The Blizzcon announcement that they’d removed player-customized attributes, and that D3 items would not have attribute requirements to equip, caused a huge stir of controversy. Blizzard said it made the game easier to balance and kept noobs from wrecking their characters. Angry players lamented the loss of character customization and the “training wheels” approach to D3. Approving players said they’d never liked attributes in the first place and pointed out that everyone on Battle.net has the same build anyway. There were more arguments than just those, of course, and they’re all summarized in this installment of On the Drawing Board.
The article also features a thick slab of discussion about larger game issues: what effect does the removal of item attribute requirements have on overall game design? How will this affect HC characters? What other attribute-tweaking game mechanics and equipment options (such as the Talisman) do we not know about yet? There’s a lot more, so read on for all the details, and join the discussion afterwards.
On the Drawing Board #3: Preset Attributes
One of the most controversial features yet announced for Diablo III is a big change from how characters were created in the earlier games in the series. The character attributes, which are strength, dexterity, vitality, and willpower in Diablo III, will be automatically placed when your character levels up. Attribute points still exist, and the values are displayed in the character window, but you don’t set them yourself. When your character levels up; they are automatically allocated in a mixture the D3 Team has devised. The characters at Blizzcon could only level up to level 9 or 10, so we have no idea how the points will be distributed long term, but from what we saw while playing, they were fairly-evenly distributed. A level 10 Barbarian didn’t have 80 strength and 5 willpower, for instance. It looked like each attribute gained 1 or 2 per level.
Player reaction to the auto-attributes announcement has been mixed, to put it kindly. This change was first revealed in a news item we posted from Blizzcon, immediately after hearing this news from Jay Wilson, and it spurred massive discussion there, and especially in a now 29-page thread in our main D3 forum. Much of the early discussion centered on an error; we’d reported that there would still be attribute requirements for equipment, which turned out not to be true. Jay hadn’t said there would be, but he hadn’t said there wouldn’t be either, so we just assumed that basic game feature would return from D1 and D2. The fact that there will not be any attribute requirements for items in D3 defuses a lot of the “But what if my Wizard wants to wear heavy plate?” complaints, but there were other arguments than that.
Incidentally, there will be some item requirements, they’ll just be based on Character Level, possible other unknown factors, and sometimes on class. We saw a few class-specific items in the Blizzcon build (axes that only Barbarians could use) but on the whole, characters will be able to wear or use the vast majority of items they find in Diablo III, if they so desire.
The fact that characters will be able to use most items they find is a great concession to Diablo’s gameplay style, and a sign that not everything in D3 is influenced by WoW’s design, but it’s only a small part of the larger issue. What does the lack of assignable attributes change? What are the pros and cons behind this decision, and what are some of the larger gameplay issues that have to be considered? This article will attempt to sum up the points on both sides of the issue, and bring some clarity to a murky topic.
Pros
It’s noob-friendly, and attributes were largely irrelevant in D2 anyway. Jay Wilson stressed this in our interview, and Bashiok reinforced it in a forum post:
For the most part attribute spending in Diablo II was a great way, when you didn’t know how to play the game, to break your character. Most people didn’t know where to put them and when they found out the answer was always kind of weird like “Put 5 points in Energy and then all the rest of the points in Vitality.”—Jay Wilson, video interview
To expand on it a bit more when you don’t know what you’re doing you’re essentially lost, and you sort of spend points how you think you might want to. When you finally have an idea of how a character should be built, stat distribution generally comes in the form of “this is exactly what you need for x build” and there’s little variation. At that point it’s pretty easy to remove that system and instead offload the potential build ‘requirements’ to something more interesting and something that’s actually more engaging and fun.—Bashiok
Part two of this point is mentioned by Jay above, and it’s that attribute points weren’t much of a source of character differentiation anyway. Most builds in D2X have very specific attribute requirements, which generally break down to, “Enough strength for your heaviest armor or weapon, enough dexterity for max blocking, and all the rest in vitality.” This varies somewhat between classes and builds, but with experienced players planning out most of their equipment in advance, and most characters using a lot of the same equipment, the end result is that the vast majority of D2 characters of the same type have about the same attributes. That’s exactly the D3 Team’s point: what’s the use allowing players to assign their own attribute points if 1) it leads to broken characters, and 2) all the well-built chars end up with the same values?\
It enables the D3 team to better balance the game. Another benefit of setting the attributes is that the team can balance items and skills and monster difficulty and quest rewards more precisely. With pre-set attributes they have a rough idea of how many hit points, how much mana, how much stamina, how much damage, etc, to expect from a character at level X. (And they know how level X correlates with dungeon Y from play testing.) This isn’t something of much concern or notice for players (unless/until you get way off the usual progression path and find things pointlessly easy or impossibly hard.) but it’s a major concern for the design team. Quests in D3 reward the player with magical and rare items of pre-set types (one early quest reward in the Blizzcon build was always a pair of boots) and that’s a design element that helps the team balance the game, since it gives them a rough idea of how powerful (or weak) characters will be as they progress. Knowing the character’s attributes serves the same function.
