Theorycrafting Crafting - Part 1: Itemization


Staff member
(Note: This is a "What Would Nephele Do" post. Any crazy opinions and insane ideas expressed here are entirely my own and do not necessarily represent the rest of the Pantheon Crafters staff.)

Vjek’s recent constructive criticism of the statements in the developer roundtable got me to thinking – what would I do if it were me designing Pantheon? So, I thought I would put it here and see what people thought of it. At first, I was going to do a single post but realized that would be absolutely huge, so I’m going to break it up a bit. This is Part 1. Expect more to follow over time.

First, some disclaimers:

Disclaimer 1: I don’t pretend to be all that smart. I’ve had the opportunity to see a lot of good and bad in MMOs over the years but that doesn’t mean that what I envision would be perfect or not potentially require tweaking.

Disclaimer 2: I’m a moderate, and an optimist. I still believe that some of the classic tropes of MMORPGs and their economies can work, long-term. I don’t think we have to throw out concepts or ideas just because they’ve been poorly implemented or poorly managed in other games.

Ok, all the disclaimers are out of the way. Here’s what I would design in Pantheon if it were entirely up to me.

Part 1 - Itemization

Before I can talk about anything else, I need to talk about itemization. This is important because every other design decision in the crafting sphere ultimately hinges on how itemization works.

I want to start off by saying that it’s been my experience that level-based systems lead to mudflation no matter what you do. When you give people a ladder to climb, and their gameplay revolves to a large extent around climbing that ladder, it’s just the nature of the beast. When they get to the top, they’re going to eventually want the ladder to grow bigger. And since items in a level-based system are also a component of the ladder, those items will have to continue to grow bigger as well.

With that said, however, I think it’s possible to mitigate mudflation to a very large degree by doing smart itemization. I will try to illustrate what I mean by that:

First things first – rather than having items boost raw power directly, what you want is for more powerful items to increase in terms of the potential benefit. This means that things like damage dealt and damage mitigated are determined primarily by other aspects of your character and that while better equipment helps, each piece is not a significant bump in character power by itself.

I have two examples to illustrate what I mean here:

The Broadsword

In most games, you could expect a broadsword made of bronze to have a damage range of, say, 5-8. A broadsword made of iron might have a damage range of 7-11. Steel, 9-13. Mythril, 11-15. And so on. The problem with doing this is that when you’re scaling combat encounters, you have to start thinking about what equipment the player will have as well as other factors like bonuses from levels, attributes, skills, or etc. And that means that the power of the enemies will have to rise proportionately to the power of the weapons.

Instead, imagine if every broadsword in the game had a base damage range of 6-10. It doesn’t matter what that broadsword is made of, on its own it doesn’t do any more or less damage than any other broadsword. Sounds boring, right? But the damage isn’t the only stat. What if the “quality” of a broadsword (based on materials used in its construction) contributed to a secondary stat. For these examples I’ll call it an “effectiveness” rating

The “effectiveness” rating is essentially a stat that determines how well the broadsword can be used in the hands of a skilled wielder. That bronze broadsword is heavy and off-balance, so maybe it only has an effectiveness rating of 80. That steel broadsword, however, is well-made, so it has an effectiveness rating of 100. The mythril broadsword is lighter and stronger than steel, so it has an effectiveness rating of 110.

Effectiveness makes sense when you think about attack damage being influenced by other properties of the character using the weapon. A warrior might have high strength that gives him a bonus to damage. Or, an experienced swordsman might have specialized training that gives him a bonus to accuracy and a greater chance to land critical hits.

What does effectiveness do? That’s simple. It acts as a multiplier on those bonuses. This means that a specialist warrior with high strength using the mythril broadsword is going to be more effective than if he were using a bronze broadsword – and much more effective than a novice warrior would be. However, that effectiveness boost is a relatively flat curve and does not represent exponential growth in character power. If our skilled warrior couldn’t afford a mythril broadsword, he could still pick up a steel one and do fairly well. Fights might be a little tougher for him, but not to the degree that they would be in most games where that mythril broadsword is effectively tuned for enemies 10 levels higher than the steel broadsword is.

The Shield

For my second example, consider the humble shield. A shield provides a bonus to armor and to the chance to block an enemy attack, mitigating its damage. Pretty straightforward, right? In most games, the amount of those bonuses is determined by the construction of the shield. So, a flimsy wood shield might only grant 3 AC and a 5% block boost, while a steel shield might grant 12 AC and a 12% block boost. Because of this, each “tier” of shields offers a fairly large boost in terms of power from the previous one. This in turn also leads to mudflation, where higher-level monsters end up having to hit much harder in order to justify the differences in equipment tiers.

So, what if all shields had the same AC value and block percentage boost based on their form, rather than their construction? Meaning that a buckler might provide 2 AC and a 10% block boost, while a kite shield provides 4 AC and a 7% block boost (because it is bigger and heavier and harder to move to intercept a blow). These numbers would be the same regardless of whether the shield is made of wood, steel, or adamantite. But like my example with broadswords, shields also have an effectiveness rating that scales based on their construction. And likewise, people using shields have skills and attributes that largely determine the results they get when using a shield (or any other piece of armor). Thus, effectiveness contributes in the same way to the results – as a multiplier on the maximum potential of those skills and attributes.

Why do it this way?

