Aurora 4x
C# Aurora => Spoilers => Topic started by: Steve Walmsley on September 05, 2022, 03:22:20 AM
-
SPOILER WARNING!!
I am taking this from the bug thread into a new thread. You could argue raiders are unbalanced (explanation below), but it definitely isn't a bug.
Raiders are broken.
Experienced 400k of raider tonnage with ECM10 on year 17 of the game. (normal TS start, research rate at 100%). This is about 7 years after they got triggered. They went from sending in 20k tonnes in ships to 400k in a single attack.
They also have ECM10. Even if I went 100% in on ECM I'd only get to ECCM3 (and my engine tech would be 0)
I looked at DB, and they have a lot of ships (400k t+), but they don't have a lot of killed ships.
Edit: Reddit thread on same issue. Seems at least a couple more people hit sudden huge ways of raiders at 15-20 years
ECM 10 is the first level of ECM, giving a 10% penalty to chance-to-hit. Was that 400K including non-warships, such as troop transports or salvage ships? Also, the raiders continue to build ships from game start. The system trigger is when they first attack.
OK dove into DB some more. By year 17 of the game they have 32 ships * 10k tonnage + 16 * 10k + 20k * 4 = 560k military tonnage. This military costs (assuming build cost =BP cost) 128k BP. They also have commercial ships but I am not counting any of them (5 slave ships, 3 salvager, 2 troop transport).
It looks like (from hull numbers) that they actually had 1/3 more ships at some point that were destroyed. (me taking out couple raider ships here and there over the past 7 years).
It's just a really huge disconnect of fighting off 2-3 ships at a time (thinning their numbers) to suddenly having 400k military at home star and 100k+ on colony.
And you are right, ECM10 is nothing. It was the cloaking that made it difficult to use missiles on some ships.
While I understand you are unhappy about the number of raiders, this isn't a bug. I'll reply here so if you want to continue this debate, it won't clutter the bugs thread.
Raider forces are spawned using the following chance:
GameTimeMultiplier = ((GameTime / GlobalValues.SECONDSPERYEAR) / 20) + 1; (effectively 2x normal rate after 20 years and 4x normal rate after 60 years)
OGChance = (int) Math.Floor(10000 * ROI * (double)GameTimeMultiplier); (ROI is the portion of a year that passed in the construction phase)
RN = GlobalValues.RandomNumber(10000);
if (RN <= OGChance)...
Generate Operational Groups
So about 1 new operational group a year will be spawned at the start but about 2 formations a year after 20 years, etc.
When a new group is formed, the raiders choose the next from a list of operational group types. When the list expires, they start it again.
Lone Raider
Salvage Group (Salvage, 2-4 raiders, 1-3 escorts)
Slave Transport Group (Transport, 1 escort)
Raider Squadron (2-3 raiders, 1 escort)
Troop Transport Group (Troop Transport, 1 raider, 1 escort)
Lone Raider
Salvage Group
Slave Transport Group
Lone Raider
Troop Transport Group
Raider Squadron
Lone Raider
Slave Transport Group
Raiding Fleet (1-2 large raiders, 2-4 raiders, 1-2 escorts)
There are several factors influencing the number of ships. Obviously time passed is a major factor, as is the influence that time has on the multiplier plus the annual chance of generation. Another is the size of the randomised groups and the number of large groups in the list.
One obvious thing that stands out is the size of the Salvage group. Cutting that down to 1 escort, 1 raider would reduce the list by a few ships.
Opening this thread up to comments regarding the spawn rate and how it could be improved - or is it fine now?
-
The only issue I have with them is they only grow from a time factor, instead of the way an NPR takes into account your empire size. I have a game where the growth has been very slow (due to game settings), my empire only reached the 10th system mark at around 200 years in. Suddenly I had the raiders turn up with around 3000 ships, in about 600mt worth.
The issue this presents is you cannot have them active in such a game unless you build huge military forces up, but this has zero RP reason in such a game.
If the raiders had their starting numbers based on your empire size and military tonnage, and then their future growth based on the current time development model, this would allow for all play types. You wouldn't need to build X amount of ships just before system 10 for an artifical meta reason (that has to be done otherwise you will simply get wiped out), but also after they turn up you will have an in game reason to have to then maintain a patrolling force.
-
If you have raiders disabled and turn them on mid-game, will they start out with nothing and generate their yearly fleets from that point on with the normal spawn chance, or will they instantly receive a bunch of fleets, depending on how late during the game you've turned them on?
If turning them on late prevents them from stockpiling huge amounts of fleets during the time period where the player hasn't explored enough systems to trigger their raids (default 10), maybe that should actually be the default behavior?
