Expeditions: Rome Developer Diary - Legion Battle Improvements

Soon after Expeditions: Rome launched, we got this developer diary outlining some areas of the game that required improvement in the short term. One of those areas were the rather obtuse legion battles, and as a result, we now get this new diary that talks about some upcoming changes to the system and the original ideas behind it.

Check it out:

Ave, and welcome to our 18th dev diary – the second after the release of Expeditions: Rome. As we mentioned last week, the first major update is currently undergoing testing. We were hoping to release it last week, but QA did their jobs well and found a new issue that had slipped into the game due to one of our stability fixes, so we had to push the patch until this week so we could get that fixed first and give QA time to re-test the build.

In the meantime, those of us not already working on the first piece of DLC have set our sights on a much-requested set of improvements to a certain system in the game, namely the legion battles.

The legion battle system has been a challenge right from the start. We knew we wanted to represent the large-scale battles of a Roman legion somehow, and give the player a feeling of being a grand strategy commander handing out orders from on high, but at the same time we didn’t want to cram a whole wargame into our tactical RPG. During the prototyping phase we intended battles to simply be resolved by a die roll based on the strength of each army – but it seemed too simple. Later we briefly considered representing it by a re-interpretation of a Roman dice game called Tali – but we decided that would be far too abstract. We knew we needed to do something that would sell the fantasy without ballooning out of scope to take over the whole project.

What we settled on was a resource management system where you balance your losses against the risk to your centurions, a desire to scatter the enemy troops, the likelihood of getting some loot out of it, the morale and experience of your legion, and so on. Success is still essentially a dice roll based on your legion’s strength compared to the enemy army, and you would always be able to win as long as you paid just a little bit of attention. However, you would have some control over how well you come out of it in the end: how great your losses, how good your loot, and whether the enemy survivors would retreat to fight another day.

Now that the game has been out for a few weeks, it's clear that the system was not received well. Although Expeditions: Rome is a premium single player game, we’re committed to giving it the post-release support it needs to remain an excellent and beloved game for years to come, and the big thing we need to address is how to make this legion battle system more interesting.

There are limits to what we can do. We are still not planning to turn this into a real-time grand strategy game. It must remain a relatively minor part of the game loop that doesn’t overstay its welcome and get too much in the way of the core gameplay of Rome.

We have perused all your feedback and identified the following two major criticisms: First, it’s too difficult to understand what’s going on and how it works; second, it feels random, not giving you enough agency over the outcome of battles especially in the beginning before you have a chance to unlock new stratagems and level up your centurions.

With that in mind, we’ve decided to make the following changes. First, we will remove the “challenges” that pop up during each phase of a battle. These are the “attacks” and “defences” you see getting either succeeded or failed by your chosen commander, based on his character class and his command specialisations. This will also have the benefit of making each phase shorter, for those who don’t wish to skip them entirely.

To replace this system, we plan to add new trackers that compare how many points your centurions have in each specialisation compared to the enemy commander. To refresh your memory, the specialisations are Infantry, Cavalry, Artillery, and Logistics. Each specialisation your centurions have will be added to your count, and your chosen commander will get their specialisations doubled for this purpose. The enemy’s specialisations are deducted from your count, and each tracker can go into the negative this way. To help you visualise it, here is an early mock-up of what two of these trackers may look like in the UI. Each line marks a point where a new effect will be unlocked[...]

In each subsequent phase, every stratagem you play will add points to or remove points from these trackers. This will simplify the stratagem cards as well, reducing the plethora of largely unexplained effects to a combination of specialisation points. The enemy will also play stratagems that affect these trackers, and you will be able to see at least some of what the enemy is going to do before you make your own choice.

At certain tiers, each tracker unlocks a new effect: a bonus to your defence or aggression, a new loot crate, a morale bonus, ultimately even forcing the enemy army to scatter rather than retreating. This means the new system allows you to aim for tangible goals: rather than a card just making you lose slightly less manpower, you might be trying to build up to reach a certain tier of Artillery in order to unlock a particular effect, and which cards you draw in the following phase will determine whether you succeed or fail. Inspecting these effects and deciding what to aim for based on what the enemy seems to be focusing on, and what stratagems are at your disposal, will give you a better sense that you’re formulating and following an overall strategy, and that you’re going up against another commander with their own strategy.

To give you some idea of what these effects could look like, here’s a screenshot directly from the design document. Please note that this is extremely work in progress, subject to all sorts of pending iterations, and reproduced here over the dead body of the technical designer (he will be missed)[...]

Since we’re removing the challenges that were formerly based in part on the chosen commander’s class, each class will instead apply a specific overall bonus to the outcome of the battle, similar to how certain perks (namely Cautious, Reckless, and Medicus) already add specific bonuses or penalties to a battle when a character with those perks is selected. These bonuses will not only be more significant and tangible than the previous effect of the commander’s class, but will also be completely orthogonal to the specialisations, creating more variation in possible outcomes based on who you choose to command the battle.

In addition to these core changes, we have a few minor tweaks planned. First of all, losing battles will now affect the legion’s Morale directly: before, the morale effect was tied to the specific set of stratagems you would have to choose from if you lost a battle, but this just seems like the game is forcing you to make a bad choice. Simpler to just deduct the Morale directly if you lose. Second, winning a battle will always give the legion Experience. In the current system, Legion Experience is based on relative losses during a fight: if you loose 100 more manpower than you kill, you lose 10 Experience, and vice versa. However, losing Experience feels bad, so we’ll remove that.

And finally, this is not a simplification but rather an effort to bring the system in line with expectations and with the core fantasy: attrition during each phase (how many men you lose or enemies you kill) will be based on the relative active manpower of each side, rather than now where it is essentially a predefined random range that increases in each phase.

Once we have these changes in place, we need to test it and iterate on it, paying special attention to how well we explain how the system works and what you as the player should be thinking about during a battle. Though we are removing some of the more confusing aspects of the system, we are adding new elements to it as well, and we want to make it clear this time what’s going on during each phase and how your decisions have affected the outcome.

We’d love to hear what you think of these changes. Please leave your comments and questions below and we will be guaranteed to read them. We will as always have a Dev Stream on Twitch this Wednesday, February 9th at 1:00 PM Eastern / 6:00 PM GMT at http://twitch.tv/thqnordic, where Senior Producer Brad Logston will host Creative Director Jonas Wæver to talk about our plans for this system and what other things we’re working on for future updates. We will answer any questions left on this diary, and we’ll of course try to answer any other questions you throw at us during the stream.

Until then, Valete!