The point of the effectiveness concept is to ensure that while items contribute to character power, they don’t cause character power to grow at a rapid or exponential pace – one that would then precipitate vast amounts of mudflation as the character climbs the level ladder. The other benefit of doing something like this is that items will now push players towards horizontal progression – improving skills, abilities, and the like. If our warrior gets his mythril shield and mythril broadsword but doesn’t have the training to really take advantage of them, they’re not that much more useful than steel versions would be.

I want to be clear here that there are plenty of other ways to achieve similar results mathematically. I just invented a stat called effectiveness to illustrate the concept, but you can do the same thing with attribute bonuses, skill modifiers, or whatever you want. The goal is to keep the power curve fairly flat in order to contribute to mudflation as little as possible, while still providing people reasons to seek out better items than what they currently have.

I also want to be clear that no matter what itemization method you use, in a progression-based game there is always going to be some amount of mudflation. It can’t be avoided, since you are allowing character power to increase as those characters advance, and you want to ensure that they are met with appropriate challenges at every step of the way. This would even be true in a purely skill-based progression system without levels. But, by keeping the numbers smaller and more manageable, and by focusing players more on horizontal rather than vertical progression, you can set the game up to be much more resilient to the impact of things like expansions that allow players to progress further than they originally could.

Finally, I want to recognize that one of the risks of a flat power curve is that item upgrades may not feel as compelling for players (although from experience, many players will obsessively pursue even the smallest bonuses if they think it will give them an advantage). It should be possible to get around this however by giving items unique and useful secondary effects. This could be anything from bonuses against certain types of enemies (everyone knows to use silver against werewolves and cold iron against wraiths), to unique special effects, to unlocking special attacks or abilities. There’s plenty of room for additional things on items in a game so that you don’t always have to increase raw power – it just requires some imagination.

I realize that I didn’t really talk about crafting at all, yet – and that’s on purpose. I strongly believe that a game can have the best crafting system in the world and none of it will matter if a poor itemization strategy leads to massive mudflation and obsolescence over time. This is the first of probably a half dozen posts I’ll make, and the others will talk more about crafting. I just wanted to get this one out of the way first since it’s so critical to everything else.

Comments, ideas, and thoughts are of course welcome.

Theorycrafting Crafting - Post Series Index
Part 1: Itemization
Part 2: General Crafting Concepts
Part 3: The Crafting Process
Last edited:


Staff member
Staff Writer
Well said Neph.

There are some cool sudo-physics mechanics we could work in between iron, steel and mithril (i.e. kinetic energy = ½*mass*velocity^2). Different weighted items could swing at different speeds with center of gravity also effecting the speed and angular momentum. Finally armor penetration could be a function of total energy over the contact surface with a relative hardness between the materials vs the unit area resistance of the armor to see if the attack ignores the armor. A faster weapon could also have a higher chance to hit an unarmored portion of the body all together. (different armor construction types could have different spring constants that decrease a certain amount of effective weapon speed)

While this may sound complicated but it would actually be fairly easy to model up the relationships between damage types then create another table with material properties that can be looked up on the fly.

There are also only so many ways to make an attack based on humanoid anatomy. Each of these attacks would have a different result based on the swing speed, center of gravity, total mass, contact area shape and material hardness. The type of attacks you make would be dependent on the type of weapon you use and its individual balance. For example a spear and a rapier work great in a lunge attack where the center of gravity is directly in line with the contact point and would be very fast in a lunge and best of all the character can put their own weight into the lunge. In a cutting swing a spear is very slow due to how far out the center of gravity is from your hands while the overall weight can be high the contact area of a spear swung sideways is very large so armor would be very effective and it is hard to put much additional weight behind it. A rapier on the other hand is still mostly balance near the hilt and therefore can be swiped very quickly but its mass if fairly low so its ability penetrate armor is almost non existent so if you do not hit an unarmored area no damage will result.

These are just two relatively similar weapons and each group behaves vastly different than others and that is before we even consider things like reach.

The point of this though is to show all the relative categories that itemization could tap into and tweak without falling back directly on +X average damage. Even the concept of Average could be heavily explored if the effective attack dice could be modified as a part of itemization.

Which leads to the final main point. Misses matter! We have gotten spoiled by getting to the “always hit” or the “too hit soft cap” fairly easily making all progression a matter of magnitude rather than increasing consistency.

So to tie it all together:
An iron bastard sword could have a weight of 2.2 kgs and a swing speed 26.5 meters per second and a hardness of 32 rockwell C, the 2.4 kgs Steel sword can by swung at 25.4 (a function of character strength and dexterity relative to the distance to center of gravity) with a Rockwell C of 45 and a mithril bastard sword with a mass of 1.8 kgs, swing speed of 29.3 and a hardness of 50. The types of weapons and attacks will favor different materials based on if thrusting, slashing, hacking or crushing mechanics are being utilized.

Potentially maces and mauls would favor iron over steel or mithril as crushing mechanics will ignore trying to penetrate armor in favor of just overwhelming it. Steel will be better for hacking attacks as it has a good balance of mass, speed and hardness and mithril will favor thrusts and slashing as it is fast enough to hit unarmored areas and strong enough to pierce in a thrust despite the lower overall mass.

Now for crafting progression a poor bastard sword could have a 1d1024+0 chance to hit, a low quality one could be 2d512+10, an average 4d256+20, a high quality 8d128 +30 and a masterwork 16d64+40. Each tier is an improvement in accuracy but not a huge one and much more consistent than the previous tier without the difference between poor and masterwork making poor weapons completely useless.