The other side of that coin is that giving items requirements other than attributes prevents expert (or lucky/accidental) players from exploiting the game with twinks, or overpowered items that are inappropriate for the low level characters who are using them. In D2 items had Clvl requirements, but those were often rather low, especially before the end game. This resulted in Barbarians and Paladins putting their first 8 levels into strength and getting into item sets and uniques (or just plain magical or rare items) that were clearly designed to be used by much higher level characters. Anyone who wore a full Sigon’s Set at level 8 could simply ignore the monsters until about Act 4. (This sort of thing could be prevented by tighter Clvl requirements as well, of course.)
It works in World of Warcraft. Whether this point relieves and infuriates depends on how you feel about Blizzard’s other RPG. The fact remains that attribute points are automatically allocated in the most popular MMORPG ever, and no one seems to mind. The games aren’t exactly comparable, but lots of RPGs have preset attributes, and it doesn’t break the games. Their design teams simply need to put in enough other ways for characters to be customized and varied, and the D3 Team is confident they can meet that challenge. (More on this issue in the discussion of larger game features after the “cons.”
Attribute respecs. Even if there were player set attributes in D3, any customization from them would be largely undone by a respecing feature. The D3 Team has committed to providing some sort of skill respec system, and it’s a sure bet that if they’re doing that, they would have done something similar for attributes, since all the same arguments apply. They don’t want to force players to reroll if they make some point-allocation errors and “break” their characters, so there would have to have been some sort of attribute respecing. And you know if they included that feature, every player would have taken advantage of it to completely cookie-cutter their character for the end game. (What will keep that from happening with skill respecs remains to be seen, but that’s an issue for another column.)
Also, attribute respecing would have encouraged very unbalanced, exploitive game play that would have had to be balanced against by some major changes to the monsters or other factors. For example, if every player could put their first 50 points into strength to boost their damage, knowing that they could just reassign those points later to other more useful long term attributes, every player would. (Which would force the design team to boost the monsters in some way to counteract it, which would doom players who didn’t go for that build style.) Not allowing that sort of thing is another point in the design choices that let the D3 Team better balance the game, since they can count on characters being somewhere in the Goldilocks range, at least in the early-mid game.
It boosts the importance of items. The D3 Team feels that items are a core mechanic of the game, and that features that emphasize the importance of items are good features. They’re adding a lot of new types of items, and adding much more variety to the mods on known item types. If modifiers provide bonuses to attributes far above, or far differently than they did in D2, it could be possible to customize a character nearly as much as we did in D2, simply from equipment.
Late Game Equipment Finds. In a point related to stat-respecing, the preset attributes make it possible not to have attribute requirements on items, which removes the problem of high level characters finding a great item they can’t use. It’s frustrating to many players to build a character with 120 strength, get to a high level, and then find an awesome weapon that requires 150. It might be possible to juggle other equipment, or change around charms, or socket something, but the best solution in D2 was almost always to reroll; to build that same character type from scratch, with 30 more strength (or whatever attribute the item required). The D3 Team doesn’t want players to have to make a new character to use a new item. (They are considering other ideas to keep players making new characters, so that over time the early and mid-game areas don’t entirely hollow out, as often happens in RPGs.)
Cons
Reduced Character Variety. This is the most common argument against preset attributes; that it forces players to play in pre-determined ways and eliminates the possibility of indulging in odd, amusing, or truly wacky builds. What about melee mages and singing Barbarians? What about Titans, pure Dexazons, and all theother interesting builds inventive players have discovered? If a player wants to make a Sorceress with 2000 mana and 200 hit points, why should the game not permit it? Is there a law that says every character must be a square peg in a square hole, and equally strong and functional? This issue might be addressed in some of the extra game info, below.
