I wanted to post here and get some input in regards to how others plan to handle T3 loot for Naxx.
My guild uses Zero Sum dkp, and it's worked fairly well as a system. Unfortunately, we find the majority of some classes more negative than others. It's partly our own fault. We had the priest and hunter epic quest items for free. Meanwhile, rogues and warriors were paying more for their weapons (because they took, in the case of rogues, at least 2x1H, and some more who took both swords and daggers at different times). It's never been a problem, really, with multi-class items by and large, because we put class priority on the most coveted drops (no shaman with Spinal Reaper in our guild). Also, there's some disparity in class vs. class dkp simply because some classes are more overpopulated (warriors, rogues, and shaman) than others (priests, druids, and warlocks).
Now the question becomes, how do we determine loot priority when the average Priest is +400 DKP and the average warrior is -250 DKP? We're considering addressing it by collapsing some of the positive/negative ranges, but I'm not sure that is the best solution. If armor pieces just dropped as they did in MC and BWL, it wouldn't be an issue. Warriors could go to -2000 and druids could be +2000 and it wouldn't matter on the set pieces.
Is this something that will affect other guilds, or are we the only ones who screwed up melee wpn / OH pricing?
It seems like this always ends up being a problem with warriors that want to dps, but still need to pick up tanking gear. I'd imagine if you look up the amount warriors spend on armor compared to the other classes, you'll see a pretty hefty difference. If you want however, you could always do a weekly rotation, since the tokens probably are 100% like the qiraji bindings and such. One week healers get it, the next week tanks, the next dps.
Your pricing is bad. Even if you somehow made a "minimum/maximum" amount of dkp you can be at, it just means people are going to hover around one of those two thresholds forever.
So you really have two problems 1) how to fix the short-term (if you think it deserves fixing), then 2) fixing your pricing in some manner. If you don't attempt to fix both of these things separately, then you're going to encounter the same problem again and again every couple of months.
Your pricing has to reflect the fact that some classes (warriors, priests) require a 2nd set of gear for grinding/pvp, while others (rogues, hunters) generally do not. Thus your pricing needs to be set up in some way that reflects this without getting in the way of your pve gearing progress. If you do not do this then you will have warriors who have to buy two full sets of gear + 2 sets of weapons (a 2h for pvp and 1h+shield for pve), and they'll always be negative. Personally, we set 2h weapons and dps gear to be extremely extremely cheap, we don't want it to rot, but we want our tanks using their dkp on items that help them tank.
As for the short-term fix to get rid of the dkp spread... who knows. No matter what you do is going to create some drama from the people high on dkp, it basically depends on how much of a nerf you can shove down their throats without a rebellion.
This seems to be a common problem, zerosum doesn't balance well in regards to cross-class items. You see this weapons and archetype specific items.. Token system's rock for loot diversity, but as far as tracking and dishing out loot they can be quite the pain in the ass when dealing with loot in high demand by many classes.
I can see a few options to solving this, though they aren't really practical.
Possibility A.) Seperate DKP system for the affected classes, only for the tokens. Will be the most fair, but it will be a pain to manage. Multiple DKP tables can be a real cluster.
Possibility B.) Do nothing, let the imbalanced zerosum system play out. It will balance out over time, but the people low on the list aren't going to see these T3 tokens.
all in all, depending on JUST how imbalanced your system is... option B will probably be the easiest, though it will have some teething issues... Just depends on how much fuss you want it to cause =/
Funny you raise this issue -- we've been working through a lot of this internally. Our warriors, in particular, are way behind everyone else because our system was initially designed never imagining that a warrior and a priest would compete for the same drop. Basically we're just going to say "fuck zero-sum" for a bit, give warriors +15% to their earned point across the board when Naxx hits, and leave it at that. We're restarting our loot system when the Expansion hits and will fix all the flaws then.
