June 05, 2020

Aurora Tutorial Part 9

As we make our way through Aurora, we've finally reached the point where we can talk about offensive missile warfare. Because offensive missiles require an understanding of how to defend against missiles, I'm going to be relying heavily on the previous parts about defending against missiles with beams and anti-missiles.

The first step to take when using missiles is to decide on a standard size for your offensive missiles. It's possible to run multiple sizes on different types of ships, but I'd recommend against it. That way lies micromanagement madness. On the whole, I'd recommend sticking to missile sizes between 4 and 6. Any missile of size 6 or smaller is picked up by active sensors and fire control at the same point, while larger missiles are picked up further out, so that as a cap maximizes your ability to penetrate defenses. Going smaller means more missiles per salvo, and a shorter interval between salvoes, but less-capable individual missiles. Anything smaller than Size 4 is likely to pay too high a price in capability.

The next thing to establish is your range for the first generation of missiles, which is most likely to be limited by your fire-control tech. The C# sensor model is sharply non-linear, so the VB6 trick of building enormous fire controls and dominating half of a star system doesn't really work any more. You're going to want more than one fire control per ship (redundancy and salvo size) so it's probably best to keep the size down in the 1-2 range. I'd keep the fire control at R100, as most warships tend to be bigger than that, although I sometimes fit a backup R20 FC to make sure my ships won't get blindsided by small craft.

At this point, we have nearly all of the factors required to do the final design of our missile. The only thing remaining is to decide on the size of the warhead. For my primary missile, I generally fit a warhead that takes up somewhere between 25% and 40% of the weapon. Because of the way that the game's damage model works (the wiki has more details if you're interested), you'll get the best armor penetration with warheads of 4, 9, 16 and so on, although this often has to be compromised because of space limitations. Once you know this, you can start playing with combinations of fuel, engine size and multiplier, and agility to get the most effective missile for your desired range and warhead. The best option here is probably to pick the target speed (your own fleet's standard speed is a pretty good option here) and try to get 100% accuracy on ships of that speed, while meeting your range target and maximizing speed. Or you can download a Java app that will do it for you. If the missile is too slow, then shrink the warhead or the range and try again. Even if you use the Java app, I'd still recommend that you play with the missile designer yourself to get an intuitive sense of the tradeoffs involved.

So now that we have the missiles, we need ships to fire them. This is pretty much like designing any other warship. The only point of special note is that I generally go with one FC per 6-8 launchers, to exploit some of the limits on AMM fire control. And pay attention to how many salvoes the magazines can hold. There's not a firm value to aim for, but missile combat can run through a lot of weapons very quickly, so if you only have three salvoes, then you should probably remove some launchers and fit more magazines. Make sure you have a decent active sensor, although (particularly if you're pushing for really long range) it could make sense to only fit the big actives to a few ships. Offensive missiles can be fired from reduced-size launchers, which allow you to cram a bigger salvo into a ship, at the cost of reduced ROF.1 In practical terms, a single massive salvo is more effective against a fleet defended primarily with beam weapons, but less effective against one using AMMs. Unfortunately, you're not likely to get this information until it's too late to be of any use, but both can work well, so the choice is up to you.

The other thing you should do as soon as you start a missile fleet is to start building missiles. These are built by ordnance factories on the Industry screen, switching from Construction to Ordnance in the dropdown in the upper left. You're going to need a lot of missiles, and while you'll undoubtedly research better ones in time, the outdated stockpiles are valuable if you run through the newer ones. The downside of missile warfare is that managing the logistics of the missiles can turn into a major headache on both the cost and micromanagement fronts.

One of the great advantages of missiles is their flexibility, and it can be a good idea to build specialized missiles for certain purposes. The two I use most often are missiles intended for high-speed targets like fighters and heavy-warhead missiles for killing things that are already disabled. The problem with these is that they take up magazine space and add more micromanagement, and I go back and forth on their utility, even over the course of a given game. But I'd encourage you to experiment and see what you can come up with.2

As your technology advances, you'll be able to build better missiles. Researching a new missile is relatively cheap, so the only reasons not to upgrade every time you get a new tech that boosts missile performance are (a) you're going to get more new stuff very soon, and want to incorporate that, and (b) logistics, because you do want to produce enough to be able to keep your fleet more or less on a common standard, instead of spending a bunch of time figuring out your loadouts. Remember that the point of missile warfare is to have all of your weapons arrive on target at the same time, so having some ships with upgraded (faster) missiles and some without can make your fleet significantly less effective than if all of them were using the older missiles. This isn't to say you should never upgrade your missiles, of course, but there will be a transition period.

