by Saturday-Saint Thu Apr 25, 2013 8:20 amI was chatting with some gentlemen on another forum about high and lofty game DESIGN ideas. Specifically the topic was behavioral differences vs. statistical differences. Behavioral differences would be something like, "Zweihander has a different moveset than Longsword." Statistical differences would be something like, "Zweihander does a bajillion more damage than Longsword." There are points at which the line blurs (Claymore attacks recover faster than MSGS attacks. Is that a behavioral or statistical difference?) but overall it helped me organize some of my thoughts with regards to how builds work in Dark Souls. I don't think it gave me any great insights, but it makes it easier to communicate my ideas about them. To celebrate that, I'm going to share my Expert Opinions on Buildcraft with you. While this article is aimed towards people who are interested in making powerful min-maxed builds, any player should be able to find useful information in it.
"There are cards that are bad, but seem good. There are cards that are good, but seem bad. There are cards that are strictly better than other cards in most situations, but far worse in very specific situations. There are cards that are incredibly good, but no good deck can be built around them. There are cards that are bad, but are used in the very best deck because the rest of the deck is utterly dominating and desperately needs a certain ability, even if the best available source of that ability is on an otherwise terrible card.
Perhaps the hardest lesson to learn, the lesson that every Magic player seems to have to learn over and over, is that there are great cards that come together to make great decks that ultimately do not win because they aren’t as good as something else that’s out there."
-"What Makes the Best Player?" from Playing to Win by David Sirlin. Talking about Magic: The Gathering. Relevant to Dark Souls Buildcraft. Also I wanted an excuse to plug this. You should read the whole thing. It's not very long and it's all available online, just ask Google. Worth reading if you're into any PvP game (and I assume you are if you're reading this). Just make sure you read my far more important Expert Opinions on Buildcraft first.
Expert Opinion of High Ledgen Saturday-Saint #1: The Axioms of Buildcraft.
Axiom 1. A build is composed of equipment, spell selection, and a character's statistics.
Axiom 2. A build's behavior is primarily determined by its equipment and spell selection.
Axiom 3. A build's statistics should be designed to support its behavior.
Perhaps I did not thoroughly explain the difference between behavior and statistics earlier. I will broadly define behavior as, 'how something acts,' and statistics as, 'the degree to which something acts.' A Zweihander's swing is fundamentally different than a Longsword's in many ways. Its swing arc covers a different area and it is much slower. On the other hand, its damage is simply higher than the Longsword's. The swings act differently. When they connect, however, their function is the same: they deal damage. Only the degree to which they deal damage is different. This is what I mean by behaviors and statistics.
The First Axiom merely points to a build's components. Equipment includes consumables and anything that may be hardswapped onto a build in the middle of combat. Worth noting is that if you are not going to hardswap equipment in the middle of combat, merely changing your equipment or spell selection constitutes a change in build according to this definition.
The Second Axiom points to the importance of equipment and spell selection. More than your character's statistics, this determines your build's strengths and weaknesses, and its behavior. A Strength build wielding a Demon Great Machete is more similar to a Quality build using a Zweihander than it is to a fellow Strength build wielding a Great Club. While the two Strength builds would be statistically more similar, their behavior is quite different, as they are wielding different weapons. The Zweihander-wielding Quality build, while being statistically dissimilar, is behaviorally nearly identical to the DGM-wielding Strength build.
The Third Axiom points to the purpose of your character's statistics (Vitality, Attunement, Endurance, Strength, Dexterity, Intelligence, and Faith). They are a supportive structure for your spell and equipment selection, which as the First Axiom states, are the primary determiners of a build's behavior. Your character's statistics should be chosen to support the behavior determined by your equipment and spell selection.
I would like to point out that these axioms are not designed to have any ontological value. Rather, they are designed to help you think about builds and how they are designed. I would also like to point out that 'statistics' can refer to the statistics of anything, including your character, your weapon, armor, et cetera. A build's statistics are determined by the statistics of all of its components. When I want to refer specifically to Vitality, Attunement, Endurance, Strength, Dexterity, Intelligence, and Faith, I refer to them as the character's statistics. When I want to refer specifically to scaling, attack rating, requirements, etc., I refer to them as the weapon's statistics. And so on and so forth.