Here's an alternate suggestion that we'd considered also -- I'm curious what others think of it:
If your loot system was designed around non-tokenized drops and less cross-class competition, why not turn tokens back into BWL-style drops, except without the potential for loot rot? If a token can be "Mage, Warlock, Warrior, Druid" either /random 1-4 when it drops or cycle through the four classes on weekly basis (obviously skipping to the next if no one of the chosen class wants/needs the item). It feels a bit odd, but if you think about it, you could achieve a result that's identical to the way drops worked in BWL, with the advantage of making sure you never get the dreaded "2xNemesis" off the boss.
all in all, depending on JUST how imbalanced your system is... option B will probably be the easiest, though it will have some teething issues... Just depends on how much fuss you want it to cause =/
We did this in AQ. But in Naxx, having warriors get no tank gear until everyone else has their stuff is not an option. If a guild's DKP system allowed that it would be pretty damn stupid.
...cycle through the four classes on weekly basis (obviously skipping to the next if no one of the chosen class wants/needs the item). It feels a bit odd, but if you think about it, you could achieve a result that's identical to the way drops worked in BWL, with the advantage of making sure you never get the dreaded "2xNemesis" off the boss.
This is what we do, and it has worked rather well for us. The only disadvantage I can ascribe to this system is the effort of remembering (and confusion when forgetting) which class's turn it is for the various tokens. Any other issues that have been raised so far I consider to be problems of perception. I think it is right to view the token system of AQ40 in the way you described--BWL with really nice advantages.
Alternatively, the traditional loot system of MC/BWL can be understood as the most extreme token system with 64 unique tokens, one for each item. Towards the other extreme would be an 8-token system, one token for each equipment slot of the set regardless of class. In this system it's as if a guild can assign the drop rate of the set pieces for themselves. Fewer tokens equates to giving more control to the guild in how it distributes loot, and that's always nice.
Like many guilds, we have a similar situation, and I'm about to propose a change the system, retro-fitted, to try and address this. (And the issue of priests needing 2 sets of gear as well).
Basically, every item in every slot can be upgraded, at a cost of X, which is the difference in the dkp cost between what you are using and what the item costs, + 5. (The +5 is there to give everything a cost and to not trivialise looting).
So, for example. A warrior has bought all his shiny tank gear, like a good chap. He then needs to start buying DPs gear. Instead of paying for a full new set of gear, he just pays the difference, +5 for each piece. Same for the priests and shaman who want to start buying dps gear for out of raid play, or the mages who want to buy resistance gear.
It works for weapons as well, with the only difference being that a 2 handed weapon counts as a seperate slot to your 1 hander + offhand "thing". Purely because it seems most classes swap between the 2 now depending upon your stage in the game.
As we price dps and alternate raid gear at roughly the same price of sets, it should mean that most people pay the minimum 5 DKP for an item, once they have bought *something* for that slot at that tier.
Obviously, it requires a lot of tracking, and it also requires your guild mates not to go bad, and end up with priests with + 600 damage and mages with Angelista's Grasp, but I'm hoping it will work though. So far havne't spotted any *massive* flaws with it, and am preparing a test copy of the DKp database to see how things will look when it is retro-fitted on.
It's the same for us. Warriors are among the lowest in DKP for us too and it is a bit of a stinger that you're guaranteed to be behind 15 other people for an armourpiece you want. It was particularly bad on Huhuran/Emps where Priests geared up in 3/5 of their set generally before any other class, sometimes taking two Bindings in a row.
We are already talking about refining our loot system for Naxx, even splitting it off from other instances as to prevent pointwhoring the "easier" instances like BWL and AQ for Naxx lootpieces, so I'll keep my eye on this thread and will welcome every suggestion.
We use a seperate dkp pool for (mc/bwl/onyxia/out door bosses) and AQ and will probably implement a third pool for naxxramas. This enables our shitty system to never get completly out of hand ;p . Since the full time tanks don't nuke each other that badly on tanking gear they will still have points left for armour items.
It's a pretty bad system but it works allright most of the time as long as people team play, if they don't we kick em.
