I punched the stats from these boots into this excel sheet that I made to evaluate item budgets using the data from this thread... and they both came out to the exact same budgeting, right around ilvl 219. I use a socket value of 16 for budget purposes, however this causes pretty much every item with sockets to come in grossly "underbudget" so something is amiss with that... Regardless, the two items you linked end up using the same total budget (within 0.1 points) Tempered Mercury Greaves = 218.37 (should be 226) and Greaves of Swift Vengeance = 218.32 (should be 236).
What is going on with this? Are socket bonuses factoring into item budgets more than previously thought?
Also, I need to do some of my own figuring on stuff, but the current school of though on sockets does not seem to be working, at least when I do my calculations. For 226 items with sockets, my calculations has pretty much all of them coming in at 218-222. This is with using 16, as I stated above. I have to use a socket budget of 28-30 to get items that are perfectly budgeted or even slightly over budget.
Any one have any thoughts? I can post my rudimentary excel sheet if that would help.
Two things come to mind as possible causes. The first is stamina is being overvalued. This occurred with a lot of block rating gear in 3.0 where stamina was being counted as 1 ipoint rather than 2/3rds. Now another stat like intellect or something else could have been misvalued similarly. The second is that sockets are being valued as one of the stats already present on the gear most likely according to socket color and as the stat in largest amount so they are being taxed accordingly.
If you read this thread you will see that it's been established that sockets probably count as if the item had 16 more of one of the stats on it (or 24 stamina). Which stat is chosen and if this is the actual method at all is hard to impossible to figure out/prove since items are not always exactly on budget, but that's the general guideline.
Also, if you are going to claim that an item is underbudgeted, it helps if you include your calculations. This makes it a lot easier for people to detect any errors in the calculation and verify the results.
I might start doing some true crunching of numbers because, frankly, it seems like a whole slew of Ulduar epics are under budget.
I think it's important when making these calculations to keep in mind that ilevels are guidelines, and not hard rules.
It's likely that Blizzard does not start with a ilevel budget and dole it out to stats from there, but rather they just assign stats as they go and simply check back on the ilevel once they're done to see if they missed the mark.
Items that will miss their ilevel budgets by 1~2 points will probably always be the rule, rather than the exception.
I find it hard to believe Blizzard just "assigns" some arbitrary stats and then backtracks to see if the item level matches what they intended. The numbers are too random for that. If they would assign stats, you'd see a lot more round numbers instead of all these 33 spirit, 31 critrating or 103 spellpower.
One could argue that the numbers are so weird *because* they assign them manually and then adjust them to make the item match the intended item level. But if that was the case, alls items should be exactly on budget, which is clearly not the case now. Unless of course the ilvl formula being used is not correct or blizzard is intentionally putting in items over or under budget.
So, keep in mind, this is done by humans. They are making an item and it gets tagged "iLvl 213" (or insert appropriate value) in a static database field. That's not the calculated value, it's the "don't forget this is the right value" field that you can datamine. Then there are the actual stat values, that, obviously, many have attempted to reverse engineer the formula for and clearly have come at worst very close.
Now, if the "item creation tool' is separate from the database, it'd be very easy to build an item there, have it check out at the correct iLvl and then get re-entered in the database wrong. It'd be easy for it to get created slightly off from optimal because someone decided to check with someone else before allocating the last stat points and forgot to get around to it, etc. etc. Finally, it's very very possible an item is allowed to be with +/-1 stat of correctly valued. In that case a 199.34 would pass and a 211.5 would pass if the checker was rounding first (a bad choice, perhaps, but Blizzard uses a LOT of rounding). It doesn't explain the 215 getting through, true... Read on.
I mean this is being done by humans.
Now, if the database would flat out not allow something tagged iLvl 219 to only have 215, what if the person entered 216 in the iLvl static value field? Well, then a 215 would indeed pass through.
I'm not saying the formula is right or wrong, but it's not hard to imagine how a few things could slip through the cracks, especially when there are ~400 in Ulduar alone this patch and several hundred more on the badge vendors, in the Argent event, etc.
I'm not sure if this is specifically an iLvL mixup, but are some of the new blacksmithing plans from Ulduar underbudgeted / overbudgeted? As one of their Wowhead comments suggests, has it been confirmed that something is wrong with these items?
Both sets of items actually appear to be slightly overbudget coming in at 229 epics rather than 226 epics. To get the item level of the DPS boots, I just gave it an extra 16 Strength and 24 Stamina, which mirrors a similar stat distribution to the tanking boots if you give it an extra 48 Stamina. Doing the same thing to the belts yields similar results. When it comes to white stats, all four items are essentially the same and the green stats wind up equalizing themselves. These items just seem to lend more credence to the fact that the color of the socket doesn't necessarily correspond to any particular type of stat, but rather it's 16 points subtracted from one stat and then turned into a socket of any color.
The problem seems to just be an intentional design of making plate DPS gear offset with more stamina rather than being purely about offense. I'd have to dig into some old screenshots during the 3.1 PTR, but I remember seeing a lot of the plate DPS gear having more stamina compared to leather or mail items.