The last thing a missile fleet needs is some plan for resupply. If you're just venturing out into the cosmos for the first time, then sending your ships home to reload makes sense. If you want to pursue a more aggressive policy, then you'll need to reload in forward areas. Transferring ordnance from a colony requires either a spaceport or an ordnance transfer station, while an ordnance transfer system is required if you want to get it from another ship. Fortunately, you can build an ammunition ship at a commercial yard, and this is probably the approach to take. Just remember that civilian magazines have a 100% chance of exploding if destroyed, so it's probably best to keep it well back.

At this point, you should be able to make an informed choice between beams and missiles for your fleet, and can start building your actual warships. It will probably be some time before you actually run into an enemy, and said enemy can range from a small and easily-overwhelmed power to one of the spoiler races,3 who can make mincemeat of even a good early-game fleet. These can be turned off in game options, but to some extent, it's all part of the challenge.

The last four posts have been a detailed look at the basic pieces of Aurora combat, and with that background it's time to zoom out again. There are a number of systems I've ignored so far, and it's time to look at how they interact with both individual ships and the fleet as a whole. We'll pick up there next time.

1 Do note that the smallest launchers, box launchers, can only be reloaded in a hangar bay or at a maintenance facility. While fitting your ships with a VLS sounds cool, I'd recommend against it for a major warship in Aurora.

2 The best semi-hack I've come up with is decoys. These are size-1 missiles designed to be very slightly faster than the regular missiles, launched from somewhere outside missile range. They would soak up a bunch of the enemy's defensive fire, although changes to beam FC in C# have made them less effective. Also, at the time my missiles were based entirely around size-1 penetrators bundled into pods for range, because I was in the last stages of an arms race with myself on missile defense/attack, and had the tech to pull that off. That's not to say that decoys couldn't be useful, particularly because different types of missiles group into different salvos, so a regular attack with a few decoys in it will have three different salvos instead of one. But there's a significant increase in micromanagement to go along with it.

3 There are three races, the invaders, precursors, and star swarm, who are all very powerful and tied into the game's rather loose mythology. More details can be found on the Aurora forums if you're interested.


  1. June 06, 2020John Schilling said...

    Why are large missile salvos less effective against targets defended against AMMs? It would seem to me that if the AMMs are limited by fire control or launcher ROF, large missile salvos would overwhelm them in the same way they overwhelm beam defenses. If the AMMs are not so constrained, then in the extreme case it doesn't matter whether attacking missiles are launched in one salvo or many, but there should never be a case where launching one large salvo would be less effective that many smaller ones.

    Also, why are people using these silly "warhead" things? At reasonable space-combat velocities, using the same mass as propellant to accelerate the missile faster is going to add about as much impact energy as the warhead would provide, while reducing time of flight and making the missile harder to intercept.

  2. June 06, 2020Alexander said...

    Apparently the game doesn't use Newtonian physics (ships use fuel to keep moving, not slow down) so perhaps kinetic weapons don't perform as expected. Though rail guns and Gauss weapons exist, so I suppose there must be some 'impact' from a chunk of metal slamming into you at high velocity.

    Maybe part of the issue with large salvos is retargeting AMMs if their first target is destroyed before they arrive. Since they don't have to cancel their velocity, they can instantly attack another missile in the same salvo, but wouldn't have the endurance to wait for the next wave. Largely speculating, since I've only read about the game, not actually played it.

  3. June 06, 2020AlexT said...

    there should never be a case where launching one large salvo would be less effective that many smaller ones

    Wouldn't multiple small salvos from different directions mess with interceptor retargeting? The more distinct the trajectories of the small salvos, the harder it would be to redirect interceptors between them, as opposed to between individual missiles in the same salvo.

