
  • #50

    OK, not willing to gain any fame etc.
    Just contributing.
    25 full runs of Inferno Whimsyshire with 5 stacks of NV and MF gear (263% /w stacks).

    iLvL # %
    63 75 12% \
    62 125 20% 44% total
    61 75 12% /


    And now a little question. I've heard somewhere, that pony land is considered to be a part of Act III-IV on Inferno difficulty. So the drop rates for that location have to be somewhat equal to the named Acts.
    Now, today, there was a patch increasing drop rates, which if I remember well was as follows:

    iLvl 63: 16.3% \
    iLvl 62: 21.7% 65.1% total
    iLvl 61: 27.1% /

    I don't see them to be very similar... And that is 263%MF!


    You need more items to estimate things. You have found 625 items in that data pool and while it is a lot of items it does not give a completely representative picture.
    If you look through 2.3 in the original post you can see that my %'s are also varying from what they are supposed to be and that's with almost 3.2k items.
    None-the-less: I think the amount of items you have found having an iLvl of 61 is extremely small compared to the 27.1% it should have. Are you by any chance capable of increasing the sample size? Around 1.5k items should be viable, 1k items at least are needed to raise suspicion.


    I think an interesting thing to notice here, is that MF increases in value the more MF you have :) (unless I've misinterpreted the item quality graph, in which case my excuse is it's late and I'm tired :D).


    There are too few data points to make such a statement yet. While it is a thing that would be extremely interesting (especially because at this point it is believed that the value of MF decreases by the more MF you already have), I will need more sampling with high MF to see if such a trend is true.

    - - -

    Edit: I would also like to add a general point:

    People are more than welcome to contribute to this research; just like nosound did above. All I am asking for - if they wish to contribute to the research of section 2.2 - is to keep track of white/magic/rare items dropped with the MF they are running with (and if NV is included note the stacks as well) and note what they are killing (elites/bosses/gobos/ponies/whatever). The research I am doing at this moment is focused on goblins, but I won't be killing them forever and it would be nice to see if elite farming has similar results (even though they are influenced by guarantied rare drops). One thing that would be very interesting is to make a large sample with 0 MF as this gives a hint about the baseline %'s.

    I hope that the original post will eventually be so well constructed that it will serve as a single thread where people can get all of their questions about MF answered. And in order to get there I most likely need a bit of help.

    People are also more than welcome to find relevant blue posts / similar official statements or previously performed research on MF (such as how the rolls are made) and contribute with this, as the original post will be continually updated with such. Just make sure it is reasonable.
  • #51
    If people didn't go out of their way to make stupid assumptions since day 1 you wouldn't have had to do all this work to prove something people already knew.
  • #52

    If people didn't go out of their way to make stupid assumptions since day 1 you wouldn't have had to do all this work to prove something people already knew.


    There is still some interest about the research in matter of computations.

    - - -

    [Update]

    I added another data point in 2.2 (MF = 270%) and changed the conclusion and outlook to two separate sections. In the outlook I have included the general response in post #50 above.
  • #53
    Great work, it's great to see someone having effort this big to help the whole community and trying to figure the things out. Keep up the good work!
  • #54
    --sticky--
  • #55

    --sticky--


    Yay
  • #56
    and now front page =)

    Was shocked by the amount of work put into this, please keep it up!
  • #57
    Greatest post I've seen on this site since Beta
    These guys I'm sure agree...
    "
    187 user(s) are reading this topic

    108 members, 75 guests, 4 anonymous users

    Duh, Bigginz89, twopac187, neiljo, Molster, Daegal, Hestedraeber, MasterFischer, Seditiar, Daeveren, Nekari, Arqentum, Sae, killer625, Sham0wn, Icebear,Trashing, Gensui, xvire69, venoxis, Duckintheriver, iceeagle, Tsueah, Dilusion, Xhion, El3ss4r, barra84, leisuRe, Ashurin, Artemus, Penetier, Lastdoughnut,Bille, akilinik, nugiyen, Akisto, oilam, Quisari, Ulto, far0sino, Dominictus, Grozak, Microdiablo, chaud, Number1337, Juul, TheCurse, Sonicboom22,Shnackaran, Mikoma, Moppi, Dentoin, Warguyver, sarbehandlad, Akiyama, Memo85, Witver, Frelphin, Onkel, Hesten, Swiftdeathx, EnderWiggin, Riezier,xandrac, marcecko72, Bloomy, darkwarrior42, ayra, DaveC524, Jack_The_Jackal, kektonic, vaelrock, darchii, PaavoLepisto, Xathian, Braindog, Epialt,Freshman, LaMu2560, Gulvklud, Ironleg, Ploben, C55S, Datatap, Redmogg, MeraxesIF, Striderka, Crumak, Unrated1992, Plocer, DiabloSin, Slipknot06,Darabus, drewaz, Puttah, sp00n82, keiwo, adammichaelchess, Rageweawer, neolgs, head0r, Shadowspyes, Tyburn, happyh7, lMarcusl, Swiem, SolHellyon,kangaxx"
  • #58
    Amazing job. Interesting read!
  • #59
    Very good work, but im not sure why this was needed, MF was/is well understood from diablo II and is on several wikis containing the same information as you posted. Blizzard blue post confirmed that MF works the same in D3 as it does in D2.

    Perhaps now I wont have to reiterate in threads that MF does not increase the chance of getting higher ilvl now tho.
  • #60

    Very good work, but im not sure why this was needed, MF was/is well understood from diablo II and is on several wikis containing the same information as you posted. Blizzard blue post confirmed that MF works the same in D3 as it does in D2.

    Perhaps now I wont have to reiterate in threads that MF does not increase the chance of getting higher ilvl now tho.


    You pretty much answered your own question.

    Edit: It will also - if enough data gets collected - serve as a sample for computations. One thing I am hoping for is to compute the % increase for a random number of MF and then see if that increase is solid for other farming situations such as elites.
    I did not make all this work to come and say "Hey look what I just figured". It is to back up the statements that are out there and as far as I know there is no graphical display of the effect of MF. The future plans for the research simply just tagged along.
  • #61
    awesome work!
  • #62
    MAn, i know what is your deal... you are a engineering student!!! =DDDDD
  • To post a comment, please or register a new account.
Posts Quoted:
Reply
Clear All Quotes