I think most of us originally thought the socket bonus was too high and would be corrected to +16 AP in the future. However, blizzard instead went the opposite route, leaving the socket bonus intact but removing a small portion of another stat to keep it on budget.
The interesting thing about this is that socket bonuses appear to be part of an item's budget, instead of a free bonus for matching all socket colors on an item.
I think most of us originally thought the socket bonus was too high and would be corrected to +16 AP in the future. However, blizzard instead went the opposite route, leaving the socket bonus intact but removing a small portion of another stat to keep it on budget.
The interesting thing about this is that socket bonuses appear to be part of an item's budget, instead of a free bonus for matching all socket colors on an item.
I would not take this as full proof of that, changing socket bonuses on existing items causes the already existing socket bonus to stick on the item regardless of it's gemming, while also potentially giving someone the new socket bonus at the same time if the gems continue to match; I used to have an old [Justicar Faceguard] which had both it's old socket bonus (+4 block rating) and it's new one (+4 dodge rating) at the same time as it was changed while I had it in my posession.
So changing the socket bonus itself on the Guise to 16 attack power would actually result in people who already have one getting 48 AP total from the socket bonuses, while people who don't already have one would only get 16 AP. Removing 16 AP from the item is, in other words, the only clean solution Blizzard has.
buff /bʌf/ Pronunciation[buhf]
–verb (used with object)
- to reduce or deaden the force of
I would not take this as full proof of that, changing socket bonuses on existing items causes the already existing socket bonus to stick on the item regardless of it's gemming, while also potentially giving someone the new socket bonus at the same time if the gems continue to match; I used to have an old [Justicar Faceguard] which had both it's old socket bonus (+4 block rating) and it's new one (+4 dodge rating) at the same time as it was changed while I had it in my posession.
So changing the socket bonus itself on the Guise to 16 attack power would actually result in people who already have one getting 48 AP total from the socket bonuses, while people who don't already have one would only get 16 AP. Removing 16 AP from the item is, in other words, the only clean solution Blizzard has.
You didn't actually get two socket bonuses. It was just an UI issue. But maybe they're trying to avoid similiar isues.
You didn't actually get two socket bonuses. It was just an UI issue. But maybe they're trying to avoid similiar isues.
I was actually getting two socket bonuses, it was not an UI issue. I compared my stats prior to equipping it and after equipping it, and I always gained 4 block rating which I should not have had.
buff /bʌf/ Pronunciation[buhf]
–verb (used with object)
- to reduce or deaden the force of
According to patch notes, and the last time I remember it happening, 2.1 was the last time that Blizzard changed a socket bonus. Chicken's right, the reason socket bonuses aren't changed is because you'll get the benefit of the old socket bonus as well as the benefit of the new socket bonus. This blue post, while it doesn't go into the underlying reasons why socket bonuses aren't changed, confirms that Blizzard does not change socket bonuses anymore.
As far as i know they are the only epics without other epics at their item level.
To me, that indicates that the item level for these items fits their budget exactly: ie, if the formula we have is correct, we should get 143 & 206 as the respective iLevels.
Also : [Dark Herring] is the only iLevel 187 green weapon in existence.
Can anyone else come up with more 'odd' items (not under-budget, just unique)
I've got [Arcanum of the Earthen Ring] and [Arcanum of the Earthen Ring] coming in at 139 and 200 and [Dark Herring] coming in at 190. It's interesting that they chose non-standard item levels for the rings because item level only affects stat distribution when it comes to rings. On the fish, I may need to work on the Uncommon formula a bit more, but I haven't been too concerned with it. With the rings, I think it's just that the item designers were having some fun and, since they weren't constrained, they just put in any number that was higher than what they needed.
On a different note, I've been working through armor calculations and I've run into something interesting. In TBC, epic armor increased almost linearly based on item level. This was true for cloth, leather, mail and plate. In Wrath, it gets a little bit tricky. There's no longer a linear increase for armor, instead it's different. I haven't found an equation yet that matches the data points I have. An interesting thing about this new armor formula is that as item level increases, the amount of armor gained per point of item level also increases. During TBC, each point of increase of item level added approximately 12 armor to a plate chest piece. In Wrath, going from an item level 200 to an item level 213 plate chest piece only added approximately 4 armor per point of item level, yet going from an item level 226 to an item level 239 plate chest piece adds approximately 5.5 armor per point of item level.
Cloaks and cloth chest pieces follow the same kind of increase, which made me reach for my favorite cloak, the datamined 252 cloak from beta. Armor seems to increase uniformly across all types of gear, just to different scales. Using this fact, I can estimate the expected armor of a 252 plate chest piece. To get plate armor from an equal item level cloak, the formula is approximately
plateChest = 15.02 * cloak - 12.544
So, given that a 252 cloak has 170 armor, a 252 plate chest piece might have about 2541 armor. It looks like Blizzard's trying to prevent players from reaching the damage reduction cap too quickly, but being able to ramp up armor as the player progresses through content.
