Itemization

[Quark]

Just saw an awesome post on the Raids and Dungeons forums. Blizzard has hinted (in their reasoning of the +def nerf) at the ways they come up with items, using mod values to determine how much of what can be added to an item. Hyzenthlei worked out the numbers, and shows you Blizzard's values for stats, and how much value blues versus greens can have and such.



Snippit:

Item Creation Gone Wrong - Mislabelling

Let’s use this to examine a blizzard blooper.

Everyone probably knows of the Drape of Benediction (ilvl 67)
http://www.thottbot.com/?i=35923

It is a cape that seems horribly underpowered for where you get it from, Azuregos. Let’s see how it shapes up using these tools.

When you work out the numbers, the actual ItemValue of the Drape is 48.3, but an ilvl 67 Epic cape should chime in at 65.3 so something seems amiss.

Maybe its ilvl was too high? Lets see at what ilvl an epic cape should have an ItemValue of 48.3. This works out to be ilvl 48, seems a little low and random, maybe we are missing something….

Let’s see what the ItemValue of a rare (blue) ilvl 67 cape would be, maybe that will tell us something. This value is 48.9 (with a lower bound of 48.2). The ItemValue of the drape of benediction falls right into this window.


So now we have the mysterious ilvl 67 rare (not epic) cape, drape of benediction. All of you who said it shouldn’t have been an epic are exactly right. It was designed as a rare and flagged as an epic (a mixup like the snowblind shoes that are epic but originally tagged as a rare, though they fixed those).

[oldmandennis]

Freaky good post. That guy must have really driven up Thottbots bandwidth costs getting the data to do this.

[Olon97]

Posts like those are why I still pop into the Blizzard forums from time to time. Very thought provoking.

Now if only thott could put a little "budget summary" below each item for handy reference. :)

[Skandranon]

Hyzenthlei worked out the numbers, and shows you Blizzard's values for stats, and how much value blues versus greens can have and such.
[right][snapback]88717[/snapback][/right]

The real killer in the formula is ^1.5 on every ItemValue. What was Blizzard *thinking*? The explanation for crappy caster itemization is completely there in that one little number. If Blizz can't keep raising the +damage numbers on items, all we get is more useless int and spirit and slightly less useless stamina. How doesn't Blizzard understand that once you're at 3k/6k, +damage is everything?

[Rinnhart]

The real killer in the formula is ^1.5 on every ItemValue.  What was Blizzard *thinking*?  The explanation for crappy caster itemization is completely there in that one little number.  If Blizz can't keep raising the +damage numbers on items, all we get is more useless int and spirit and slightly less useless stamina.  How doesn't Blizzard understand that once you're at 3k/6k, +damage is everything?
[right][snapback]88789[/snapback][/right]

They need new, dropped spells. Flat out.

[mjdoom]

Just saw an awesome post on the Raids and Dungeons forums.  Blizzard has hinted (in their reasoning of the +def nerf) at the ways they come up with items, using mod values to determine how much of what can be added to an item.  Hyzenthlei worked out the numbers, and shows you Blizzard's values for stats, and how much value blues versus greens can have and such.
Snippit:
[right][snapback]88717[/snapback][/right]

Just read through this thread and the most interesting thing to me was the restrictions of the formula on increases in ilvl. If the system stays in place as currently constituted (linear increase in ilvl, exponential increase in value of stats as a component of ilvl) item improvements will grind to a halt as one poster mentioned. Without running any numbers it seems that by tier 3 or 4 the differences in items will be almost entirely insignificant...

This says to me that Bliz has to change their item formula in some way at some point, and probably fairly soon with Silithus on the horizon. If they do this in any logical way it stands to reason that tier 2 may indeed be improved as people are apparently clamoring for (I personally haven't looked too closely at Tier 2 gear yet). These improvements may not come exactly how the player base wants them (with Bliz you'd almost expect it to not be what you want) but any improvements will be welcome. I'm reminded of a voice chat conversation just the other day about Bliz's lack of foresight in previous Diablo games and it seems to have caught them again. The budget system seems to be a good idea but some of the implementation details are seriously flawed as the system scales.

This almost leads me to pity the poor raid dungeon designers. Any of us who have dealt with Blackwing Lair have probably been frustrated in some way (probably an understatement even). Part of this is that the dungeon is designed with some of the harder encounters very early which slow any significant progress. But in light of knowing more about itemization it stands to reason that the guilds that are finally getting some of these encounters down and progressing are actually gaining less power overall with the loot and thus it is not necessarily making those encounters any easier. This is in contrast to earlier dungeons where early attempts required high technical working as a group but as your gear improved you could blow through the encounters easily.