Funny you raise this issue -- we've been working through a lot of this internally. Our warriors, in particular, are way behind everyone else because our system was initially designed never imagining that a warrior and a priest would compete for the same drop. Basically we're just going to say "fuck zero-sum" for a bit, give warriors +15% to their earned point across the board when Naxx hits, and leave it at that. We're restarting our loot system when the Expansion hits and will fix all the flaws then.
Here's an alternate suggestion that we'd considered also -- I'm curious what others think of it:
If your loot system was designed around non-tokenized drops and less cross-class competition, why not turn tokens back into BWL-style drops, except without the potential for loot rot? If a token can be "Mage, Warlock, Warrior, Druid" either /random 1-4 when it drops or cycle through the four classes on weekly basis (obviously skipping to the next if no one of the chosen class wants/needs the item). It feels a bit odd, but if you think about it, you could achieve a result that's identical to the way drops worked in BWL, with the advantage of making sure you never get the dreaded "2xNemesis" off the boss.
all in all, depending on JUST how imbalanced your system is... option B will probably be the easiest, though it will have some teething issues... Just depends on how much fuss you want it to cause =/
We did this in AQ. But in Naxx, having warriors get no tank gear until everyone else has their stuff is not an option. If a guild's DKP system allowed that it would be pretty damn stupid.
We also have this problem currently, and I can't be bothered to do anything massively elaborate to fix it seeing as Burning Crusade will necessitate a new system anyway. The 'take it in turns' thing is all I could think of doing to stop the priests and druids from getting 8/9 while the warriors are in 0/9, and I imagine that's how we'll handle it.
What do people think about altering item prices to reflect the fact that some classes need more loot than others, for instance the way someone above mentioned pricing DPS plate very cheaply? Having upgrade/downgrade paths makes that particular example irrelevant I guess, so perhaps something like reducing plate costs in general compared to cloth of a similar iLevel and quality?
It's not something I've done and it seems like an awkward bandaid for the problem really; personally I quite like having neat concise values for items based on level and utlility, irrespective of what item type it is. But that said, it would be beneficial to 'artificially' alter values in order to even things out.
BTW, as an amendment to what I posted above - retrofitting the system in this way has meant that the warriors who have earned a lot of DKP have suddenly shot up the charts, because they are no longer paying double for each slot and ever increasing DKP for ever decreasing weapon upgrades.
Some casters are still further up the chart, but that's because up until now they haven't been buying any alternate gear because it nukes their points for very little gain. (I have NW Bracers, why buy Bracers of Arcane Accuracy?). One of the benefits of this "mass upgrade" system is that they will all be much more keen to buy sideways items...
We use a split DKP pool with one for classed items and one for non class items; our line on this has been, if it has the tag "Class:" it comes from your Class DKP pool. With respect to the problem of certain classess having more DKP than others, the answer was: They have more DKP because they had less drops, this is a nice way to even it all out.
This solution also neatly steps around the issue of feral druids, shadow priests, dps warriros and so on who would rather spend all their DKP on dps toys, but are still required to pick up their tanking/healing gear to fulfil their primary raid role. We use loot prioritization on non-set items within reason; our second Ashkandi went to a paladin because it was his birthday so we arent total monsters.
I dont forsee major loot issues for us when Naxx hits, because any advantage one class has over another will last about 1 drop, given we use an open bid, no minimum system. Essentially people can pick up the last 2/3 pieecs of their Tier2 sets for 0 DKP if noone else from their class needs it, while people with 12930123 DKP will have to outbid people with 12930122 DKP for their first piece of tier 3, and so on. I imagine our DKP tables will be stabilized by cross class competition in AQ before we see our first Naxx drop.
We use a split DKP pool with one for classed items and one for non class items; our line on this has been, if it has the tag "Class:" it comes from your Class DKP pool. With respect to the problem of certain classess having more DKP than others, the answer was: They have more DKP because they had less drops, this is a nice way to even it all out.
