• 0

    posted a message on Patch 2.1 Roundup - Minor Changes
    Another minor - though undocumented - change: There now weapon racks in Pandemonium Fortress that are unique to that place.
    Posted in: Diablo III General Discussion
  • 0

    posted a message on Gambling Drop Rates Summary, Part 2: Share Your Post-Hotfix Loot From Kadala
    • Item Type: Amulet
    • Number of Blues: 22
    • Number of Rares: 119
    • Number of Legendaries: 3

    Posted in: Diablo III General Discussion
  • 0

    posted a message on Loot 2.0, discovering its true potential.
    If only half of these changes were implemented, I'd be spending most of my game time in Diablo III again. Leave it to a player to realize what keeps players coming back to a game, I guess. Heck, I'd pay for these changes to be implemented. All of them substantially increase the replayability and the freshness of the game (save perhaps the adjustments to the legendary beam of light, which is purely cosmetic: you merely bring the "wow" moment from the point in time when you identify the item to the point in time when it drops... it's a novelty that will wear off after a few dozen drops - but then again, it's so easy to implement, the art/dev team might as well go for it).

    I sincerely hope Blizzard is listening.
    Posted in: Diablo III General Discussion
  • 2

    posted a message on What do YOU do differently compared to the cliché/cookie cutter builds?
    Hello, fellow monks! :)

    Lately both these boards and the official forums offered little to no inspiration for exploring other aspects or variants of monk playstyle, itemisation or stat focus. Most threads either ask for gear/upgrade advice or are guides to farming something as efficiently as possible, with the occasional newbie question popping up here and there, but the clear common focus of almost all these threads is efficiency.

    While I agree that to avoid feeling like you're not progressing at all, you can't choose a very inefficient build to do your farming and there are some mandatory choices regarding skills, I'm a bit astonished to see how many people abandon the concept of thinking for themselves and finding the answers to itemisation or stat focus questions on their own, and instead rely on guides, tutorials and premade builds - heck, even gear purchased by someone else - to get their characters higher on the efficiency scale. This is great for instant gratification and noticable leaps in xp/loot gain over a given period of time, but I believe that in the long run this causes you to lose interest in the game much quicker and become a "burnout".

    So in this topic I'm looking for things you enjoy doing differently from what the guides, tutorials and forum posters tell you, regardless of whether it makes you more efficient at farming or not. Let's share our unique (or not-so-common) quirks, variants and obsessions and maybe some of us will be inspired by the choices of others and start experimenting, providing a breath of fresh air in their adventures.

    I'll start off by listing my two guilty pleasures as a monk:

    1. Life regeneration

    This stat is by far my favourite, so much so that I have 2500-3000 life per second depending on my equipped items and aura. Most monk guides and tutorials fail to even make mention of this affix, presumably writing it off as useless, but I've found that it has tons of utility in almost any combat situation, because it's the only source of healing that doesn't require you to do anything. This can be a life saver for hardcore players experiencing a lag spike, but even in softcore it means that you can focus a lot less on life steal/life on hit for weapons and jewelry, resulting in a cheaper way to upgrade these crucial equipment slots. Also, you don't feel like you have to keep attacking when surrounded by elemental damage sources such as desecrator, molten or plague pools, since with high enough life per second you can counterbalance the damage taken from them, providing you a safe way to reposition yourself if necessary.

    Life regeneration also allows you to crank up the damage output against reflect damage packs some more. After a certain dps mark, life steal alone can get completely nullified by damage reflect, making you slowly but steadily lose health if you increase dps further, whereas coupled with the always active life regeneration, this dps mark will be much higher, meaning you can go all out on them without having to watch cooldowns.

    I've been pleasantly surprised by how people react to this unconventional twist in my monk. Most people I've done ubers with doubted my ability to survive MP9-10 back when I had only 32k health and low life steal/life on hit values. But since every time I use serenity or seven-sided strike, I replenish a lot of health without taking damage and all three uber fights have some damage source that's present almost all the time, life regeneration is the primary source of safe healing for when things get rough. I've had people whispering me that they were sure I was going to lie on the ground the entire time with the defensive stats I had, and it felt good to prove them wrong. Some people even looked me up on the life regen leaderboards on DiabloProgress and added me as friends to exchange ideas.

    2. Sword & Board

    Another thing that most monk guides rarely recommend is not going dual-wield. But I just can't help it: Sacred Shields look awesome and I've always wanted to make a tanky monk to hold any kind of monster at bay while my friends are laying the smackdown on it. Not much to explain here: you trade critical damage and attack speed for critical chance and defensive stats. Still, you can arrive at a decent DPS total if you're lucky enough to find a nice shield in the auction house, and good shields are generally much cheaper than good off-hand weapons.