  4. June 06, 2020bean said...

    Basically, it's a matter of how the system is most easily overwhelmed. Yes, you can absolutely throw a big enough single salvo to beat AMMs, but it's probably going to need to be a bigger salvo than for an equivalent tech in beams. Worked example: the enemy AMMs have a minute to shoot between the time you enter their range and the time your missiles reach them. We'll say they've got good tech, and each AMM launcher can fire 12 missiles during that time with a 33% PK. That means each AMM tube will take out 4 missiles, and you're going to need a huge edge in launcher numbers.

    But what about a second salvo? We fire a second salvo 15 seconds after the first. (All of the numbers are guesses, but reasonable ones in my experience.) Now, they have to stop both salvoes in a total of 75 seconds (15 increments), which means that each incoming salvo faces only 7.5 shots per AMM tube instead of 12. Now, you can overwhelm them with 2.5 launchers per AMM launcher. Fire a third salvo, 15 seconds after the second. Now, you've got 18 shots split across 3 salvos, and you're down to an effective ratio of 2 tubes per AMM, and so on as you continue to fire more shots. The ultimate limit case in this particular scenario is obviously 1:1 if you have enough salvoes stacked. That sort of thing can get very expensive, but so long as their engagement time is longer than your reload interval, you can wear them down. If it's not, the missile defenses are like beams, and you just have to kill them all at once. But if their missiles are like that, you have nothing to worry about anyway.

    In terms of ship design, the smallest non-box launcher you can get is 0.3 size, x100 reload. That's a significant increase in throw weight, but you're firing one salvo every 25 minutes or so in exchange, which means a lot less firepower in total.

    As for warheads, the answer is that this is all Trans-Newtonian, and objects in Trans-Newtonian space don't do nearly as much damage during a collision as objects in the universe we live in.

  5. June 06, 2020bean said...

    In terms of salvoes, I've been kind of sloppy with my terminology. There's in-game salvoes which are a bunch of missiles of the same type, fired by launchers linked to the same FC. They move together and are targeted together. In that case, more small ones is better, because of targeting, as discussed last time. But there's also "how many missiles am I throwing at once?", which is what I was talking to John about. In that case, yes, the same number of missiles will always be more efficient all at once rather than spread out. The tradeoff there is that you're going to sacrifice total missiles over time to get that, and when facing an AMM-heavy defense, that's probably not a good trade.


    The AMMs can only hit targets in the same FC salvo. If a missile's target is destroyed and it doesn't have onboard sensors, it self-destructs. Most of the time, onboard sensors aren't worth it. You definitely can't retarget.

  6. June 06, 2020echo said...

    So I'm trying to plan active sensor coverage, and oh god help.

    For area search (and missile early warning) big passive sensors seem like the way to go, if you don't care about knowing exactly what's there. Do you really need active sensor range much beyond your FC range? If you do, should you put it on some sort of AWACs cruiser so the huge search radar doesn't light up your battlefleet?

  7. June 06, 2020bean said...

    I usually match active sensor and FC range for missile forces. For beam ships, I go with mostly R1 sensors, but usually will include some sort of medium-size (maybe 1 HS) R100 or so system to make sure I have good tracks on any ships I’m facing. You’ll probably pick up the big ships, or anything radiating, at long range, but I often find smaller ships (5-7,000 tons) showing up first when they enter my active range.

    The game doesn't model anything like bistatic radar. As for AWACS ships, given the sensor model changes (a size 50 sensor is about 4x the range of a size 3 now, where it used to be linear), there's not a lot of point in building ships with huge actives. It wouldn't be a terrible plan use FACs or fighters to push the sensor perimeter out, but the problem is that we don't have the tools for it any more either. Hopefully, Steve will restore them to us in a future update.

  8. June 07, 2020bean said...

    I should add that freaking out over your sensors isn't necessary, because the AI isn't sophisticated enough to need it or appreciate it. If they see you in their system, they'll probably dispatch ships to meet you. Some will be radiating, and they definitely won't do anything like trying to trick you into deploying on one threat axis, and then attacking on another. Nor can you do the same to them. For that matter, there's not really any reason to deploy on a threat axis. You might be able to get a few extra percent out of it, but it's not worth the micromanagement. I only ever used the formation tools in VB6 when I was feeling fancy, and it never really paid off. If there was multiplayer, things might be different, but there isn't.