It seems that the current itemization scheme will put a quick halt to this and will also make developers of new dungeons walk a tightrope of giving more challenging encounters while not making them impossible. The current system means that Tier 3 and Tier 4 dungeons would be forced to have similar difficulty to BWL (and not anything more) because gear is simply not improving fast enough to make encounters significantly more difficult without making them impossible. Unless it is Bliz's plan that as each tier progresses you need more and more (and almost to the point of basically ALL) of the loot from the previous tier to even stand a chance at the current tier.

- mjdoom

[Kevin]

Unless it is Bliz's plan that as each tier progresses you need more and more (and almost to the point of basically ALL) of the loot from the previous tier to even stand a chance at the current tier.
[right][snapback]88801[/snapback][/right]

And this is exactly what I think Blizzards plan is. Play ad nauseum to get the gear for the hardcore so that they can move on to the next challenge. At this point casuals will be getting up to get the gear and they'll have a few hardcore sprinkled in with them to help making the encounters more trivial. The hardcore will still attempt the new content while "under geared" and still get the challenges of that but will be able to go back and get the gear they need pretty easily. If itemization is character progression this is about the only way to do it without turning all the content into a joke very rapidly.

[Skandranon]

Just read through this thread and the most interesting thing to me was the restrictions of the formula on increases in ilvl.  If the system stays in place as currently constituted (linear increase in ilvl, exponential increase in value of stats as a component of ilvl) item improvements will grind to a halt as one poster mentioned.  Without running any numbers it seems that by tier 3 or 4 the differences in items will be almost entirely insignificant...
[right][snapback]88801[/snapback][/right]

It isn't quite that bad. By tier 3 you'd expect the items to uniformly be oranges, which have a higher cap on allowable points which is still not set. So there'd be a difference, but the cap would have to be much higher. Furthermore, there would probably be only one tier of orange items and then one tier of red/gold, as any ilvl difference would be tiny compared to the much larger type bonus that Blizzard would have to add.

[MongoJerry]

The real killer in the formula is ^1.5 on every ItemValue.  What was Blizzard *thinking*?  The explanation for crappy caster itemization is completely there in that one little number.  If Blizz can't keep raising the +damage numbers on items, all we get is more useless int and spirit and slightly less useless stamina.  How doesn't Blizzard understand that once you're at 3k/6k, +damage is everything?

I don't quite understand your complaint, considering that most of the high end caster items added over the last six months have involved some sort of +damage or +healing stats on them.

My main itemization worry is that the offense has been growing in power faster than defence. High end caster gear should have more +stamina and even better would be some +defence rating on them to take into account all of the incredibly powerful weapons and +agility/str/crit gear that's being added to the game. And btw, if you're running around at 3k/6k, you'll have a miserable time while PvPing. You'd better get up to 4k life if you don't want to be one-shotted by every rogue or warrior sporting the latest easy to get epic weapons and armor.

[Leshy]

One thing that I noticed about that thread, was the complaint that Tier 2 gear was only very slightly better than Tier 1, at the tradeoff of Blackwing Lair being much more difficult than Molten Core.

It seems to me that this explanation is not correct, because the items are actually staggered across instance tiers: You collect your Tier 0 set in order to do Molten Core. There, you collect your Tier 1 set in order to do Blackwing Lair, where you collect your Tier 2 set to do the Temple of Ahn'Qiraj.

With the exception of Ragnaros, there really aren't any bosses in Molten Core that you are required to have your Tier 1 gear for, for example. Now how hard Nefarian is, is as of yet unknown, but it is not unlikely that he will require better gear than just Tier 1. On the same note, Blizzard has already announced that TAQ is going to be only very slightly more difficult than BWL, but that the end boss will be "*extremely* difficult".

By this logic, the big differences in Tier gear seem adequately explained in relation to the Tier dungeons. From Molten Core to Blackwing Lair is a very large jump, as is the jump from Tier 0 to Tier 1 gear. From Molten Core to the Temple of Ahn'Qiraj is only a minor difficulty increase, as is the upgrade from Tier 1 to Tier 2 gear.

Of course, this is just my own speculation ;)

[Skandranon]

I don't quite understand your complaint, considering that most of the high end caster items added over the last six months have involved some sort of +damage or +healing stats on them.

