Dev Blog: General Skill Tree
Dear Settlers,
A "Dev Blog" provides a glimpse behind the scenes of The Settlers Online. It's written by our Game Design Team and it's not translated in order to avoid information loss or unintentionally changing important facts. This time, BB_Vierauge provides a lengthy insight into The General Skill Tree! Let's have a look!
<h1>Dev Blog: General Skill Tree</h1>The wait has finally come to an end with the arrival of the General Skill tree at The Settlers Online!
Most of you have been waiting for this feature since the science system was initially introduced in 2013 (and not just you – we even found marketing art from back then!), so we want to give you some insight as to what went into the production of this highly anticipated update.
Because the skill tree for quartermasters is pretty straightforward, I will be focusing on the one for combat generals. Thus, unless otherwise stated, I will be referring to the combat general skill tree from now on.
<h2>Our Goals</h2>The obvious primary objective for the skill tree was what you'd expect: To offer an alternative way of spending books to permanently make generals more efficient.
This can be achieved by speeding up various tasks, reducing unit losses (usually by making your army more powerful), and increasing rewards.
However, we had certain secondary goals in mind:- Make the two bottom rows of skills specifically useful for newer players;
- Offer meaningful choices in as many rows as possible;
- Create opportunities to use new army compositions for old camps.
However, the approach for some of these goals changed over the course of the testing phase.
<h3>Favouring Low-Level Players</h3>While balance dictates that more powerful skills should be found at the top of the tree, we wanted some early skills to specifically favour low-level players, offering them the possibility of catching up to the masses more quickly with easily accessible skills.
The most extreme example of this was Home Field Advantage, a skill that greatly increased the attack damage of all units on the home island. Home Field Advantage was removed during testing, but other skills based on this philosophy still exist in the final version of the skill tree. Can you identify them?
Anyway, in the version of the tree that we initially proposed, the first two rows basically offered options between optimised skills for low-level players and those that speed up various tasks (with the exception of the powerful Overrun, which was available in the first row). Higher-level players were justifiably disappointed in having basically no combat relevant choices or even being forced to invest books on skills that were perceived as subpar.
Consequently, we removed the most specific offender (Home Field Advantage) and moved First Aid and Hostile Takeover into these rows, swapping their positions with speed increasing skills. These two skills have the benefit of offering the same absolute value to high-level players while being relatively more useful to low- and mid-level users whose recruit production represents a more relevant obstacle. These skills are also useful at all times, without the need to optimise armies for their use or read resources available online.
<h3>Build Diversity</h3>Choosing skills is supposed to be interesting and, therefore, we tried to offer equally powerful options in every row. Our approach to this during the first few versions on the test server was trying to group similar skills in the same rows:
The first iteration offered three-speed enhancing skills in the second row and the second iteration concentrated almost all powerful combat skills in only two rows.
While this may sound like a reasonable approach, there were ultimately a couple problems with it:- Choosing between multiple combat enhancing skills is simply a math problem. While different setups are stronger against different camps, there will be an optimal choice (possibly based on a number of conditions) that the community will eventually figure out. At this point, there is nothing interesting about the choice.
- Diverse combat configurations do not mix well with adventure guides. While not all players use them, player created adventure guides play a major role in the community. Their existence result in two problems: first, if a variety of combat configurations exists, strategy maps will have to cover all of them. Second, if guide makers agree on an optimal configuration to support, players will have to adapt that one, defeating the original point.
Realising these problems led us to go with a different approach: instead of letting players choose between multiple skills with similar benefits, we are now mixing different types of skills in the same row.
You will find that most rows will now feature two skills that boost combat performance, one that speeds up a task and one that gives out rewards at the end of a battle.
This way, players can learn 5 out of the 6 existing ranks of combat skills, or decide that they value a particular benefit more than a combat boost in that specific row. This lessens the impact of the guide problems only moderately but guarantees choices even if the mathematical part is solved. Preferring travel speed over reduced boss HP is a question of playstyle and preference, not math.
<h3>Promote New Army Setups</h3>While this was a minor goal, we wanted at least some skills to make players use new unit configurations in their armies to get the most out of them.
The obvious skills that do this are the ones that buff specific, and often underused, unit types. For example, Cleave turns Elite Soldiers into heavy hitters with splash damage, previously a niche among non elite units.
Battle Frenzy can shift setups towards more defensive armies in general. While you cannot go too defensive (you still want to reduce the damage you receive by taking out enemies), drawn out battles do become a more viable strategy.