We won’t be noobs forever. On a related issue, is it wise to base so much of the game design on a feature that’s largely designed to function as training wheels for new players? We’ll live and learn, and besides, do players really expect their first character in an RPG to be their best, or their permanent/forever character? Everyone reading this made a bunch of D2 characters in the early days that we loved, even though we can look back and know they sucked. We’ve all got fond memories of struggling through the game with a Bowazon wearing Frostburn with 100 points in energy since we didn’t know about mana leech. Or a Paladin with 200 strength and 50 dexterity that couldn’t hit a quill rat in Nightmare. They were fun at the time, and we learned from our mistakes, and we enjoyed it more when we built out next, better-designed character.
It was impossible to build a bad character in Diablo I, since the attributes all capped at set levels, and you could only increase your spell levels over time. (You could ruin your character on bad shrines, but that’s another issue.) D2 added great variety over D1’s cookie cutterism, but that variety and customization came with a risk. One of our forum mods hit this nail on the head in post #186 in the mega-thread.
There seem to be a lot of people willing to accept hand-holding just to “make things easier for newcomers”. What I don’t understand is why people these days are so opposed to learning and failing. Why should you be able to come to a game you never played before and immediately create a powerful and successful character? What exactly is wrong with screwing up a char or three? I’ve had a lot of fun in D2 finding out what worked and what didn’t. The reward when you find out what works for you is fantastic. But no one seems interested in it these days.—Mad Mantis
People are still thronging Battle.net more than eight years after it was released, and it was Blizzard’s best selling game until World of Warcraft unseated it. Clearly the risk of making a sub-optimal character through a few misspent attribute points didn’t turn off too many players.
No Attributes = No Skill. In a related issue, skillful D2 players could choose to specialize in order to maximize their returns. If you were good enough not to die, or didn’t care if you did, you could ignore vitality and put every point into dexterity and strength, and see your power increase commensurately. A low level Sorceress could load up energy early on to boost her mana pool and killing speed. Etc. These early game strategies had to be balanced with long term build goals, but if a player knew they wanted 150 strength and 90 dexterity at level 80, it was up to them to decide when they’d put in those points.
Playing the D3 demo, even at the low levels the characters were (they started at 5 and could be leveled to 9 or 10 before the dungeons were entirely cleared), I wanted to customize them. My Wizards were constantly out of mana and low on life, while the dexterity and strength did nothing for me. I didn’t need any vitality on the Witch Doctor since my Mongrels were tanking so well I hardly got hit. My Barbarian was a killing machine, but never missed a swing; I wanted more strength to deal even more damage.
It’s quite possible that all three of these examples are actually evidence to support Jay Wilson’s “noobs will build bad chars if you let them” thesis, and it’s impossible to say how I would have missed those other attributes if the game hadn’t been spending them for me, but I missed not having the option to try, or to see what each attribute actually did for my characters.
Rerolling is fun. This one is speculative, since we don’t know how long the character development process will be in D3. Having to reroll in D2 wasn’t a big deal, since a few hours a day, a few days a week, was plenty to zoom a character up into the high levels in short order. Reaching a really high level can be quite a grind, especially since v1.10+, but most D2 characters are into their end game gear and using their skills at nearly their maximum power by level 70 or 75. WoW characters take much longer to level up (barring insanely long hours or paying leveling services), so of course respecing and a game design that keeps them from being built poorly is essential. (Plus, players feel more entitled to be catered to when they’re paying $15 a month for the privilege.)
What about hardcore? The D3 team has a point that most characters end up with very similar stat allocations, but what about Hardcore (HC) characters? These days, after so many patches full of mudflation, years of accumulating charms and other loot, and so many rounds of rune duping, the differences between high level HC and SC character aren’t that large. But in the early days, playing HC was a very different experience. Characters were built with substantially more vitality, often to the point that they valued items very differently. I remember socketing a resistance jewel into a Doombringer> on an early D2X MF barbarian, since I did enough damage to take out Meph without adding more % damage, I had leech covered elsewhere, I was low on fire and lightning res, and using a champion sword worth something like 350 hit points since I could pump vitality instead of meeting the higher strength and dexterity requirements for a colossus.