Yes, they have. The point I was making was: look at the mathematical implications of the formula. After about 50 some damage, the point cost for +1 additional damage exceeds the point cap increase for +1 ilvl. In short, at some point, Blizzard literally cannot put more damage on items.

As I said, orange and red/gold items change this, but the fact is that casters primarily depend on one stat which is exponentially penalized. Raising the orange and red/gold cap helps, but not by much. What this formula reveals is an inevitable growing discrepancy on caster items versus melee items.

To keep caster items competitive, Blizzard will eventually, in tier 3 and 4 gear, have to raise the caps such that numbers of say +65 to +80 damage to all are possible, representing a DPS increase of perhaps 8 over tier 2 gear. However, if they raise the cap that much, equivalent melee items are going to end up hugely powerful.

I present to you the following Tier 3 items, assuming an ilvl formula that caps their points at around 155.

Caster Head Item:
+80 to damage/healing with all spells

Rogue Head Item:
+25 strength
+25 agility
+25 stamina
+30 attack power
Improves your chance to get a critical strike by 1%
Improves your chance to hit by 1%

And of these two items, the Caster Head Item has the higher itemvalue.

For PvE, mages and warlocks are stuck. We don't want more int; all that does is let us cast longer, and unless Blizzard intends to continually scale up encounter length (and let's face it, they won't do this), it doesn't impact our DPS. We don't want more spirit; it does the same bloody thing. We don't really want more stamina; we do, kind of, but only because Blizzard puts in deadlier and deadlier RSTS attacks. What we want is +damage. +spell crit and spell hit help, a little, but that's it.

The system penalizes items that are "long" in any one particular stat. Mages and warlocks have no choice but to go "long" in +damage, no matter how much spell crit and spell hit get added. Rogues, warriors, and hunters, in contrast, benefit from a variety of stats. Rogues get more DPS from +agi +str +AP +crit and +hit. Warriors do as well, though only fury warriors are seriously aided by agi. Hunters can stock up on +agi, +AP, +RAP, +crit, and +hit.

Raise the caps high and you get comparisons like that one; keep the caps low and casters can expect something like +3 damage per piece when jumping up a tier. Casters get screwed either way.

And btw, if you're running around at 3k/6k, you'll have a miserable time while PvPing.  You'd better get up to 4k life if you don't want to be one-shotted by every rogue or warrior sporting the latest easy to get epic weapons and armor.
[right][snapback]88818[/snapback][/right]

Why yes, Mongo, I do PvP every now and then. I know that, and have a PvP set of items for extra hp. However, it's not strictly germane to the discussion. Not everyone views WoW end-game as PvP. While I enjoy it, my focus isn't completely there, nor, I believe, should Blizzard's.

[Warlock]

Of those two items, the Caster helm is worth around +27 dps to a 3.0 second spell talented to 2.5 seconds (more for elemental shaman, less for untalented spells). The Rogue one is worth around +6 dps plus specials (very generously another 6 dps) and the hit/crit/dodge chance. I'd rather have the 27, thanks.

It's also very possible to build caster helms with a wide stat spread - using six entries as that Rogue one does could give something like 1% spell hit, 1% spell crit, +25 Int, +25 Stam, +30 damage/healing and +30 to a random spell school (or something near that, I haven't calculated the cost).

[Skandranon]

Of those two items, the Caster helm is worth around +27 dps to a 3.0 second spell talented to 2.5 seconds (more for elemental shaman, less for untalented spells). The Rogue one is worth around +6 dps plus specials (very generously another 6 dps) and the hit/crit/dodge chance. I'd rather have the 27, thanks.

Keep in mind that rogue damage also scales up by weapon. Rogues aren't supposed to get the same dps off the same equipment as casters; Skyshroud Leggings offers +11 dps and Shadowcraft Pants offers only +2.5, but no one disputes that they belong in the same category. Proportionately, the gap begins to close as +damage is required to stack higher.

It's also very possible to build caster helms with a wide stat spread - using six entries as that Rogue one does could give something like 1% spell hit, 1% spell crit, +25 Int, +25 Stam, +30 damage/healing and +30 to a random spell school (or something near that, I haven't calculated the cost).
[right][snapback]88849[/snapback][/right]

Recognized. However, you only need a few such items for PvE. Once you've gotten around 3k/6k, the int and stam become, as they've started to become on current items, essentially meaningless. 1% spell crit and 1% spell hit spread things out a little, but again, there's only so much spell hit you can use. Rogue items can replace +tohit with dodge or an incremental increase in previous stats; casters have to pile damage.