Expert Opinion of High Ledgen Saturday-Saint #2 + #3: Originality is Overvalued/The Purpose of a Build
http://mmdks.com/21l6
This is a very generic build that happens to be stronger than what 99% of Dark Souls players could come up with on their own. Many people want to come up with their own builds. Many people think that using a generic build like this is somehow wrong or unoriginal. To that I say: buildcrafting is not what Dark Souls is about. The choices you make in a fight are far more important than the choices you make in mugenmonkey. There is more room for creativity in a single match between two generic builds than there is in the entirety of buildcraft. A build is just a tool. It has two purposes:
-To have a behavior that you enjoy playing.
-To have statistics that make that behavior powerful.
Therefore, there are only two purposes to making your own build:
-If no build available has behavior which you enjoy.
-If you think you can make a more powerful build with similar behavior.
To know what behavior you want, how to create a build with that behavior, and how to make that behavior effective, all take a lot of experience. I started playing Demon's Souls PvP in early 2010. Up until around mid-2011, all the cookie-cutter builds served my purposes just fine. I never found any fault in them, so I never made my own builds. Now I'm not saying everybody has to get 15+ months of PvP experience before they're allowed to make their own builds. What I am saying is that it took me 15+ months of PvP experience before I could look at a standard build and say, "I can do better." Until you can say the same thing, there is no reason for you to bother making your own build. It doesn't matter if it takes you a week of experience or a month or a year or never. If you can't make an honest evaluation of a standard build and find something worth fixing, don't. Instead of trying to fix something that isn't broken, use it. Improve your skills at PvP while using it. Ultimately it is that experience in PvP where you learn which behaviors are strongest. And when you know which behaviors are the ones you prefer, then start making your own builds.
To demonstrate these ideas, I am going to go over two builds that I use and the thought process behind making them.
http://mmdks.com/21lf - Generic Dexterity build. There are five attributes of this build that I'd consider 'non-standard' for a min-maxed build that are personal touches I added. They are:
-Gravelord Greatsword Dance
-Cloranthy Ring
-Endurance gouge on a scaling build
-Dark Hand
-Replenishment
This build was primarily designed around taking the behavior of a generic uchigatana build (which I like) and making it more powerful (because I can). As somebody with an Expert Opinion, I am well-equipped for the task of making builds stronger.
-Replenishment gives an effective boost to health, usually between 300 to 500 (not including bonus from Synergies, which is +120 health per synergy), and nullifies chip damage dealt by buff users while I kite them. Massive boost to survivability.
-Dark Hand is the one decision I made because of behavioral preference. I like having something that reliably blocks whatever damage type I'm up against. I dislike hardswapping equipment in and out, even before a fight starts. I also like the pushback.
-Endurance gouge. I tend to value defense more highly than most people, I think. I think that increasing endurance to use armor with stronger defense often improves survivability more than gouging vitality past 50.
-I usually don't equip stamina-regenerating items. I almost never use Cloranthy Ring. But in this case, since I wanted to use Dark Hand and Replenishment, it left me with only the Child's Mask and Cloranthy Ring. I decided to use Cloranthy Ring since it gives a bigger boost, but decided that Cloranthy and Child's together was overkill.
-Gravelord Greatsword Dance exists because it only costs 1 stat point and converts 0-damage BS's into ~500 damage and another knockdown, which is basically what I'd get if I BS'd somebody with an Uchi anyways.
This example is to emphasize that if you are not satisfied with the cookie-cutter builds out there, you often don't have to go to much length to change it into something you enjoy. Rarely do you need to make your own build from the ground up.
http://mmdks.com/21lo - Almost as if to sabotage the statement I just made, here is a build that is radically different from any other build I have played as or against. Even compared to other Greatbow builds, it is very different. This build was designed from the ground up to fit a specific behavior. I wanted to take what I enjoyed about the Dragonbone Smasher from Demon's Souls and put it into a build. I enjoyed that the DBS was a very defensive weapon which focused on turning a single hit into a series of mix-ups that would kill the opponent. This build is that playstyle turned up to eleven. It is extremely defensive only needs to win one or two mix-ups in order to kill. I'm not going to go over every decision I made with this build like I did the last one. It would take too long. This build is highly iterative and I've been working on it almost since Dark Souls came out. But I'll give you a general overview of its current state.