  9. June 07, 2020echo said...

    Thanks, this game is hell for overthinking things.

    Also, is it just me, or is some of the advice on the wiki just bad?
    The point defense article suggests layering beam pd in a way that seems to make it all useless against any fast-moving missile.

  10. June 07, 2020bean said...

    Checking the history, that page is grossly out of date. The last edit was in 2016, and it's mostly unchanged since 2010, which is when I started playing. There's been some substantial changes to the engine model since then, which I think (this was in 2013, so it's been a while) have made missiles rather faster. That's a major hit to the utility of area-defense ships.

    It's also possible that whoever wrote it had managed to find an equilibrium that felt good and worked well enough, and hadn't spent enough time fighting themselves to work out the problems. A lot of my stuff dates back to the days when I was a bored college student furiously competing with myself to squeeze every last drop of efficiency out of my defenses. But Aurora is (weirdly) forgiving enough you can do a reasonable job without that kind of thing.

  11. June 07, 2020echo said...

    It'll take a while to learn a game making all my own spreadsheets at an hour or two a week, but it's a lot of fun anyway. And a whole lot easier with this walkthrough, which seems more optimized than any of the youtube videos I could find.

    One thing that's been bugging me: you can almost never see the year in "Completion Date". It usually looks like "Thusday, December..."
    This is not exactly optimal.

  12. June 07, 2020bean said...

    I suspect that youtube isn't the best medium for this. It's fine for the early bits, but if you're doing design philosophy, it's just going to restrict you. That's why I haven't given specific instructions since about Part 5.

    As for the dates, Aurora uses the default Windows date display. Check the forums for how to change that.

  13. June 07, 2020echo said...

    So I was getting ready to design a jump drive, but got a "Cannot design component without all required technology" error. Do I need efficiency 4 and the squadron size researches to actually make a jump drive?

  14. June 08, 2020bean said...

    You'll need both of those, and the squadron jump radius tech, too. Jump Point Theory is just "we know what jump points are". You need the tech for finding them and making jump drives separately.

  15. June 29, 2020huscarl105 said...

    One thing that worked well for me in VB6 was multi-drive loiter missiles. I expect using them in c# will be less effective and more difficult but I'm, almost at the tech where I can try it. Basically a size 5 missile with an additional 1 hs 'booster stage'. The speed of the booster is the same as you missile ships's max speed (or slightly under). You fire and move toward your target, adding another salvo ever 20 or 30 seconds. The booster are good for 10m km and then they release their payloads somewhere 20% or so inside the second stages max range, at whatever your fleet missile speed is. It is difficult to time this on an unknown opponent, they could reverse a course and the salvo will be lost. They are absolutely deadly vs bases, slow ships, and opposing ships that have you have good reason to suspect will continue to close. I expect the improved AI will make them more challenging to use this time.

  16. June 30, 2020bean said...

    That's quite a good idea. I don't see any reason that wouldn't work in C#, too. The enemy tactical AI hasn't changed all that much. You're more likely to be thwarted by the changes in missiles and sensor ranges.

  17. August 05, 2020huscarl105 said...

    Swarm Spoiler info.. .. .. .. .. .. .. .. .. .. Well it worked, but it was difficult to arrange. I managed to get an 800 missile time on target strike vs a swarm mothership (the 200,000 ton size). It only managed to shoot down about 350 of them. Sadly I thin the code is doing something wonky. After a few failed experiments and lots of missiles wasted, it looks to me that the code determines an intercept point for the missiles and their target and the launched the sub-munitions at the designated distance from that point. This can be much further away from the target at time of launch and is dependent on the course and speed of the target. Fortunately the swarm can be led around by a fast scout pretty easily. I glad this worked. I could not have taken out that ship or its escorts with anything else I had.

Comments from SlateStarCodex:

Leave a comment

All comments are reviewed before being displayed.

Name (required):

E-mail (required, will not be published):


You can use Markdown in comments!

Enter value: Captcha