    Shields, however are frowned upon even worse than life per second on gear (probably because they can't tell you're stacking that by looking at your character). I've been turned down by not one whirlwind barbarians for uber parties with the explanation of "LOL shield monk, no thx". When I ran into one of these barbarians in someone else's uber party, he made perfect use of his whopping 225k dps by lying dead on the ground during the majority of all three fights, while my shield (and life regeneration) kept me alive easily so I could tank attack the ubers non-stop and provide a DPS boost to the party by recasting MoC/Overawe every 3 seconds (since nothing during the fights required me to use any of my defensive cooldowns). Many people - especially those playing barbarians and DH-s with their ridiculous paper dps numbers - disregard the fact that it's not enough to have good dps on paper, you have to be able to continuously attack and not have to run around (or lie dead) to make the most of it. If a shield helps you achieve that without dropping billions of gold into upgrading each gear slot, it's a worthy investment.

    So, what is everyone else's way of breaking free from cookie cutter monk conventions? Come on now, don't be shy, there's no wrong answer. :P Remember, this isn't about efficiency, like most of the other threads out there, this is about creativity, personal preference and - most importantly - personal enjoyment of the game.
    Posted in: Monk: The Inner Sanctuary
  • 0

    posted a message on Magic Find and its efficiency: A statistical insight
    Quote from ztking

    so are you saying there is a guaranteed drop at each stack but only the last two are guaranteed to be rare?

    From my personal experience, minibosses have always dropped 3+x items, where x=the number of NV stacks. Back before 1.0.3 (I think it was...), you used to get 2 guaranteed rares at NV5 (and no guaranteed ones at NV4), then they nerfed it to one guaranteed rare, then buffed it again to one at NV4, a second at NV5. But the additional drop per NV stack has always been there, from what I witnessed.
    Posted in: Theorycrafting and Analysis
  • 0

    posted a message on Magic Find and its efficiency: A statistical insight
    Quote from Kozik

    Sny83, in post #787 above, it appears that you used the old droprates. Here is the blue post excerpt describing the change in droprate.

    D'oh, I had completely forgotten about that droprate fix between the two major patches... I'll go and update the values. Kudos for pointing this out.

    EDIT: The values are now based on the drop chances from the hotfix patch notes.
    Posted in: Theorycrafting and Analysis
  • 0

    posted a message on Magic Find and its efficiency: A statistical insight
    Quote from thundersteele

    So, all that being said, I went to the AH to check on sets, and at least for DH's and Monk's the class specific armor piece (chest/head respectively) is in higher supply than the non class specific armor pieces. I found about 30 pages of monk spirit stones and demon hunter cloaks compared to around 20 pages for the other armor slots. Of course that's hardly evidence for anything, since not all set items are equally good.

    Hehe, I actually did the same with the EU Auction House. The 6 possible ilvl63 set chests and Tyrael's Might each yielded between 7 and 11 pages worth of results, while Natalya's Embrace and The Inquisitor (ilvl63 set/legendary cloaks) had about 15 pages each. They may not be significantly different, but I'm inclined to believe that class-specific armour pieces have a smaller chance of dropping than normal armour, so that might explain why there wasn't 20-25 pages of cloaks listed.

    To exclude the possibility of class-specific items skewing the results, I looked at Vile Ward (the only ilvl63 legendary shoulders) and the three ilvl63 legendary/set pants available (Depth Diggers, Blackthorne's & Inna's set pants). Sure enough, there were 45 pages of Vile Wards compared to only 20 pages of Depth Diggers, 11 pages of Inna's Temperance and ~14 pages of Blackthorne's Jousting Mail. Since all these items are fairly coveted, not to mention expensive (at least according to EU prices), any drops are very unlikely to be salvaged/sold when found. The fact that the amount of pages are similar might indeed be an indication that the drop probability of a legendary drop of a certain ilvl and type is divided amongst all legendaries for that ilvl/type.
    Posted in: Theorycrafting and Analysis
  • 0

    posted a message on Magic Find and its efficiency: A statistical insight
    Quote from thundersteele

    I don't think you can use the abundance of Legendaries on the AH as a measure of anything.