-Greatbow. This is the primary focus of this build. The Greatbow is an extremely defensive weapon and powerful control mechanism. It has virtually no offensive capability, but once you have drawn the bow, it can handle almost any attack thrown its way. It fires instantly, so it can trade with nearly any attack on reaction. It always staggers or knocks down its target, so you can't poise through the attack. It can first-hit swap escape (FHSE) to cancel its huge recovery after a trade allowing you plenty of time to tickstab people who poise through its knockdown.
-Everything else. Supports the Greatbow. The Greatbow itself is not very powerful. It deals really weak damage, you can rollstab it for free, shieldpokes shut it down 100%, et cetera. I threw in some tools to take care of this. Black Flame is a 'get off me' button. Fast, 60 poise damage. Mm-mm. Dark Silver Tracer turns any hit into massive tickstab. Dragon Roar crosses up shield users and makes DST BS for like 1,600 damage. Dark Hand is a good shield, light-weight, 80% all block, mostly used for FHSE into parries if I get hit during bow draw animation.
-0 Poise. I often say that 0 poise isn't viable, and I'm right. This build is an exception. 0 poise opens up a lot of options for this build thanks to first-hit swap escape. Also the usual weakness of 0 poise—that you lose attack trades with everything in the game and have no good response to light/medium weapon R1 spam—doesn't really apply to Greatbows. If you FHSE then as long as you don't get knocked down, the Greatbow will win every trade regardless of your opponent's poise and it will stop any attack in the game. Uchi rolling R1 is still annoying since arrows tend to fly over their head if you're locked on.
I also want to touch on new players and originality. New players often sabotage themselves when they make their own build, not only because they make a crap build, but also because they make a build that will not help them learn how to play. Many of the standard builds are not only powerful and easy to use, but reward simple, fundamental play. I often seen new players create a build designed around some gimmick, much like the above Greatbow build above. They have an idea for a build, and the design the entire build around supporting that idea. There are two major problems to this. 1) Gimmick builds that don't suck are hard to make. You have to know a good gimmick to build around, and then you need to know exactly how to . Like I said, that build took over a year of iteration, and that was after I already had something like two years of Souls PvP experience. And 2) Gimmick builds teach you how to use a gimmick. Beginners should concern themselves with learning the fundamentals of the game. Positioning, accounting for lag, knowing when to attack, having a sense of timing, anticipating the opponent, etc., are all valuable things for a beginner to learn. How best to trick people with Camouflage > Wrath of God shenanigans is something that might be valuable for a bored expert to learn.
Expert Opinion of High Ledgen Saturday-Saint #4: The Secret Fourth Axiom
I kept this axiom a secret because it isn't particularly useful for buildcrafting. It is mostly just a curiosity, perhaps more useful probably for game design (this whole thing was spurred by a discussion on game design) than build design, but it's a good enough thing to know.
Axiom 4: All behaviors can be expressed as a statistic.
A weapon's swing animation can be expressed as a collection of X,Y,Z co-ordinates that define the space the weapon occupies over time. A weapon's swing speed and hitstun can be expressed by their durations. Et Cetera. Anything we call behavior can ultimately be reduced down to a bunch of numbers and thus expressed as a statistic. Actually it's probably how the software the game is made on is thinking about everything. It's really not useful to think about behavior in this way, though. The value of knowing this is that it points our a contrast between behavior and statistics and how we control each. Near the beginning I brought up the issue that there is a fuzzy line between behavior and statistics. It is hard to say if the increased recovery on a MSGS's swing is a statistic or a behavior. Axiom 4 tells us we don't need to care about that fuzzy line, because it's basically all the same.