Our priests don't have less drops, they just don't take anything because it's a sidegrade/situational thing and they'd rather save up for regalia and those token items. Even cheap resistance items that go for 5 points have to be forced sometimes. Yes, this needs to be discussed, but the issue remains that most of the times things get sharded because noone WANTS to take it rather then it not dropping.
Then again, you should hear the roaring from the back when us dps warriors refuse to pick up a tank plate item.
Originally Posted by Judia
This solution also neatly steps around the issue of feral druids, shadow priests, dps warriros and so on who would rather spend all their DKP on dps toys, but are still required to pick up their tanking/healing gear to fulfil their primary raid role. We use loot prioritization on non-set items within reason; our second Ashkandi went to a paladin because it was his birthday so we arent total monsters.
Eh, as a warrior I must disagree on the monsters part ;)
Originally Posted by Judia
I dont forsee major loot issues for us when Naxx hits, because any advantage one class has over another will last about 1 drop, given we use an open bid, no minimum system. Essentially people can pick up the last 2/3 pieecs of their Tier2 sets for 0 DKP if noone else from their class needs it, while people with 12930123 DKP will have to outbid people with 12930122 DKP for their first piece of tier 3, and so on. I imagine our DKP tables will be stabilized by cross class competition in AQ before we see our first Naxx drop.
I think the whole switching the class who can take it every week is probably the fairest. The biggest thing for me was certain classes/people getting e.g. Bindings 2,3 times in a row. Tier 3 is an upgrade for everyone and DKP should still reflect what you can earn for your effort in my opinion, just no hogging it for 2 months straight until every priest has it, then the rest of the guild can start.
nb: I use priests here as an example, insert any possible class when applicable ;)
We use a split DKP pool with one for classed items and one for non class items; our line on this has been, if it has the tag "Class:" it comes from your Class DKP pool. With respect to the problem of certain classess having more DKP than others, the answer was: They have more DKP because they had less drops, this is a nice way to even it all out.
Our priests don't have less drops, they just don't take anything because it's a sidegrade/situational thing and they'd rather save up for regalia and those token items. Even cheap resistance items that go for 5 points have to be forced sometimes. Yes, this needs to be discussed, but the issue remains that most of the times things get sharded because noone WANTS to take it rather then it not dropping.
Then again, you should hear the roaring from the back when us dps warriors refuse to pick up a tank plate item.
This wouldnt happen to us.
We dont have a minimum bid, therefore the item would be /rolled on for 0DKP if it would actually help PVE progress. Our system was specifically designed to stop any item that would help our raid progress being sharded because items such as resistance drops were seen as not worth the DKP at the time they dropped.
The only time items get sharded is because everyone has it already or it isnt worth a bank slot to anyone.
Originally Posted by Rane
Originally Posted by Judia
This solution also neatly steps around the issue of feral druids, shadow priests, dps warriros and so on who would rather spend all their DKP on dps toys, but are still required to pick up their tanking/healing gear to fulfil their primary raid role. We use loot prioritization on non-set items within reason; our second Ashkandi went to a paladin because it was his birthday so we arent total monsters.
Eh, as a warrior I must disagree on the monsters part ;)
We had 3 prot warriors and 2 fury warriors who all agred he could have it as a Birthday present.
Originally Posted by Rane
Originally Posted by Judia
I dont forsee major loot issues for us when Naxx hits, because any advantage one class has over another will last about 1 drop, given we use an open bid, no minimum system. Essentially people can pick up the last 2/3 pieecs of their Tier2 sets for 0 DKP if noone else from their class needs it, while people with 12930123 DKP will have to outbid people with 12930122 DKP for their first piece of tier 3, and so on. I imagine our DKP tables will be stabilized by cross class competition in AQ before we see our first Naxx drop.
