PC Gaming News
Page 4 of 6 FirstFirst 123456 LastLast
Results 31 to 40 of 53

Thread: magic find

  1. #31
    GWOnline Content Team
    GWOnline Site Pal
    Achievements:
    Social10 PostsVeteranCreated Blog entry10K Posts
    Alaris's Avatar
    Server

    Kaineng
    Guild

    The Order of Dii [Dii]
    Posts

    22,615
    My comment was about people not collecting data but relying on their impressions. Data is good.
    == Alaris & clone ==
    Proud Officer of The Order Of Dii [Dii] - join us
    You can tell the quality of life of people by what they complain about

  2. #32
    Quote Originally Posted by Alaris View Post
    My comment was about people not collecting data but relying on their impressions. Data is good.
    Indeed, if I get a magic find item I will indeed do my best to note how many good items I get from random drops. Shoot, if I get good items at all I'm sure I'll take note. I got like 1 or 2 good items during the last bwe, but none that had upgrade components on them already. Probably not surprising since I didn't break level 10.

  3. #33
    My comment in response to yours, Alaris, was that it's literally impossible to test harniq's hypothesis. There is literally no amount of data that is feasible for us to collect that would prove if his hypothesis was right or not. I don't think ANet is so stupid as to give as an effect that is indistinguishable from not having it, so we don't need to test it anyway.

  4. #34
    Quote Originally Posted by Alaris View Post
    NOOOO! Me first!

    The 5% risk is the risk of false-positive, which is not really a concern here. Magic find probability is low, so we'll need to have a very large sample size to actually test for the effect at a good power. For a 1% base rate and 300% magic find (so 4% magic find rate), we need to test using a 0.01 binomial with a sample that would give us good power to detect a difference with a 0.04 binomial. Alternatively we can plot the number of rare drops per kill as a function of magic find, and get a linear correlation.

    Over the total drops combined across players over a weekend, it'd be trivial to find a significant effect.
    See, you managed to explain a bit better =) Also do not forget that correlation does not imply causation. If MF correlates with more magic drops, it doesn't directly prove that MF is causing more drops. Real prove is in the formula... which we know since virtually every game with MF has the same base formula. Google "diablo 2 mf" if anyone doesn't believe me....

  5. #35
    The thing is, I think the hypothesis that magic find is multiplicative instead of additive is incorrect, and that even if I'm wrong it's still the harder of the two to prove.

    If the hypothesis that magic find is additive is correct, then all you really have to do is get a player or two to use gems to get 50% magic find and see if magic items drop more than half the time. If they don't drop nearly that often then the hypothesis is incorrect. So go ahead and test the multiplicative hypothesis at that point. Unless of course, you can think of another option that would also be easier to test.

    Quote Originally Posted by harniq
    Real prove is in the formula... which we know since virtually every game with MF has the same base formula. Google "diablo 2 mf" if anyone doesn't believe me....
    Yes, because Anet copies all of diablo's formulas just like it copies the holy trinity from WoW. We should totally expect them to use the same formula because diablo 2 was good, even though the percents are completely different from each other (what, 200-1000% in diablo 2 vs 1-50% in gw2). If anything we should do the math and realize that they probably won't be the same. Also, googling "diablo 2 mf" only gets me a bunch of diablo 2 sites, shocking I know.
    Last edited by djacob; 12-05-2012 at 00:07.

  6. #36
    GWOnline.Net Member Skyy High's Avatar
    Posts

    5,223
    Quote Originally Posted by djacob View Post
    Yeah but... if people are saying they noticed a difference, I'm going to assume that it is an added percent and not a multiplied one
    Yeah....I'm just going to go ahead and assume that if people are saying they noticed a difference, they're wrong. There is no possible way anyone could actually run the trials to test how MF impacts drop percentages with this little time spent in-game.

  7. #37
    People seem to be forgetting that if harniq is correct, not even a computer will be able to detect the benefits of MF. If djacob is correct, it would be entirely plausible to see prelimary data that fits the trend. It wouldn't confirm anything, but getting 50% better drops (presumably indefinitely since we get all those free gems) you could notice.

  8. #38
    During the stress-test I managed to get an item with +5% magic find (I think). So based on that getting 100%++ sounds quite reasonable.

  9. #39
    Quote Originally Posted by Fluffball View Post
    People seem to be forgetting that if harniq is correct, not even a computer will be able to detect the benefits of MF. If djacob is correct, it would be entirely plausible to see prelimary data that fits the trend. It wouldn't confirm anything, but getting 50% better drops (presumably indefinitely since we get all those free gems) you could notice.
    Indeed, as Fluffball points out I'm not saying there's any confirmation here whatsoever, just that if people do notice a difference it might not be their imagination.

    Quote Originally Posted by Wethospu View Post
    During the stress-test I managed to get an item with +5% magic find (I think). So based on that getting 100%++ sounds quite reasonable.
    Unless there's a hard cap, you know like anet loves to put on things like ias, ims, armor, etc. I wouldn't be surprised if there was a hard cap somewhere between 50-80%. That way there would be no need to spend money on the magic find buff if you have the right equipment, but you might still want to so that you can equip a stronger set of armor. Ya'll remember this is anet right? They're trying to make it so the cash shop isn't required to play the game efficiently. At least they were the last time I checked, regardless of what some of you think.

  10. #40
    GWOnline.Net Member
    Posts

    6
    Playing through stress test, at some point i noticed a consumable adding +50% magic find for an hour. No idea where it came from, i suspect personal storyline, as i was doing them intensively at that time.

    I found three green items during next 20 minutes (green focus and rifle, both with an inscription and green pants), but i did move to queensdale swamps, so not sure if increase in drops wasn't caused by location change.

Posting Permissions

Posting Permissions

Smilies are On
[IMG] code is On
HTML code is Off