Don't get caught up in the technicalities of the axioms. Don't use them to justify making dumb decisions, and don't get confused if you find something that contradicts them. The purpose of the behavior/statistic dichotomy is to help us organize our thoughts and come up with a better thought process for Buildcrafting. As I said in the closing of my first opinion, the axioms have no ontological value. They aren't here to tell us what a build "really is" and Axiom 4 shows us that. It is not important to define whether swing recovery a behavior or a statistic. What matters is that if you want to reduce your recovery time, you have to get a different weapon.
"There are cards that are bad, but seem good. There are cards that are good, but seem bad. There are cards that are strictly better than other cards in most situations, but far worse in very specific situations. There are cards that are incredibly good, but no good deck can be built around them. There are cards that are bad, but are used in the very best deck because the rest of the deck is utterly dominating and desperately needs a certain ability, even if the best available source of that ability is on an otherwise terrible card.
Perhaps the hardest lesson to learn, the lesson that every Magic player seems to have to learn over and over, is that there are great cards that come together to make great decks that ultimately do not win because they aren’t as good as something else that’s out there."
-"What Makes the Best Player?" from Playing to Win by David Sirlin. Talking about Magic: The Gathering. Relevant to Dark Souls Buildcraft. Also I wanted an excuse to plug this. You should read the whole thing. It's not very long and it's all available online, just ask Google. Worth reading if you're into any PvP game (and I assume you are if you're reading this). Just make sure you read my far more important Expert Opinions on Buildcraft first.
Expert Opinion of High Ledgen Saturday-Saint #1: The Axioms of Buildcraft.
Axiom 1. A build is composed of equipment, spell selection, and a character's statistics.
Axiom 2. A build's behavior is primarily determined by its equipment and spell selection.
Axiom 3. A build's statistics should be designed to support its behavior.
Perhaps I did not thoroughly explain the difference between behavior and statistics earlier. I will broadly define behavior as, 'how something acts,' and statistics as, 'the degree to which something acts.' A Zweihander's swing is fundamentally different than a Longsword's in many ways. Its swing arc covers a different area and it is much slower. On the other hand, its damage is simply higher than the Longsword's. The swings act differently. When they connect, however, their function is the same: they deal damage. Only the degree to which they deal damage is different. This is what I mean by behaviors and statistics.
The First Axiom merely points to a build's components. Equipment includes consumables and anything that may be hardswapped onto a build in the middle of combat. Worth noting is that if you are not going to hardswap equipment in the middle of combat, merely changing your equipment or spell selection constitutes a change in build according to this definition.
The Second Axiom points to the importance of equipment and spell selection. More than your character's statistics, this determines your build's strengths and weaknesses, and its behavior. A Strength build wielding a Demon Great Machete is more similar to a Quality build using a Zweihander than it is to a fellow Strength build wielding a Great Club. While the two Strength builds would be statistically more similar, their behavior is quite different, as they are wielding different weapons. The Zweihander-wielding Quality build, while being statistically dissimilar, is behaviorally nearly identical to the DGM-wielding Strength build.
The Third Axiom points to the purpose of your character's statistics (Vitality, Attunement, Endurance, Strength, Dexterity, Intelligence, and Faith). They are a supportive structure for your spell and equipment selection, which as the First Axiom states, are the primary determiners of a build's behavior. Your character's statistics should be chosen to support the behavior determined by your equipment and spell selection.
I would like to point out that these axioms are not designed to have any ontological value. Rather, they are designed to help you think about builds and how they are designed. I would also like to point out that 'statistics' can refer to the statistics of anything, including your character, your weapon, armor, et cetera. A build's statistics are determined by the statistics of all of its components. When I want to refer specifically to Vitality, Attunement, Endurance, Strength, Dexterity, Intelligence, and Faith, I refer to them as the character's statistics. When I want to refer specifically to scaling, attack rating, requirements, etc., I refer to them as the weapon's statistics. And so on and so forth.