Splitting the damage between +to all and +to specific works, too. The problem is that you can't reasonably design such equipment and expect that everyone will wear it. There will, for example, be three mage helms: each with +40 some odd to all and +30 to either frost, fire, or arcane. Even assuming the ZG token system is reused and you get to pick, you're essentially barred from respeccing once you've gotten the first item of your set. It's a restriction on your level of flexibility which isn't acceptable, and for that reason I don't expect Blizzard to pursue this path.

[lfd]

you're essentially barred from respeccing once you've gotten the first item of your set.  It's a restriction on your level of flexibility which isn't acceptable, and for that reason I don't expect Blizzard to pursue this path.
[right][snapback]88861[/snapback][/right]

It might be taking it a shade too far, but they could always go with the transformable-items approach here; transform your +fire hat into a +frost, transform that into +arcane, transform that back into the +fire version.

[Skandranon]

It might be taking it a shade too far, but they could always go with the transformable-items approach here; transform your +fire hat into a +frost, transform that into +arcane, transform that back into the +fire version.
[right][snapback]88862[/snapback][/right]

That's a good point, but then you'd never be able to enchant your hat (or you'd be blowing the cost every time you respecced, which, especially with the new ZG enchants, pushes the cost of respeccing up 300+ gold - essentially a ban). Not that I don't see this idea as a workable adaptation of the system, but it's far better to fix the system.

[Quark]

Keep in mind that rogue damage also scales up by weapon.  Rogues aren't supposed to get the same dps off the same equipment as casters; Skyshroud Leggings offers +11 dps and Shadowcraft Pants offers only +2.5, but no one disputes that they belong in the same category.  Proportionately, the gap begins to close as +damage is required to stack higher.

Of course, many Rogues avoid SC for just the reason that it's not adding enough damage compared to other blues. Only SC worth wearing, once you have your choice of blues, is the cap. That and my biggest pet peeve about Dagger Rogues right now - until MC, we're not scaling damage as we go. Barman Shanker is the Rogue weapon, and it's from freaking BRD.

Recognized.  However, you only need a few such items for PvE.  Once you've gotten around 3k/6k, the int and stam become, as they've started to become on current items, essentially meaningless.  1% spell crit and 1% spell hit spread things out a little, but again, there's only so much spell hit you can use.  Rogue items can replace +tohit with dodge or an incremental increase in previous stats; casters have to pile damage.

I don't want dodge, I don't want any more +hit. I want crit, attack power, and hp. That's it. The difference isn't in how many things we want. It's that strength and agility further my goals - crit and ap - while int and spi don't further yours.

[Lissa]

The real killer in the formula is ^1.5 on every ItemValue.  What was Blizzard *thinking*?  The explanation for crappy caster itemization is completely there in that one little number.  If Blizz can't keep raising the +damage numbers on items, all we get is more useless int and spirit and slightly less useless stamina.  How doesn't Blizzard understand that once you're at 3k/6k, +damage is everything?
[right][snapback]88789[/snapback][/right]

Hit the nail on the head.

I don't know how many of you out there have really looked at the Tier 1 sets vs. the Tier 2 sets, but the difference between Felheart and Nemesis is fairly small. Nemesis has about +20 more damage total than Felheart and has only slightly more +Int, +Sta, +Spirit, and various resists. The fact that the difference between the tier 1 set and the tier 2 set of about 6 DPS for long cast spells is horrible.

To further show how bad itemization is for casters, consider this. The typical Blue dagger is around 40 to 41 DPS with a near 60 level requirement. The typical Purple dagger from MC is around 48 - 50, sometimes higher. Now, for a caster, Int improves crit rate at a factor of 1% per 100 Int. For a Warrior, every 2 Str is 1 MAP (Melee Attack Power) and for a Rogue every 1 Agi and 1 Str is 1 MAP. The facts that MAP increases very easily and very quickly for the Melee classes means that that jump of about 8 to 10 DPS on a weapon is closer to a 20 to 30 DPS improvement due to stats.

As such, while non-Caster DPS increases dramtically as they go up each tier, Caster DPS languishes on improvement. Blizzard has to sit down and do something to make it better for Casters. While Casters making up the vast majority of damage in the pre-40 man raid instances, they are dwindling quickly in the the 40 man raid instances (last few MC runs with Avarice has had a Rogue at #1 each time and has had atleast 2 to 3 Rogues in the top 5) due to horrible itemization concerning +damage/+crit.

