Act 3 Inferno spreadsheet to record ilvl 63 droprates

  • #1
    https://docs.google.com/spreadsheet/ccc?key=0ApnaR8CZeYdDdGZkeW5OMk16cDM3UVJMY0tQdDhLeHc

    will update it constantly

    letm e know if there is anything you would like me to add to it
  • #2
    can anyone help me with as formula that will calculate the percentages of c16 +q16 and q16 + r16. I have been doing those manually and its becoming a pita and would like it to just automatically update
  • #3
    For practicality you should probably just record everything in your inventory when you empty it and not try to track individual enemies.
  • #4
    Not sure why you are tracking this. Are you trying to disprove the presented drop rates or proved them?

    Your sample size will never be enough unless you get about 1000 people contributing data points.

    All your spreadsheet is showing is your personal experience which isn’t indicative of actual drop rates.

    Additionally magic find only increases quantity not quality. I got the impression from your spreadsheet that you assumed it increased quality. MF is really extraneous data unless you are trying to decipher how much MF will cause x amount of loot to drop.


    -edited for typos
  • #5
    Quote from Tenaciouzd

    can anyone help me with as formula that will calculate the percentages of c16 +q16 and q16 + r16. I have been doing those manually and its becoming a pita and would like it to just automatically update
    I don't understand what you want. You're adding C16+Q16 and Q16+R16, but what do you want them to be percentages of?
  • #6
    Quote from darsh333

    Not sure why you are tracking this. Are you trying to disprove the presented drop rates or proved them?

    Your sample size will never be enough unless you get about 1000 people contributing data points.

    All your spreadsheet is showing is your personal experience which isn’t indicative of actual drop rates.

    Additionally magic find only increases quantity not quality. I got the impression from your spreadsheet that you assumed it increased quality. MF is really extraneous data unless you are trying to decipher how much MF will cause x amount of loot to drop.


    -edited for typos


    MF is quality not quantity.


    EricGORe: In Diablo 2, from my understanding, magic find took the base item of a predetermined loot table, and gave you a higher chance of rolling a higher quality version of that item. In Diablo 3, this no longer seems to be the case and with loot tables being so vast now, magic find seems almost too arbitrary, it just doesn’t feel like it’s worth it. Could you explain exactly how magic find affects item drops in Diablo 3, as a lot of people seem to be unsure how the mechanic has changed from Diablo 2?

    Jay Wilson: The mechanic is exactly the same as Diablo 2, and as you describe it hear.
    And the loot tables are not more vast than D2 because we don't allow items to drop below level 50 in Inferno.
  • #7

    Quote from darsh333

    Not sure why you are tracking this. Are you trying to disprove the presented drop rates or proved them?

    Your sample size will never be enough unless you get about 1000 people contributing data points.

    All your spreadsheet is showing is your personal experience which isn’t indicative of actual drop rates.

    Additionally magic find only increases quantity not quality. I got the impression from your spreadsheet that you assumed it increased quality. MF is really extraneous data unless you are trying to decipher how much MF will cause x amount of loot to drop.


    -edited for typos


    MF is quality not quantity.


    EricGORe: In Diablo 2, from my understanding, magic find took the base item of a predetermined loot table, and gave you a higher chance of rolling a higher quality version of that item. In Diablo 3, this no longer seems to be the case and with loot tables being so vast now, magic find seems almost too arbitrary, it just doesn’t feel like it’s worth it. Could you explain exactly how magic find affects item drops in Diablo 3, as a lot of people seem to be unsure how the mechanic has changed from Diablo 2?

    Jay Wilson: The mechanic is exactly the same as Diablo 2, and as you describe it hear.
    And the loot tables are not more vast than D2 because we don't allow items to drop below level 50 in Inferno.



    Very nice explanation sir and seems the above poster took his assumptions and tried to make them fact. Nice little punch in the teeth.
  • #8
    Quote from darsh333

    Not sure why you are tracking this. Are you trying to disprove the presented drop rates or proved them?

    Your sample size will never be enough unless you get about 1000 people contributing data points.

    All your spreadsheet is showing is your personal experience which isn’t indicative of actual drop rates.

    Additionally magic find only increases quantity not quality. I got the impression from your spreadsheet that you assumed it increased quality. MF is really extraneous data unless you are trying to decipher how much MF will cause x amount of loot to drop.


