The air combat section of Xenonauts 2 has gone through a number of iterations over the past few years, and with the project approaching Early Access we've taken the decision to switch to (an upgraded version of) the realtime air combat mechanics from the first game rather than pursuing the alternative turn-based model I've been experimenting with. I'll explain the reasons for this change below, but let's start by discussing the realtime mechanics and the planned improvements.
Realtime Mechanics:
We're already working on implementing the realtime air combat mechanics from the first Xenonauts and we're hoping to have them in the next major release (V8). This will also include various supporting strategic systems such as the ability to manufacture advanced aircraft on the Engineering screen, many of which require some extra work now the "classic" base update changed the way Hangars worked.
The goal for V8 is therefore to literally have the air combat from the first Xenonauts in the second game as a starting point to test our improvements. We'll probably chuck the same planes and weapons in the game with the same stats, and fit them into the tech tree in roughly the same place, and do the same for the UFO stats. This will make it easier for both us as the developers and you as the community to spot missing features or things that aren't working properly, and it also ensures that the strategy layer has reasonable balance / progression to allow us to test the new features we're planning to experiment with in future builds:
- Interceptor Components: on the Aircraft screen there are additional slots for new types of equipment that did not appear in the first Xenonauts (armour, engines). One of the main things I want to experiment with is to have fewer types of interceptor but more possible upgrades, making the tech tree more interesting ensuring each type of interceptor can potentially stay relevant for longer.
- As an aside, it might be interesting to give each aircraft type a Power stat and have the various weapons, engines and armour types draw a certain amount of power. So even basic aircraft can still use highly advanced equipment but can support less of it than the more advanced fighters.
- Also, if much of the cost of an aircraft comes from its components rather than the aircraft itself, we could re-implement permadeath for the aircraft itself but make most of the equipment recoverable when a plane is shot down.
- Clouds: these would provide cover on the battlefield and the amount and position of them would be randomised each battle. The idea is that combatants can move through clouds freely but they would block the fire arcs of weapons (and missiles wouldn't make course adjustments while flying through them).
- Hit / Evade Chances: this an experimental change we're going to try, where combatants have % Evade scores and weapons have % Accuracy scores and Evade modifiers. The Evade roll will no longer be manually triggered and will just play an evade animation (without moving the plane laterally) when an Evade occurs. We'll see if this improves the game and if not we'll return to the old system where weapons would always hit if in range.
- The % Accuracy on weapons is somewhat required if we're going to add Pilots to the game, as the obvious thing for pilots to do as they gain experience is provide an Accuracy bonus to their weapons and an Evade bonus to their interceptor. The same is true for upgrades like targeting computers or so forth; in the old X1 air combat there's just not many variables to play with and that limits the equipment and upgrade choices we can give the player.
- Relative Battlefields: in X1 the boundaries of the battlefield are set at the start of the combat, but in X2 the boundaries will always be a fixed distance from the main UFO. This will allow us to set some combats up as a chase where the UFO is trying to get far enough away from your planes to push them off the edge of the map, while peppering your pursuing interceptors with fire from a rotating turret weapon (or relying on their escorts to cover for them). It's not a huge thing but in X1 literally every UFO would just turn and fly towards your interceptors so it'd be nice if in X2 some UFOs tried something a little different.
- Special Equipment: we'll also likely be experimenting with some other types of equipment that weren't in X1, such as turret weapons that are capable of rotating their fire arcs, or shields. Not sure how many will provide practical but we've got a few ideas!
The main intention of these changes is to add a bit more variety to the air combat. One of the problems in X1 was that a combat featuring a particular UFO versus a particular combination of interceptors would almost always play out the same way every time, and there's a few things we can do to mitigate this. The addition of clouds means that the battlefield itself may cause the tactics to be different in different battles, and making weapons use % hit rolls should also ensure a bit more variation (e.g. a combat may play out quite differently if a long range volley of missiles at the start of combat scores 4 hits compared to if it scores 2 hits).
Depending on how the combat changes play out, I think the strategic side of the air combat may also become more complex and interesting. In X1 you were continually building steadily more advanced planes and getting rid of the older models once they became irrelevant, but if specific aircraft gain combat experience through a pilot system and are also more upgradeable than before then I can see more interesting choices becoming available to the player. Do you replace your experienced starting interceptors as soon as a better interceptor becomes available, or do you give them some upgrades and keep them around? Or just play aggressively with them until they get shot down, and then replace them? Etc.
What happened to the turn-based air combat model?
Up until V7 the game featured a turn-based air combat model. The plan was to add increasing complexity to this turn-based system until we got something that was complex enough to be fun, but was ideally a bit faster-paced than the X1 air combat and used a more similar skillset to the rest of the game.
The tun-based air combat in the public builds never got to the stage where it became fun. After the last iteration it was obvious that air combat needed proper 2D unit movement (rather than just 1D moving forwards / backwards) if it was to be interesting enough to support the more complex strategy layer that Xenonauts has compared to XCOM or classic X-Com. With overly simple air combat not only are the interceptions more boring, there's also less scope for research and UFO behaviour on the rest of the strategy layer too.
Unfortunately, when we set to work implementing this it became clear that trying to handle complex 2D movement in a fast-paced way was going to be impractical in a turn-based system. Obviously asking players to issue orders to all of their planes every few seconds wasn't an option as every combat would take hours, so we instead developed an "automated" move system based on auto-calculated moves towards your target enemy unit (or movement waypoint). We were hoping it would provide a realtime feel while retaining the turn-based system under the hood, but in practice it didn't work well - it was difficult for the user to understand what was going on and it didn't feel as natural or responsive as the X1 realtime system.
This is a bit of a shame, as the turn-based system we had planned had been paper prototyped and worked rather well as a board game. But if the fundamental building blocks of the system don't translate well onto the screen, there's point pursuing it further - it seems like we've taken the turn-based model as far as it could go. The best thing to do would just be to pluck out some of the interesting systems and merge them into the X1 realtime system.
I certainly think there's some scope to do this. Ideally, I want to try and minimise the amount of time players have to spend pausing / unpausing to try and pick the optimal split-second to do something (like rolling their planes to dodge incoming fire), so making Evasion auto-trigger on a % roll may help a lot here. Adding more variety to the air combat in general should also improve the experience and replayability for everyone, and new ideas such as clouds and the interceptor components actually work equally well under the old X1 realtime system as they do in the current X2 turn-based system.
Conclusion:
When development began I couldn't see many improvements that could be made to the X1 air combat, which was one of the reasons I was reluctant to use the same system - I felt like I'd be serving up exactly the same thing all over again. After all this experimentation I'm now pretty sure the air combat can be improved, and it's just a question of whether we can improve things a little (by adding clouds, relative battlefields etc) or if we can improve things a lot (by getting hit chances / components / pilots to work).
Perhaps if I spent a few more months working on the turn-based system we'd make some kind of breakthrough ... but the game is now approaching Early Access and we need to make a final decision on what system we want to use because the uncertainty is holding back the strategy layer. It's a pretty simple decision; the X1 realtime system currently works better than the X2 turn-based system does, so we'll be going with the X1 system.
Anyway, I'm sure some people are going to be very happy with this change and I'm sure some other people will be a bit disappointed. I can understand both viewpoints, but really the most important thing here is that a decision has been made and in the next build we should be able to start balancing and properly playing the strategy layer. Hopefully that at least is something everyone can get excited about!
I kicked it off my must-buy list for now. Phoenix Point looks much more interesting all of the sudden.