[Quark]

I don't know how many of you out there have really looked at the Tier 1 sets vs. the Tier 2 sets, but the difference between Felheart and Nemesis is fairly small.  Nemesis has about +20 more damage total than Felheart and has only slightly more +Int, +Sta, +Spirit, and various resists.  The fact that the difference between the tier 1 set and the tier 2 set of about 6 DPS for long cast spells is horrible.

Going to Bloodfang from Nightslayer gets you 28 agility, 35 Str, 2 Stam, 140 AC. It also changes +5% crit, +2% hit to +3% crit, +2% dodge, +1% hit, and immune to disarm. You actually lose +crit and +hit going from Nightslayer to Bloodfang, and that's even counting the agility difference. You have about the same Fire resistance, with a little added Frost/Arcane/Shadow resist. So the pure stats are better, but I'd only take the Disarm immunity over pure crit/hit. Dodge only helps me with circular physical AoEs, and truthfully there's not that many of those. Conical physical AoEs can be avoided, and magical AoEs ignore dodge.

To further show how bad itemization is for casters, consider this.  The typical Blue dagger is around 40 to 41 DPS with a near 60 level requirement.  The typical Purple dagger from MC is around 48 - 50, sometimes higher.

Of course, we don't care about DPS. I just did some of the math (saving for a later post), and for a Rogue with 750 AP (easily makeable), Barman Shanker of 36.5 dps does only 2.5 less total dps Electrified Dagger of 42.5 dps. Electrified would be the best blue dagger in the game, too, if it wasn't for weapon speeds. At 800 AP, Barman has taken the top of DPS again.

Now, for a caster, Int improves crit rate at a factor of 1% per 100 Int.  For a Warrior, every 2 Str is 1 MAP (Melee Attack Power) and for a Rogue every 1 Agi and 1 Str is 1 MAP.

Yes, this is the part that hits hardest for casters.

As such, while non-Caster DPS increases dramtically as they go up each tier, Caster DPS languishes on improvement.  Blizzard has to sit down and do something to make it better for Casters.  While Casters making up the vast majority of damage in the pre-40 man raid instances, they are dwindling quickly in the the 40 man raid instances (last few MC runs with Avarice has had a Rogue at #1 each time and has had atleast 2 to 3 Rogues in the top 5) due to horrible itemization concerning +damage/+crit.
[right][snapback]88888[/snapback][/right]

It's really not dramatic at all unless it's a new weapon. The new weapon is everything to the Rogue (and I haven't seen one in months, with only Electified and Gutgore in sight). The reason we've had a Rogue up top is because Arleas got tired of waiting for Dal'Rend main hand and grabbed a Krol Blade, while Yuen does his best to maximize Barman's damage range with +8 damage stones and +damage enchants. Besides, while I'm not gonna argue that other classes shouldn't be comparable to Rogues, what other class should be #1 in damage? Mages/Hunters? They don't take the risks (and bandage consumption) we take. Warlocks? Same risk as Mages, and they have great utility. Warriors? Put a good Fury warrior together (not half-fury, not part tanking. Pure Fury), they'll already keep with Rogues while having more armor and health. Oh, and they can backup as tanks too.

I do agree that caster Itemization is still the worst side, but I don't think your talking about our damage list furthers your argument.

[Artega]

For a Warrior, every 2 Str is 1 MAP (Melee Attack Power) and for a Rogue every 1 Agi and 1 Str is 1 MAP.[right][snapback]88888[/snapback][/right]

For Warriors, 1 STR is 2 mAP. 14 mAP is +1.0 DPS. Thus, every 7 STR adds 1.0 to our total DPS.

[Olon97]

The facts that MAP increases very easily and very quickly for the Melee classes means that that jump of about 8 to 10 DPS on a weapon is closer to a 20 to 30 DPS improvement due to stats.

That implies that there is some sort of multiplicative effect. Other than the weapon speed (a stat independent of DPS) serving as a multiplier for instant attacks, there isn't a multiplicative effect with weapon stats and attack power dps.

When a warrior has a 40dps 1h, they likely are getting ~60 dps from attack power. So 100dps regular attack damage. getting a 50 dps weapon with no change to attack power takes them to 110 dps regular swing damage, a 10% improvement, not the 25% improvement it would appear to be at face value.