    -edited for typos


    Just a note: Since the distribution comes from the same source for everyone, it is totally irrelevant how many people share data. The only thing that matters is the sample size.
  • #9
    I just started this last night, and the names are of relevance as you can see the Blood Clan Mauler packs gave generated more ilvl 63's. My sample size will continue to increase, and having a bunch of people with different amounts of magic find would through off the results. Just by using my own I can have a controlled expiriment.
  • #10
    And on a plus note I already figured out the formula for what percentages and they display correctly now.
  • #11
    Quote from Darkfrosty521


    Quote from darsh333

    Not sure why you are tracking this. Are you trying to disprove the presented drop rates or proved them?

    Your sample size will never be enough unless you get about 1000 people contributing data points.

    All your spreadsheet is showing is your personal experience which isn’t indicative of actual drop rates.

    Additionally magic find only increases quantity not quality. I got the impression from your spreadsheet that you assumed it increased quality. MF is really extraneous data unless you are trying to decipher how much MF will cause x amount of loot to drop.


    -edited for typos


    MF is quality not quantity.


    EricGORe: In Diablo 2, from my understanding, magic find took the base item of a predetermined loot table, and gave you a higher chance of rolling a higher quality version of that item. In Diablo 3, this no longer seems to be the case and with loot tables being so vast now, magic find seems almost too arbitrary, it just doesn’t feel like it’s worth it. Could you explain exactly how magic find affects item drops in Diablo 3, as a lot of people seem to be unsure how the mechanic has changed from Diablo 2?

    Jay Wilson: The mechanic is exactly the same as Diablo 2, and as you describe it hear.
    And the loot tables are not more vast than D2 because we don't allow items to drop below level 50 in Inferno.



    Very nice explanation sir and seems the above poster took his assumptions and tried to make them fact. Nice little punch in the teeth.

    I like how you assume that my misunderstanding of magic find means I had some sort of malicious intent and my education on the subject is a "kick in the teeth". It was a mistake and as such I apologize. In light of the clear explanation of how magic find works, to the OP I would like to see a much more accurate total of how much MF you indeed had at the time. Also you may as I stated earlier want to look for people to collaborate on the project for a better picture of drop rates perhaps even at different levels of magic find.

    I mean does MF work as a flat % increase or are there tiers in the progression?
  • #12
    Also, he is not even absolutely wrong. While Magic Find does indeed affect the "quality" of the item (being white, magic, rare or legendary), it does not affect the itemlevel at all, which might be, what he means by quality. So, you could interpret his statement like this:

    "Additionally magic find only increases quantity (= the average amount of magic, rare and legendary items from a specific set of enemies) not quality (= the average itemlevel of these items) ."

    Interpreted this way, the statement is true, even if easily misunderstandable.

    Also this means, that the point he was trying to make is absolutely valid: For a spreadsheet tracking specifically the itemlevel, magic find is irrelevant; it would just potentially increase the sample size per set of enemies.
    On the other hand, you might just as well add all white items to the sheet, since they could have rolled as magic or rare; again, for specifically tracking the ilvl, this does not matter at all.
  • #13
    All things considered, the way MF works, it shouldn't affect the relative percentages of ilvl gear. Meaning if you are looking to see if you really do get about 8% Ilvl 63 gear, MF won't affect it at all.

    Mf will affect the speed at which you get a good sample size however, because I somehow doubt that you are keeping track of white items.

    Now if you are looking to do things like stats on rare drops, or relative % of rares vs Magic... Then MF and Nef valor will make your data seem useless very quickly.
  • #14
    Quote from Küken

    Also, he is not even absolutely wrong. While Magic Find does indeed affect the "quality" of the item (being white, magic, rare or legendary), it does not affect the itemlevel at all, which might be, what he means by quality. So, you could interpret his statement like this:

    "Additionally magic find only increases quantity (= the average amount of magic, rare and legendary items from a specific set of enemies) not quality (= the average itemlevel of these items) ."


    But, you see, Darkfrosty wanted to point out how he got kicked in the teeth because he assumed that the guy was arguing about drop rates. You know, because it's impossible to have an adult discussion about conducting a proper experiment and keeping things as scientific and relevant as possible.