I think the whole switching the class who can take it every week is probably the fairest. The biggest thing for me was certain classes/people getting e.g. Bindings 2,3 times in a row. Tier 3 is an upgrade for everyone and DKP should still reflect what you can earn for your effort in my opinion, just no hogging it for 2 months straight until every priest has it, then the rest of the guild can start.
nb: I use priests here as an example, insert any possible class when applicable ;)
The risk is people who decide that because there is NO chance of their class getting loot that week, they cant be bothered with showing up. Sure for EJ or DnT you could kick that guy but in many guilds we dont have the luxiary of being so ruthless. If we ran this rotation I could see us having 2 priests and 1 rogue on druid week, and so on.
Further, because we dont use a fixed price system, noone can maintain enough DKP to win more than 1 item, to win 2 items in a row a person would have to have more than double the DKP of the person second on the list, to win 3 more than tripple. and with cross-class tagged loot the ability to maintain more than double the DKP of every other class is essentially nil.
The risk is people who decide that because there is NO chance of their class getting loot that week, they cant be bothered with showing up. Sure for EJ or DnT you could kick that guy but in many guilds we dont have the luxiary of being so ruthless. If we ran this rotation I could see us having 2 priests and 1 rogue on druid week, and so on.
Don't random until you enter the zone, and make sure there is some incentive to show up? I'd shoot our rogues if they didn't show up over something as stupid as this. It should be explained to your guild that they are *guaranteed drops* (assuming they follow the paradigm setforth in AQ) and as such, they will be rotting in short order compared to any drops in the past with the exception of AQ set pieces.
Well, if you used that system, you'd want the different pieces all on different rotations, so that it might be the rogues' and druids' turn to get bracers, but the mages' and hunters' turn to get belts, and so forth.
Unless you're using /random or something, the alternative of just letting straight DKP decide who gets the token drops regardless of class yields the same theoretical problem. When we killed C'Thun for the first time everyone knew that the shaman (not me!) and mage who were #1 and #2 in points were going to get the two chestpiece tokens, but it's not like the other 38 of us didn't want to come because of that. That's what DKP is for in the first place. If you attend, you move up in line for your shot at the item. If you don't, then other moves ahead of you. That's all the incentive you need (aside from that whole "helping your guild and having fun" thing, of course).
Funny you raise this issue -- we've been working through a lot of this internally. Our warriors, in particular, are way behind everyone else because our system was initially designed never imagining that a warrior and a priest would compete for the same drop. Basically we're just going to say "fuck zero-sum" for a bit, give warriors +15% to their earned point across the board when Naxx hits, and leave it at that. We're restarting our loot system when the Expansion hits and will fix all the flaws then.
Here's an alternate suggestion that we'd considered also -- I'm curious what others think of it:
If your loot system was designed around non-tokenized drops and less cross-class competition, why not turn tokens back into BWL-style drops, except without the potential for loot rot? If a token can be "Mage, Warlock, Warrior, Druid" either /random 1-4 when it drops or cycle through the four classes on weekly basis (obviously skipping to the next if no one of the chosen class wants/needs the item). It feels a bit odd, but if you think about it, you could achieve a result that's identical to the way drops worked in BWL, with the advantage of making sure you never get the dreaded "2xNemesis" off the boss.
I've actually considered both of these solutions to exactly the same problem. I prefer the first, but haven't thought about exactly how to implement it.
Something to point out for people who want to try it, is that, for any guild other than EJ, causing inflation in zero-sum isn't really something you can take lightly. Wait until you go through a guild phase with a high amount of turnover (we recently have). On average, people leave with negative points (we won't speculate as to why this is), so the theoretical "zero sum" gradually drifts upwards over time. And since bringing new members into a system whose sum floats around a bit is at least a nontrivial operation, it's something to be careful with.
It's not that big a deal, though. If you go with it, you can still just normalize things to overall RP generation doesn't change. This might be a pain to implement, depending on you track your RP, but that's one of the benefits of using a spreadsheet.
Yeah, I don't take messing with the zero-sum system lightly, and it's only a viable solution because I see us starting from scratch when the expansion hits. We can use cheap fixes and band-aids to keep things running until then, as we hammer out all the necessary changes.