Needless to say, that’s not the sort of choice a non-HC character is likely to make, or even consider. And that’s the point. HC players have to build their characters very differently, and that will be especially true in D3 if Blizzard is able to stop duping and hacking, and doesn’t unbalance the economy with vastly overpowered runewords in patches. But in D3 it doesn’t look like we’ll have that option.
Other Game Issues
With the pros and cons stated, here are some other points to keep in mind. These may undermine or reinforce the earlier points, on both sides of the issue.
- Attributes are supposed to be much more useful in D3 than they were in D2. All of the specifics aren’t yet known, but the D3 Team has said that they want all of the attributes to be valuable to all of the characters. For instance, strength is going to boost melee and ranged physical damage, as well as increasing the defense based on your character’s armor %, and it will help with blocking as well. Willpower will give bonus damage to spells, and will also increase the amount of healing from health globes. Dexterity and Vitality have a variety of bonuses as well, which you can read more about on the attributes page in DiabloWiki. This doesn’t argue for or against preset attributes, but it makes it seem like massive attribute specialization will be less viable than it was in D2. At the same time, it indicates that bonuses to attributes will be very valuable, since they provide so many rewards. This will make +attribute equipment far more valuable than it was in D2, where you could just make up for such bonuses by gaining a level or two, or switching around your charms. (There will not be any charms in D3.)
- If there were customizable attributes, there would have to be some method to respec them, and that would add needless complications and allow various unbalancing and exploitive play styles. This issue was covered earlier, in the pros section.
- We don’t know how long it will take to level up characters. This knowledge will weigh in, but we don’t yet know on which side of the argument.
- We don’t know anything about item requirements, yet. We know they won’t have attribute requirements, but how their Clvl requirements will be designed we do not know, nor do we know what other requirements they might have. The D3 Team has said that there will not be too many class-specific items, but we don’t know how many, or if the class-specific stuff will be the highest item types, if Uniques will be included, etc.
- Item tradeoffs. We don’t know how this will be handled, without strength reqs. There will obviously be something that makes heavy plate mail more useful for Barbarians, and makes mage characters think twice about wearing it. We don’t know what, though. Running/casting speed penalties, armor useless without strength to modify the % defense, fashion sense, etc. The D3 Team is obviously going to find ways to balance equipment between the classes, without attribute requirements to rely on.
- There will be additional items and bonuses and game mechanics we still don’t know about.
Bashiok: Definitely agree. I’m going to give you a bit of a cop-out answer, but we have quite a few game systems we haven’t even talked about. Those aside, I would argue that the rune system - something we have announced - adds quite a bit of customization, and in a more interesting way than attribute distribution.
The talisman, for instance, was going to be an “attribute modifier that used gems,” according to Jay Wilson. It was not working in the Blizzcon build, and might not make the final game, but it was part of the interface, like an inventory for gems. We have no idea how powerful it, or other things like it, might be. There were 9 spaces for gems on the talisman. If top level gems in D3 add 10 or 15 to an attribute, or 10 each to multiple attributes, then players would be able to use gems (or other such items) to greatly modify their attributes. Perhaps enough to make a Wizard viable at melee, or to give a Barbarian a huge boost in willpower, thus enabling variant builds even without player-assigned attributes.
All that stated, the issue remains open to debate. The D3 Team didn’t make this change capriciously or without giving it careful consideration. They played D2, they examined how attributes were used, and they decided that D3 would not benefit from a repeat performance. They’re implementing a variety of game features to replace or improve upon the customization of characters that attributes allowed. Their efforts may pay off, or we may see mod-makers furiously hacking away at the code so they can be the first to create a “real” sequel to Diablo II. Only time will tell.
While we’re waiting to find out, feel free to throw in your opinion on the issue. Jay Wilson has said this is a permanent change, one they’re set on and aren’t going to change, but if you’re unconvinced and want to complain now, you’ll have more right to put “I told you so” in your forum signature later.
Questions is: Considering the above, what are your thoughts on Preset Attributes?
On the Drawing Board is written by Flux. These articles examine crucial game design issues and decisions in Diablo 3 by explaining the issue and presenting arguments for and against. On the Drawing Board aims to spur debate and further the conversation, rather than converting readers to one side or the other. Conversation and disagreement is encouraged. Have your say in the comments, or contact the author directly. Suggestions for future column topics are welcomed.
