Inside the Box: Designing Krieg, Part 3 – Bloodlust

Inside the Box serves as a forum for individuals involved in the production of Gearbox Software content to share personal motives, methods, process and results. Gearbox Software projects are created by a diverse range of individuals spanning a spectrum of different backgrounds, interests, objectives and world views. The views and opinions expressed in this article are those of the author and do not necessarily reflect the official policy or position of Gearbox Software or any of its individual members outside of the author.

Welcome back to Inside the Box! I’m Paul Hellquist, the Creative Director and Lead Designer on Borderlands 2.

Today I’m going to be telling you about the creation of Krieg the Psycho’s Bloodlust tree. If you missed it yesterday, I talked about the design goals for Krieg and the evolution of the Buzz Axe Rampage action skill.

Later this week, on Friday, we’ll be releasing a short video that shows how Krieg the Psycho became a Vault Hunter — you can see a still from that video above. And in anticipation of that video, we’ll have a new Inside the Box every day leading up to it! Here’s what to expect:


Kill Kill Kill

The Bloodlust tree started out life called “Bloodbath” and revolved entirely around “kill skills” — skills that only trigger when the player kills an enemy. This theme for was inspired by powerful moments that players had in the first Borderlands where they could get 3 or more kill skills to trigger all at once, providing players with a great moment during combat where they felt super charged and would try to maintain it by getting additional kills. These moments were much harder to come by in Borderlands 2 since the skill trees were larger, making it more difficult to get multiple kill skills purchased at once. I wanted to bring those moments back to Borderlands 2.

I also knew I needed a skill tree that could really excel with the guns in Borderlands 2 despite the character’s heavily melee-based action skill. The aggressive style of trying to maintain kills to keep your kill skills always on was a perfect fit for the overall goals of getting players to become a hyper-aggressive Psycho bandit.

I filled out the tree and explored the idea of the kill skills themselves getting more powerful for each kill skill that was active. For example if there was a skill that increased reload speed after a kill it would have an additional bonus for each kill skill that was activated. This mechanic encouraged players to spread their points out to get as many kill skills as possible. It was a decent mechanic.

That theme for the tree never got past the “play it in your head” stage of development mentioned in the action skill discussion. Matt Armstrong (Borderlands Franchise Director) and I believed that the power swings of the tree would be too spikey if you were to graph it out. You would be a puny weakling, get a kill, be a god for 7 seconds, and then be a puny weakling again. This spikey nature of the power curve over time would be difficult to balance and we felt it wouldn’t encourage becoming the combat-crazed Krieg as strongly as we wanted.

Gaige’s Anarchy Skill

We also realized that Gaige’s Anarchy skill encouraged a similar kill-for-power behavior. We loved how the stack building mechanic changed the way you played and thought about gear. We decided a new twist on the stack building mechanic would provide a smoother power curve for the Bloodbath tree and could still support a strong kill skill theme. The main goal for this new stacking mechanic was that it had to be different enough from Anarchy that it provided different decisions and behaviors than when you were playing Gaige.

Bloodlust Is Born

I decided that instead of kills adding stacks that Krieg simply wanted people to feel the only thing he could feel: pain. This notion led to damage being the stack accumulation trigger. Since damage is so much easier and faster to accumulate we knew we needed a much lower cap. At first the stack limit was 50, but that turned out to be much too easy to reach to so it was quickly raised to 100. This turned out to be a great number as it allowed us during development to easily calculate the maximum bonuses players could earn. Of course, this easy math is also better for players to evaluate the skills as well.

Another important difference in the Bloodlust vs. Anarchy design is the notion of the stacks draining. Having the stacks drain away is an integral part of achieving the goal of turning players into the psycho bandit. I wanted Bloodlust Psychos to be twitchy, always on the move, running off to start the next fight since dealing more damage is the only way to maintain your stacks. Having the stacks drain quickly instills the sense of urgency and irrational aggression that makes players feel like a Psycho.

The quick draining of the stacks also changes your relationship with gear causing you to look for weapons that can deal the most consistent damage over a long period of time. Draining stacks makes the best weapons for Bloodlust accumulation the ones with large clips and, strangely, the lowest damage. Turns out Bandit weapons are exactly that. I wanted to subtly encourage players to use bandit weapons to support the flavor of Krieg without having a skill hit you over the head by saying, “You are better with bandit weapons.”

The last thing I wanted to change from the Anarchy model set by Gaige was how you purchased the skill that accumulates the stacks. In the Ordered Chaos tree the Anarchy skill is a 1 pointer on the same tier as the 4 point skill, “Smaller, Lighter, Faster.” This was a very clever bit of design by Jonathan Hemingway (Lead Character Designer on Borderlands 2) to ensure that players couldn’t explore deeper in the tree without purchasing the Anarchy skill which makes the whole tree go. In retrospect, what we didn’t like about that design was that it created a situation where the first 5 points in the tree were very boring. All of your decisions were predetermined for you. Point 1 was on Anarchy and points 2-5 had to be spent on “Smaller, Lighter, Faster.” I wanted to improve on this.

The solution was to basically give each of Kreig’s first tier skills an extra “secret” skill that enables the ability to accumulate Bloodlust stacks. This change allows players to have the normal range of interesting decisions on the first tier of the tree when spending those first 5 skill points.

Can’t We All Work Together?

Early in the life cycle of the tree there was a real problem between the relationship of Buzz Axe Rampage and the Bloodlust tree. The main problem at the time was that there were a couple of skills in the tree that provided health regeneration or damage reduction based on your stack count. These skills provided so much survivability and your guns were so effective that the Rampage was easily forgotten and never really useful when focusing on the Bloodlust tree. At the same time Andrew Hoffman (Lead Quality Assurance Tester for DLC) was telling me that whenever he used Rampage he just died all the time. What I decided to try was to take out all of the ways the Bloodlust tree could recover health. At the same time I added the ability for the Rampage to recover your health with a kill. This somewhat unintuitive notion of removing health gain to increase survivability worked like a charm! With no way to get health in Bloodlust it gave the action skill a role in that tree. It became the thing you do to get your health back during the battle.

Mmmmm…Tasty Blood

Another interesting skill in this tree that took a while to get right was Taste of Blood. For a long time it simply gave a big pile of stacks for kills during the Buzz Axe Rampage. It was pretty worthless because the stacks would generally decay by the end of the Rampage anyway. It needed something more if it was to have any value.

The next thing I tried was based on how I thought people would want to use the action skill in the tree. The theory was that since the stacks bleed away during the action skill so quickly that players would want to use Rampage when they are low on stacks to avoid losing them. The skill would provide damage reduction during Rampage where the fewer stacks you had the better the protection. Also, each kill during the action skill would add a ton of stacks to Bloodlust (a ton being something like 50). So as you get kills the damage reduction gets worse but your stacks are skyrocketing. By the time the skill ends you are probably between 40 and 80 stacks and ready to wreck people with your guns again. That was the theory anyway.

This version of the skill failed miserably! If functioned fine and it totally worked as I intended. The trouble was that I was the only one who understood how to use it. The problem was that every skill in the tree was better the more stacks you had. Taste of Blood was best in the exact opposite situation. This difference confused players and they rejected it. Taste of Blood is a great example of how you can sometimes overthink your design and, despite it making sound design sense, it breaks player’s expectations resulting in failure.

So I just flipped it around to work as players expected and then the skill received good marks from the focus groups resulting in the skill that you are playing today. I think it makes a little less sense strategically but the clarity and consistency benefits are well worth it.

That gives you a sense of the evolution of the Bloodlust tree. Be sure to come back tomorrow when I get in depth on the Mania tree!