A recent in-game change was reducing the amount of bonus armor on [The Leviathan's Coil] from 882 to 448.
If we assign a StatMod value of 0.07 for bonus armor:
The old 882 armor Leviathan's Coil has an ilevel of 237
The new 448 armor Leviathan's Coil has an ilevel of 224
Since the item itself is supposed to be ilevel 226, the 0.07 statmod seems to be accurate. I just wanted to mention this since there's no value for bonus armor in the OP.
I'm not sure this is the right place to ask, so excuse me if it's not.
For a long time warriors have been using leather dps gear because those pieces can have 4 dps stats. What is the reason that blizz can't put 4 dps stats on plate? Is it really easier to nerf and modify talents just for this purpose every few months?
Would this make plate dps too good?
They *can* put 4 DPS stats on Plate, they just don't want to, ostensibly because AP would feel redundant against STR, and they deliberately moved away from AGI going into TBC because it's redundant with critical strike rating and (probably) could produce issues with high dodge rates in PvP.
Would it be easier to make plate with 4 DPS stats instead of mucking around with talents? Possibly. If we assume Blizzard sets aside what would "feel awkward", then putting 4 DPS stats on plate would go a long way towards making them desirable, since the STR-modifying talents like Improved Berserker Stance only delay the point where leather gets better again without really addressing the problem in long-long term.
Would it make Plate DPS too good? Again, it depends. If you put AGI as the 4th DPS stat, you're not likely to excite many Warriors since their AGI to crit conversation rate is so low as a symptom of Blizzard trying to move away from that stat (although Paladins might like it). If you put something like AP as the 4th DPS stat, you're definitely making things more interesting as far as upgrade-potential goes, but not very interesting from a mechanics POV.
Functionally-speaking, adding AP as a 4th DPS stat would be no different from adding a second amount of STR that's treated as a separate stat on the itemization DR formula.
"I don't like regen stats."
The solution here, from our POV, is not to take the regen stats off of gear, but to get you to care about them. To us, that's the same argument as saying "I can get all of the hit I need through talents, so hit on gear is a waste for me. Please remove it."
"I'd rather wear leather than dps plate."
I addressed this recently. The root of the problem is that putting AP + Strength on plate feel dumb (because except for block, it's the same thing). Putting Agi + Strength on plate feels useless. A long-term solution would probably be something like letting rogues (and similar classes) derive AP from 2xAgility and just put Agi on leather and Strength on plate.
Is it me or the newly discovered mail helm from 25-man Algalon lacks about 100 Attack Power for its item level?
I've tried doing some calculations on it myself and it seems to peg it as ilvl 217, but it has a blue and meta socket. A statValue of 16 does not seem to sufficiently make up for this, so I'm not sure if it's an issue with the current knowledge of item level calculations. Them item does however significantly deviate from the usual Blizzard design philosophy on items, which seems that either an item focuses more on Agi or AP. This one seems to do neither and comes out with less on both counts compared to the T8.25 helmet. Comparing the raw statValues on the item to the T8.25 helmet, it gains only 3.66~ or so statValue and even given the nature of the ilvl formula it seems that this is indeed too low for an ilvl 239 helmet.
Has the effort of determining the ilvl calculations been dropped? It seems to have gone into a bit of a slump at least, looking at this thread.
I looked into the Algalon mail helm as well and came up with an ilevel of 219.37 - disregarding the meta socket but factoring in the colored socket via the "add-a-stat-matching-the-socket-color" method.
If the Algalon helm is 219.37 but is intended to be 239, then it's underbudget by 19.62 ilevels. Sure, we don't know how much a meta socket is worth, but if we look at other helms, we should get a pattern of how underbudget they all are, right?
Conqueror's Scourgestalker Headpiece: actual ilevel 222.79, 3.20 ilevels underbudget
Conqueror's Worldbreaker Faceguard: actual ilevel 220.25, 5.74 ilevels underbudget
Helm of the Furnace Master: actual ilevel 220.22, 5.77 ilevels underbudget
Blue Aspect Helm from Heroic Eye of Eternity: actual ilevel 216.13, 9.86 ilevels underbudget
All of these 4 helms that are intended to be ilevel 226 are all underbudget by 3-9 ilevels, much larger than the 19 of the Algalon helm, which means it's definitely lacking some stats even if we're missing the meta socket value.
Are you trying to suggest these items might be usable and still get the He Feeds on Your Tears achievement if one watches how it is gemmed, or in this case not gemmed?
No, item level is a set value and is defined on the item. What he is talking about is the theoretical item level, that is, what item level should be assigned based on the stats. In reality, stats are based off item level and not vise versa, so this method is just used to see which items are overbudget, and therefore want to be used, and which ones are underbudget, which want to be petitioned to be fixed. It has no bearing on the reality of the item level (which in this case is 239).
04/28/09, 10:33 AM
Similar Threads