If the changes I've outlined above for us work out, I'm hoping to not have to restart our DKP system come BC, as the upgrade path in all slots means that there will be no point saving up for BC items.
I *hate* restarting DKP systems, we did it once in my old EQ guild and it left a huge amount of issues and problems that were still being dealt with ages later.
Ah, but how useful is your upgrade system when the first people in your guild to hit 70 are 10-manning Kharazan and replacing their DKP-bought epics with upgrades that they /random'd on?
TBh, until we see the quality of Karazhan, I'm very doubtful whether NAxxramus gear will be worse by any means... :)
I have to assume that the 40 player dungeons in BC will give us items better than the 10 player stuff, in which case the DKP for the new 40 player items would be based upon what they last paid in the DKP system. In essence, what they get outside the system simply doesn't count.
Right, lv70 40-mans obviously will have much better stuff, and our DKP system, like yours, only looks at what item in a given slot a person has spent points on. If someone has a Jin'do neck but has never spent points on a 40-man healing neck, they have to pay full price for Angelista's Charm, for example. But anyway, as long as people are raiding actively, those situations arise relatively infrequently. 20-man and PvP loot exists to supplement 40-man loot for most raiders, and there's still a general correlation between time spent raiding and gear quality. But what about a new recruit, or someone who wasn't as active pre-expansion, but now has lots of free time and powerlevels to 70 and decks himself out in the best non-40-man expansion gear. He might have the exact same set as someone who once assembled a full tier 2 kit and then banked it for expansion gear, but he would have to pay much more for every item he loots in the expansion because he wouldn't be upgrading, or would be upgrading from lower-caliber items for DKP purposes. That doesn't make sense to me.
Plus, for us, it's an opportunity to fix all sorts of errors and misjudgments we made along the way that would be too labor-intensive to retroactively correct.
Plus, for us, it's an opportunity to fix all sorts of errors and misjudgments we made along the way that would be too labor-intensive to retroactively correct.
Can we expect to see a public description of new and improved "The Elitist Jerks Loot System" ? Frankly, im intrigued to see new verstion of it.
Originally Posted by zeidrich
Women's breasts can be modeled as a cone and measured as V = (Db^2*h*.785)/3 and since breasts can be thought of as an amorphous fluid, you just have to worry about containing the volume of the breast.
Well, our turnover on recruits is pretty low tbh - we're not completely closed like yourself, but we don't recruit often, and those that do come in get showered with MC gear before they know what's hit them... :)
I think it's an issue if Naxxramus gear is lower than Kharazhan basically - which just depends on what Kharazhan is like. However, all new recruits will pay more for thier stuff even under our current ruls, until they have a "base set" of gear to get. Although it might seem unfair, I don't know how fair it would be to put them on the same footing as the raider who spent the past year getting us all equipped in the gear that we will take to our first BC raids? (Because unless Blizzard put a *lot* of stuff in Kharazhan, you wont get a full complete set from there, plus we wont be able to get the entire guild through kharazhan geared up first anyway, so the tier 3 stuff that was earned beforehand will still have great uses).
Re. labour intensive - I know what you mean. I've just gone back and re-done the entire DKP database on a back up copy, to see what it owuld look like under the upgrade system. It helps the situation greatly, as some of our warriros have gained 60+ points, which is a big amount when your system ranges from -120 to 140 maximum.
Basically, I guess we just have to wait and see whether the system needs restarting or not come BC.
Can we expect to see a public description of new and improved "The Elitist Jerks Loot System" ? Frankly, im intrigued to see new verstion of it.
Absolutely. The main goal is to find a good way of standardizing pricing and setting up a system so that every class can expect to pay roughly the same amount for the same increase in power, taking the different slots available to different classes into account.
Hopefully we'll also be throwing together some custom software because EQDKP is a piece of crap. (Well, it's decent for what it is, but it really really really was not designed for a zero-sum system with an upgrade policy.)
05/16/06, 7:09 PM
Similar Threads