    Agreed, and I wasn't using it, merely remarking upon it. As I have said, it's just a gut instict. :)

    Quote from thundersteele

    Your calculation is interesting, but I don't see a drop scenario where it would apply. If, as you say, the item type (including ilvl?) is determined first, and then the quality (rare, magic,...), then what matters is whether there is at least Legendary corresponding to that item type or not. It doesn't matter if there are 6 Legendaries of the same type (e.g. ilvl 63 chest has that many options). For your calculation to apply, the game would have to roll separately for each Legendary/Set Item, and a chest would have six times the base probability of being Legendary.

    You're right, and it is very much possible that legendaries sharing a base item type/ilvl combination actually have a lesser chance of dropping. If I were Blizzard, though, I'd make sure that individual items for a set are equally likely to drop. I'll try to get around to making a version of the calculation taking into account legendaries sharing an item type/ilvl, but I'm afraid I won't have the free time for that until next weekend. :(

    Quote from thundersteele

    Concerning the Jewelry drops, Cyeron has mentioned about two pages back that slots 2 and 3 can indeed drop jewelry. I currently think this is done similar to the guaranteed rare drop (which ends up being 4 prop most of the time), namely if all other quality rolls fail, the last roll (1 prop magic) has a 100% chance. Maybe that's not even necessary, since the base item is a 1 prop magic, so it's the natural fallback.

    Damn, I missed that piece of info. I should read back further into the topic next time, but it just keeps growing so fast. :o Thanks for pointing it out.
    Posted in: Theorycrafting and Analysis
  • 0

    posted a message on Magic Find and its efficiency: A statistical insight
    @regenwurm007: I like this idea of using items which can't appear in normal (white) quality to determine loot generation order, although your calculations assume that all loot slots are capable of dropping jewelry. I wonder if this is the case, though, since if quality was decided before item type, items slots 2 and 3 could theoretically roll common (white) for quality and then ring/amulet/quiver/mojo/source for type simultaneously, thus producing an elite kill with only 2 drops (or 3 with NV rare added), and in our substantial amount of kill data, nobody has encountered such an elite before. Of course there are alternatives:

    - slots 2 and/or 3 can't roll jewelry or off-hand items (this should be easy to prove right or wrong, unfortunately item type information is not readily available in Cyeron's sequenced data collection)
    - quality is determined after type, and for jewelry/off-hand items, the drop probability of common quality is set to 0
    - etc. (other theories, anyone?)
    Posted in: Theorycrafting and Analysis
  • 0

    posted a message on Magic Find and its efficiency: A statistical insight
    I did some number crunching myself, primarily to find out whether it was time for me to move on to farming Act III instead of Act II, and I thought I'd share my line of thought with you guys. Please feel free to point out any errors or omissions I might have made, or any other form of constructive criticism. B)

    Before I get any further, I'll state up front that I don't share the idea of loot generation process described in chapter 2.1. of the original post. I believe item type is determined before item quality, in other words, we know if an axe is going to be an axe before we know it's going to be common, magic, rare or unique. This was actually how it worked in Diablo II as well and I see no reason why this would've changed. Determining which scenario is true is of course very hard, the best I can come up with is finding an ilvl with hardly any legendaries sharing it and finding another (ilvl63 being the obvious choice here) with which a truckload of legendaries can spawn and counting them, but since there's a lot more people farming Inferno than, say, Act III normal or Act II nightmare, and most low level legendaries never make it to the AH, this would only work in theory and not in practice. So for the remainder of the post, I'm going to assume that item type is chosen before item quality and consequently, the number of possible legendaries for an ilvl indirectly influences the perceived drop chance for legendaries of that ilvl.

    The questions I wanted to answer was: how much better - in terms of legendary and set drops - is Act III compared to previous acts in Inferno? Is the reduction in killing speed my monk will undoubtedly suffer (due to his low DPS) outweighed by the relative increase in sellable legendary loot?

    To that end, I went to D3DB and looked up all 1.0.4. legendaries with a level requirement of 60, grouped them by ilvl and summed them up. The numbers are:
    #ilvl_61: 24
    #ilvl_62: 31
    #ilvl_63: 79

    Thats a total of 134 legendaries. I had to filter out the 1.0.3. legendaries manually and it'd be pretty time consuming to verify the correctness of the data on that site, so there's a small chance that the numbers are off by 1 or 2, but they should be precise enough for the purpose of this test anyway. This means that 17,91% of end-game legendaries are ilvl61, 23,13% are ilvl62 and a whopping 58,96% are ilvl63.