Expert Opinion of High Ledgen Saturday-Saint #2 + #3: Originality is Overvalued/The Purpose of a Build
http://mmdks.com/21l6
This is a very generic build that happens to be stronger than what 99% of Dark Souls players could come up with on their own. Many people want to come up with their own builds. Many people think that using a generic build like this is somehow wrong or unoriginal. To that I say: buildcrafting is not what Dark Souls is about. The choices you make in a fight are far more important than the choices you make in mugenmonkey. There is more room for creativity in a single match between two generic builds than there is in the entirety of buildcraft. A build is just a tool. It has two purposes:
-To have a behavior that you enjoy playing.
-To have statistics that make that behavior powerful.
Therefore, there are only two purposes to making your own build:
-If no build available has behavior which you enjoy.
-If you think you can make a more powerful build with similar behavior.
To know what behavior you want, how to create a build with that behavior, and how to make that behavior effective, all take a lot of experience. I started playing Demon's Souls PvP in early 2010. Up until around mid-2011, all the cookie-cutter builds served my purposes just fine. I never found any fault in them, so I never made my own builds. Now I'm not saying everybody has to get 15+ months of PvP experience before they're allowed to make their own builds. What I am saying is that it took me 15+ months of PvP experience before I could look at a standard build and say, "I can do better." Until you can say the same thing, there is no reason for you to bother making your own build. It doesn't matter if it takes you a week of experience or a month or a year or never. If you can't make an honest evaluation of a standard build and find something worth fixing, don't. Instead of trying to fix something that isn't broken, use it. Improve your skills at PvP while using it. Ultimately it is that experience in PvP where you learn which behaviors are strongest. And when you know which behaviors are the ones you prefer, then start making your own builds.
To demonstrate these ideas, I am going to go over two builds that I use and the thought process behind making them.
http://mmdks.com/21lf - Generic Dexterity build. There are five attributes of this build that I'd consider 'non-standard' for a min-maxed build that are personal touches I added. They are:
-Gravelord Greatsword Dance
-Cloranthy Ring
-Endurance gouge on a scaling build
-Dark Hand
-Replenishment
This build was primarily designed around taking the behavior of a generic uchigatana build (which I like) and making it more powerful (because I can). As somebody with an Expert Opinion, I am well-equipped for the task of making builds stronger.
-Replenishment gives an effective boost to health, usually between 300 to 500 (not including bonus from Synergies, which is +120 health per synergy), and nullifies chip damage dealt by buff users while I kite them. Massive boost to survivability.
-Dark Hand is the one decision I made because of behavioral preference. I like having something that reliably blocks whatever damage type I'm up against. I dislike hardswapping equipment in and out, even before a fight starts. I also like the pushback.
-Endurance gouge. I tend to value defense more highly than most people, I think. I think that increasing endurance to use armor with stronger defense often improves survivability more than gouging vitality past 50.
-I usually don't equip stamina-regenerating items. I almost never use Cloranthy Ring. But in this case, since I wanted to use Dark Hand and Replenishment, it left me with only the Child's Mask and Cloranthy Ring. I decided to use Cloranthy Ring since it gives a bigger boost, but decided that Cloranthy and Child's together was overkill.
-Gravelord Greatsword Dance exists because it only costs 1 stat point and converts 0-damage BS's into ~500 damage and another knockdown, which is basically what I'd get if I BS'd somebody with an Uchi anyways.
This example is to emphasize that if you are not satisfied with the cookie-cutter builds out there, you often don't have to go to much length to change it into something you enjoy. Rarely do you need to make your own build from the ground up.
http://mmdks.com/21lo - Almost as if to sabotage the statement I just made, here is a build that is radically different from any other build I have played as or against. Even compared to other Greatbow builds, it is very different. This build was designed from the ground up to fit a specific behavior. I wanted to take what I enjoyed about the Dragonbone Smasher from Demon's Souls and put it into a build. I enjoyed that the DBS was a very defensive weapon which focused on turning a single hit into a series of mix-ups that would kill the opponent. This build is that playstyle turned up to eleven. It is extremely defensive only needs to win one or two mix-ups in order to kill. I'm not going to go over every decision I made with this build like I did the last one. It would take too long. This build is highly iterative and I've been working on it almost since Dark Souls came out. But I'll give you a general overview of its current state.