    @ OP
    Why break things out by mob? It's just adding tons of 0s to your sheet and making things abysmal to read. I'd suggest just keeping a running tally of how many of each iLevel you've found and not breaking it out by mob - perhaps break it out by player contributing as that would have some meaning. Something like 1 row per person or, I don't know. Just a suggestion.

    Also, the labels tend to be a bit confusing. I, personally, am not 100% sure what I'm looking at because things aren't all that clear and consise. I'm very interested in your results though.
    p400 :: 85.1k EK :: 2.38m TK
    Planet Express <PlanEx>
    (V) (°,,°) (V)
  • #15
    Its all just data for you to interpit how you wish. At 55 pack kills a couple of things that stand out to me is a) I have never gotten the same yellow named mob twice B) not a single ilvl 63 has dropped from a yellow named mob
  • #16
    Sample size is now at 100. As a side note all 105 rares at this point have been absolute shit.
  • #17
    Quote from Tenaciouzd

    Its all just data for you to interpit how you wish. At 55 pack kills a couple of things that stand out to me is a) I have never gotten the same yellow named mob twice B) not a single ilvl 63 has dropped from a yellow named mob


    Let me see if I can elaborate one bit. I'd like to point out that I don't hate or disagree with what you're doing, in fact I think it's much more noble than some of the shit-talking that occurs without the facts to back it up, so kudos to you for taking the initiative to bring some facts to the table, it's a mature and level-headed thing to do.

    I still don't understand why it needs to be broken down by specific mobs. Wouldn't it suffice to break it down simply by white/blue/yellow/purple? Do we really need to know if the white mob is a blood clan occultist? What does that level of granularity bring to the spreadsheet other than a LOT of 0s? 0s are meaningless data. You should consider condensing it simply to minimize the amount of meaningless data because it will vastly increase the readability.

    It would also allow us to see, much more simply, what the drop rates are across the mob types (white/blue/yellow/purple) as opposed to having to aggregate it ourselves. Perhaps you can make a cover sheet that condenses it, and then keep the detailed data on a 2nd sheet so that you don't lose that? I know that, more than anything, I'm more interested in the drop rate of 63s on blue/yellow mobs and not so much from white/purple. I find myself being unable to gauge that easily from your sheet, though. I'm not trying to sign you up for more work, I'm an IT guy and I love numbers and data and I like to help present it in better ways that's meaningful to more people. So please don't take it as outright criticism, but more curiosity.

    I also simply don't understand some of the labels on your columns, particularly the final two. Is column S (iLvl 63 rare count) / (total drop count)? Is it a ratio of something similar, but still different? Is column T (iLvl 63 rare count) / (mobs recorded)? Your column headings should be self-explanatory.

    Anyway, good work recording 100 mobs worth of drops. I'm sure it's an annoying task.
    p400 :: 85.1k EK :: 2.38m TK
    Planet Express <PlanEx>
    (V) (°,,°) (V)
  • #18
    Well initially when I started the spreadsheet, I didnt not think there would be so many different mobs spawns. But in actuallity, I havent ever gotten they same yellow mob twice, which is an interesting tidbit that the names seem to be completey randomly generated. Also yellow mobs are alot less common then the blue named mobs, yet drop absolutely shit gear. Why is this? Shouldnt the less common rare mobs be the ones dropping the better loot? I have the columns for ilv63/per pack kill, and ilvl 63/total rares because blizz never stated to my knowledge what the actual 6% was derived from. It seems closer to the champ packs you kill, rather than total rares you get. The point of the individual names was to see if there was a possibility the a specific champ pack name had a tendency to drop more ilvl 63s than others, which as of now the maulers have 2x the chance as occultists as i have gotten 3 from both types of packs but 15 kills verse 30 kills. But then again maybe I just have not killed enough mauler packs yet. Once I get this to 1000 kills i think we will have a more accurate picture, which is my goal.
  • #19
    alright i have spruced it up a little bit and hopefully its a little bit easier to read now, I have just begun pilling all yellow named mobs into 1 row and when i get a chance i will consolidate all the others into that row, letm e know if its easier to read!
  • #20
    Unneccessary, blue post confirms nerfed drop rates Act3/4 and they are looking into it.
  • To post a comment, please or register a new account.
Posts Quoted:
Reply
Clear All Quotes