    Back in a hotfix patch note shortly after 1.0.3, Blizzard disclosed the drop ratios for ilvl61-63 items for the last two acts of Hell and all acts of Inferno. Now, this might have changed in 1.0.4., but since there's no official source that states this, I've assumed it to be correct for the current build as well. I looked up this data and paired the ilvl ratios with the legendary ratios to create a value that can be used as the basis of comparison using the following formula:


    The results are:

    Hell Acts III&IV: 0,139*0,1791 + 0,0345*0,2313 + 0 = 0,03287
    Inferno Act I: 0,239*0,1791 + 0,126*0,2313 + 0,048*0,5896 = 0,1002
    Inferno Act II: 0,233*0,1791 + 0,186*0,2313 + 0,093*0,5896 = 0,1396
    Inferno Acts III&IV: 0,271*0,1791 + 0,217*0,2313 + 0,16*0,5896 = 0,1931

    By normalizing the efficiencies with the baseline being Inferno Act I, the values are as follows:

    Hell Acts III&IV: 32,8%
    Inferno Act I: 100%
    Inferno Act II: 139,32%
    Inferno Acts III&IV: 192,71%

    This is already pretty convincing stuff, consindering even if you kill half as fast as you do in Act I, you'll get the more or less same amount of legendaries and that's not even considering the fact that there'll be more ilvl63-s among them. Even when compared to Act II, there's a ~38% increase in efficiency, so if your kill speed doesn't decrease by this percentage by going Act III, it's already worth it.

    Not being fully convinced by these numbers, though, I made another go at this efficiency comparison, this time excluding ilvl61 legendaries from the equation, since most of them aren't worth much when compared to ilvl62-s and 63-s.

    Hell Acts III&IV: 0,0345*0,2313 + 0 = 0,00798
    Inferno Act I: 0,126*0,2313 + 0,048*0,5896 = 0,05744
    Inferno Act II: 0,186*0,2313 + 0,093*0,5896 = 0,09785
    Inferno Acts III&IV: 0,217*0,2313 + 0,16*0,5896 = 0,1445

    This puts the efficiency for the various acts at:

    Hell Acts III&IV: 13,89%
    Inferno Act I: 100%
    Inferno Act II: 170,35%
    Inferno Acts III&IV: 251,57%

    The numbers here are much more convincing, and based on experience with my monk so far in Act III, the ~48% efficiency gain in terms of legendaries is enough to outweigh the speed loss when farming there instead of Act II. Now why didn't I listen to my instincts saying the same when the patch came out...

    Some notes:
    - as should be obvious, these calculations don't take into account the actual drop probability of legendaries, or even magic find itself, but I felt it had a relation to the topics discussed here. :)
    - I guess that - given a large enough sample size of legendary drops (complete with act and ilvl information) - we could deduce whether item type or item quality gets chosen first when generating drops, but my gut instict tells me that there are simply too many ilvl63 legendaries in the AH in relation to ilvl61/62 legendaries for the "quality before type" scenario to be valid, and the large number of possible ilvl63 legendaries in comparison to ilvl61/62 ones provides a plausible explanation for this perceived abundance. AH/player tendencies are hard to take into account when trying to make estimations of this sort, unfortunately.
    - if anyone has conclusive evidence of the loot generation order being as described in the OP, please let me know, along with providing an anti-foot-in-mouth device and a dunce hat for making a fool of myself :P
    - thundersteele pointed out that the fact that multiple legendaries may share the same ilvl and item type would skew the picture painted by these calculations somewhat, so I'll try to provide an alternate version of the calculation in an updated post once I find the time to group legendaries sharing ilvl/type together

    EDIT: Updated the calculations to reflect the current ilvl drop rates from as per Kozik's post below.
    Posted in: Theorycrafting and Analysis
  • 0

    posted a message on Magic Find and its efficiency: A statistical insight
    All right, I managed to read up on the last one or two pages of posts and the revamped original one. Good job on that, Cyeron, looking better than ever!

    I did, however, find a minor oversight in chapter 5.3.: the two graphs appear to be outdated as they don't match up with the numerical values from the tables above them. In both graphs, the gradient of the "rare" curve between data points ~230 and 267 is negative, whereas the numerical values suggest a positive slope. Even more apparent is that in the second graph, the "white" curve is above the "rare" curve for x=267 whereas the percentage of white items found (30.51%) is below that of rare items found (31.03%) for this MF value, according to the table.

    This is almost certainly due to me having had horrible luck up until I submitted the first batch of results (~2400 items) - upon which the graphs are based - and then getting better drops for the subsequent batch (~3200 items), which would bring the curve more in line with its predicted shape.