-Greatbow. This is the primary focus of this build. The Greatbow is an extremely defensive weapon and powerful control mechanism. It has virtually no offensive capability, but once you have drawn the bow, it can handle almost any attack thrown its way. It fires instantly, so it can trade with nearly any attack on reaction. It always staggers or knocks down its target, so you can't poise through the attack. It can first-hit swap escape (FHSE) to cancel its huge recovery after a trade allowing you plenty of time to tickstab people who poise through its knockdown.
-Everything else. Supports the Greatbow. The Greatbow itself is not very powerful. It deals really weak damage, you can rollstab it for free, shieldpokes shut it down 100%, et cetera. I threw in some tools to take care of this. Black Flame is a 'get off me' button. Fast, 60 poise damage. Mm-mm. Dark Silver Tracer turns any hit into massive tickstab. Dragon Roar crosses up shield users and makes DST BS for like 1,600 damage. Dark Hand is a good shield, light-weight, 80% all block, mostly used for FHSE into parries if I get hit during bow draw animation.
-0 Poise. I often say that 0 poise isn't viable, and I'm right. This build is an exception. 0 poise opens up a lot of options for this build thanks to first-hit swap escape. Also the usual weakness of 0 poise—that you lose attack trades with everything in the game and have no good response to light/medium weapon R1 spam—doesn't really apply to Greatbows. If you FHSE then as long as you don't get knocked down, the Greatbow will win every trade regardless of your opponent's poise and it will stop any attack in the game. Uchi rolling R1 is still annoying since arrows tend to fly over their head if you're locked on.
I also want to touch on new players and originality. New players often sabotage themselves when they make their own build, not only because they make a crap build, but also because they make a build that will not help them learn how to play. Many of the standard builds are not only powerful and easy to use, but reward simple, fundamental play. I often seen new players create a build designed around some gimmick, much like the above Greatbow build above. They have an idea for a build, and the design the entire build around supporting that idea. There are two major problems to this. 1) Gimmick builds that don't suck are hard to make. You have to know a good gimmick to build around, and then you need to know exactly how to . Like I said, that build took over a year of iteration, and that was after I already had something like two years of Souls PvP experience. And 2) Gimmick builds teach you how to use a gimmick. Beginners should concern themselves with learning the fundamentals of the game. Positioning, accounting for lag, knowing when to attack, having a sense of timing, anticipating the opponent, etc., are all valuable things for a beginner to learn. How best to trick people with Camouflage > Wrath of God shenanigans is something that might be valuable for a bored expert to learn.
Expert Opinion of High Ledgen Saturday-Saint #4: The Secret Fourth Axiom
I kept this axiom a secret because it isn't particularly useful for buildcrafting. It is mostly just a curiosity, perhaps more useful probably for game design (this whole thing was spurred by a discussion on game design) than build design, but it's a good enough thing to know.
Axiom 4: All behaviors can be expressed as a statistic.
A weapon's swing animation can be expressed as a collection of X,Y,Z co-ordinates that define the space the weapon occupies over time. A weapon's swing speed and hitstun can be expressed by their durations. Et Cetera. Anything we call behavior can ultimately be reduced down to a bunch of numbers and thus expressed as a statistic. Actually it's probably how the software the game is made on is thinking about everything. It's really not useful to think about behavior in this way, though. The value of knowing this is that it points our a contrast between behavior and statistics and how we control each. Near the beginning I brought up the issue that there is a fuzzy line between behavior and statistics. It is hard to say if the increased recovery on a MSGS's swing is a statistic or a behavior. Axiom 4 tells us we don't need to care about that fuzzy line, because it's basically all the same.
Don't get caught up in the technicalities of the axioms. Don't use them to justify making dumb decisions, and don't get confused if you find something that contradicts them. The purpose of the behavior/statistic dichotomy is to help us organize our thoughts and come up with a better thought process for Buildcrafting. As I said in the closing of my first opinion, the axioms have no ontological value. They aren't here to tell us what a build "really is" and Axiom 4 shows us that. It is not important to define whether swing recovery a behavior or a statistic. What matters is that if you want to reduce your recovery time, you have to get a different weapon.