<h2>Risks and Problems</h2>As many of you who have experimented with a combat simulator will know, our game's combat system can be very swingy.
Unless splash damage is involved, a single extra hit point can double the number of attacks required to defeat a unit; first strike is usually a decisive advantage; non related changes to movement and combat speed can enable new blocks and locks that allow players to bypass entire camps.
An attentive player will notice that these things have guided the design of our skills in various ways. Many skills grant splash damage to various player units; damage boosts to swift units (and generals!) are relatively minor, and the combat speed boosting Confident Leader is competing with major combat skills in its row.
During testing, many players suggested hit point increasing skills for defensive units, but a skill that increased recruit HP had been designed, analysed and dismissed before the test phase started because it was both too powerful and too swingy, even at just a single point.
Finally, all of this means that the impact of various skills will be vastly different in different adventures. For example, because most Nords have last strike, skills that buff your units that attack with normal initiative are extra powerful since there is little point in bringing swift units along. This not only meant that we had to run a lot of simulations, it also resulted in some adventures benefitting less from the combat focused skills being, unfortunately, unavoidable.
<h2>Jog: The Story of a Problem Child</h2>Players who have followed the testing process will know that the Jog skill has gone through a drastic change before the release.
It originally affected the generals' movement speed in all situations, including attacks. The bonus back then was relatively small - 20% or 30% depending on the iteration. That version, however, had a couple problems:
The most significant one was that it forced players to have the same amount of Jog ranks on all generals that participate in the same combination of blocks when following most guides. This was an unreasonable additional cost for a skill that was supposed to have easily understandable benefits.
The second problem was that the small bonus, while theoretically significant, did not seem powerful when observed with the naked eye. To keep it in line with other speed skills and to reign in its (b)locking powers, we could not increase the bonuses to an easily observable level.
Players also debated if even equally skilled generals might throw off the timing in some edge cases. While this discussion never quite came to a conclusion, I at least wanted to mention it here because it was what drew our attention to the skill in the first place.
As you probably already know, Jog's final version does not apply to attack movements. This eliminated the problem of interference with (b)locks in any way and allowed us to significantly boost the bonus (because it works in fewer situations) to the point of being perceivably faster.
We know that many players are disappointed with this change, and we have seen quite a few good suggestions for compromises on the test forum. Ultimately, we decided on a different solution that will be implemented over the course of multiple upcoming updates:
We are going to introduce a new type of zone-wide buff that increases general movement speed. Weaker buffs will be producible (relatively cheap) in provision houses, and more powerful buffs will be available as loot and at the Merchant for gems and star coins. While this is a less permanent investment than the science system, the buffs affect all generals at once and do not come at the cost of sacrificing other skills. I hope you are looking forward to them!
We have also halved the time it takes to erect and break down a garrison when transferring it to a new location. This makes the increased transfer speed even more relevant, which results in transfers being a more viable alternative to immediate attacks when sacrificing generals or performing a so-called leech attack for shared loot.
<h2>Other Changes</h2>The skill trees focused our attention on the combat system in general, and while we cannot easily change mechanics at this point, we have taken some steps to improve presentation and communication.
Not only have we revised how information is displayed in unit tooltips, we have also implemented them for the generals themselves, and added unit type categorizations and icons that help identify a unit's rough capabilities at a glance.
The unit types displayed above garrisons and camps are now sorted in the order in which they will get attacked by opponents without flanking. This order is practically arbitrary and used to be completely inaccessible to players who did not use online resources, so providing a way to access this information was long due.
Speaking of flanking: we realise that getting used to this new term may take a while, but we ultimately believe that it was worth introducing shorter keywords for “attack weak targets first” and “extra damage to buildings”.
Revising the tooltips of generals in various places also led us to improve the tooltips of other specialists. Their tooltips in the star menu now display all skills that they have learned, so you will have an easier time discerning them.
Last but not least, we have added new archive entries and updated a few old ones. We hope that this will help players to learn about the combat system within the game itself instead of relying on external sources.
<h2>Closing Words</h2>This blog turned out a lot longer than I originally intended, but I hope you will take it as a sign of how passionate we are about this update. We are excited to see how the wider player base will react and adjust to all these changes.
We already have a lot of new content in store for the next months, and I hope you are looking forward to learning what we have planned.
For now, we hope you have fun teaching your generals some real skills!
BB_Vierauge
Please post your feedback here.