    Also, here's some info on the the legendaries I've found so far in 1.0.4:

    #   Name			  ilvl Act  Drop location
    1x  Hellrack	 		  61   II   Cave of the Betrayer
    1x  Balefire Caster		  60   II   Desolate Sands
    1x  The Burning Axe of Sankis	  62   II   Vault of the Assassin
    1x  Promise of Glory		  60   II   The Unknown Depths
    1x  Stormshield			  63   III  The Keep Depths, Level 2
    1x  Schaefer's Hammer		  63   III  Fields of Slaughter
    Posted in: Theorycrafting and Analysis
  • 0

    posted a message on How should i gear my monk for magic Find?
    I'm running Act 2 with my monk wearing (not switching, mind you) 192 MF on my gear (267 MF after reaching 5 NV stacks), and I didn't really go for dexterity on items. I found dual resist items or even single resist+dex items with a nice amount of life regeneration on them to be very helpful in keeping me alive, even with low amounts of Life on Hit. I only have around 1500 dex so my damage output is sub-par, but I also 2000 life per second using MoH, and my resists are also pretty high (over 1100 when buffed with MoH [Time of Need]). I can facetank almost any elite pack and the ones that I can't I can do hit-and-run attacks against, while regenerating health quickly due to the 2k LPS.

    Weapon-wise I think high LoH weapons and high Crit Damage weapons are both viable, each having their pros and cons. With the life-on-hit variant I could stand still versus molten/fire chains/plagued packs and just outheal them, but the larger, high-health mobs with unlucky affixes sometimes got the better of me when I was trigger happy and let myself get surrounded. With my slower, critical damage-based weapon it's the other way around. High health packs melt quickly enough to not pose a threat even with nasty affixes, but I can't reliably tank quick mobs with molten/fire chains/etc. due to lack of high life on hit.

    What I don't recommend doing is getting MF on every slot. Some items for monks are just too expensive to get already to even dream about hoping for an MF variant. Gloves, ring, helm and shield come to mind, as these are hard enough to get good stats on as it is. Keep your attack speed/crit chance/crit damage gloves and rings on if you have them, and look for MF on other slots. For helms, if you're over 35-40k health, you can unsocket your amethyst and replace it with a topaz for the added magic find and once you're ready to switch back to tank/DPS gear, just do the opposite.

    One more thing: unless you want to do gear swapping, only upgrade to an item that has MF if it's not a great deal worse than the one you're currently using. Throwing away a significant portion of your defense just for 15-20% more magic find will just make MF runs feel useless and unrewarding. If your killing speed or your ability to take hits suffers too much, you might get twice as many rares, but you'll also take twice as long to clear a zone. Take your time when scouting for upgrades.
    Posted in: Monk: The Inner Sanctuary
  • 0

    posted a message on How often are you FINDING upgrades?
    I've had three upgrades drop so far since hitting level 60:
    1. I've found a 820 dps bow shortly upon entering Act I Inferno back in May, which trumped the ~600ish one I got from the AH. My DH used that until I switched to hand crossbows.
    2. Before patch 1.0.3, I had an amulet drop with 7.5% crit chance, 58% crit damage, 100+ vitality and a socket, which was a major upgrade at that point, and as yet I couldn't afford any upgrade to this one.
    3. Just today, I had a belt drop with 166 dexterity, 85 vitality, 67 all resistances and some other goodies. Pretty much exactly the stats I want from a belt, so my DH is very happy with this one.

    All in all, that's three upgrades in ~350 hours of game time, most of which was spent farming Act II (and a little Act I & III). Can't say I'm pleased with this... under 350 hours in Diablo II, I could gear a character's every slot for hell endgame three times over.
    Posted in: Diablo III General Discussion
  • 0

    posted a message on Ghom
    I actually killed Ghom over 2 minutes after he enraged. My DPS is kind of low (only 13k) and I was using Deadly Reach with Keen Eye at the time with hardly any LoH, but I had close to 2k life regeneration per second which helped me to stay on top of things. I ran in to hit him until my health was low then ran out and let it regen and healed up using skills and transcendence. Positioning the clouds is still important, since LPS can't outheal that. Now that I have a decent LoH weapon, I'd take Thunderclap over Keen Eye so I can facetank/DPS him longer before having to run out and heal.
    Posted in: Monk: The Inner Sanctuary
  • 0

    posted a message on Unique Monster screenshots
    What luck! That's how I'm currently naming them anyway. I'll get to cropping and uploading a few over the weekend then. :)
    Posted in: DiabloWiki
  • To post a comment, please or register a new account.