Maybe the raiders should only start generating ships once you've explored enough systems to trigger their raids.
As far as I can tell, the complaints about overwhelming raider forces come from players who probably took a long time to explore the 10 systems and then suddenly get overwhelmed by a huge invasion force of raiders who spent the whole time accumulating forces.
-
Another variable to this might be how the raiders interact with NPRs.
Does the ships built up get divide among all systems with active NPRs/players?
If it does, having more NPRs in the game will make the raiders considerably easier to deal with.
In my current game I started with 5 NPRs, and in 25 years (130 systems explored) have only seen smaller raider forces in a few systems.
Another question a bit off topic, will swarm, invaders and raiders fight each other?
-
Rather than using GameTime to determine their strength, maybe use time since the raiders were triggered (ie when 10 systems or whatever is reached) as the basis for raider strength?
That should give much more consistent experiences between a standard TN start and some 15% tech, limited lab capacity, conventional start where you don't drag yourself out of Sol for a century :)
-
I think the counter is fine, as it is better to give a chance for build up to raiders so that they can be challenging, however, I think it should be possible to tweak the growth to be in line with the kind of play we are facing. Conventional, Slow Play, Fast Play, Custom Fully Developed RP Race.
Following Steve's post seems like there are a couple of variables in the formula already.
I wonder if we can set a Date on when the counter starts instead. So you'll have 100 years start build up and 10 systems to appear for instance.
-
I have been quite enjoying the raiders the way they are, the last attack was a single scout at an outpost colony, luckily I had forces on hand which could deal with it because I have been forced to take security seriously, which is what you get if you tick raiders imho. To me it means you still get combat experience, even if you play the peacemaker with NPRs.
The raid before that was 6x groups of 6 ships at Earth i.e. the 400kt worth which is no exaggeration but was defeated because I had prepared carefully and had gone for overkill after an earlier salvage group + scout raid (using player systems 1) which demolished all orbital structures. Was very satisfying to get payback against such a credible force.
It seems doable if you take the threat seriously, though probably not for a first playthrough, they do require a major commitment in resource / tech / manufacturing and management and create havoc if they break through your defences but that is what you want in a scourge class spoiler. First time around it does leave you feeling a little put upon! If I had not read Steve's description of them as a scaling threat I would not have been as well prepared tbh and it would have been overwhelming. But if you rise to the challenge it can be overcome.
I am happy with them the way they are but considering the other side of the discussion leaves me wondering what Steve thinks about the idea of an intermediate class of intrusive spoiler, which acts as a pirate scavenger. i.e. comes from the known universe, steals your stuff but presents forces on a smaller scale. Just thinking out loud.
-
Another variable to this might be how the raiders interact with NPRs.
Does the ships built up get divide among all systems with active NPRs/players?
If it does, having more NPRs in the game will make the raiders considerably easier to deal with.
It’s my understanding that they pop into systems randomly, so if there are more systems then it does seem like you would see them less often. Perhaps people habitually turn off the starting NPRs so that the game will be easier, but that inadvertently makes the raiders more difficult. That’s a really interesting game design idea: a self–balancing game where the options only shift what type of difficulty you get.
In my game I have seen neither hide nor hair of the raiders, though I did lose some geo survey ships to rakhas and the remnants. But now that I think about it, I might have had the bright idea to set them to start after discovering 20 systems. I started this game last weekend, and now I am not sure… There have been some slow downs for enemy action, and the same remnants that got one of my exploration fleet also got two alien ships as well. I guess I will see soon enough.
Personally, I think that these types of game design questions are easier to answer if you have more information than is available in a typical forum post: starting conditions, research choices, ship designs, etc, etc. Luckily we have some kind folks who record their whole game and post it to Youtube; , for example.
This one is definitely worth watching because he has really been pushing himself to be ready for the raiders, and it looks like he failed. At the very least he will suffer significant losses. After watching it, I think the reason is fairly obvious.
He settled four extra–solar colonies then garrisoned them with ground troops and a fighter wing. He moved people, maintenance facilities, and MSP to those colonies to support the fighters. It seemed like he was doing everything right, and certainly he put far more effort into his defenses than I ever did prior to 2.0.
When he discovered his tenth system (about 23 minutes in to the episode, and 21 years into the game (https://www.youtube.com/watch?v=m0OwEgTdKT8)), he was probed in four systems in fairly short order. He didn’t lose the exploration ship in the tenth system, because they all carry three fighters with them and that was enough to scare off the single raider ship he encountered in that system. I like that design more than the larger carriers with fighters and survey craft that I decided to go with. The second probe was also just a single ship raider, this time going up against a dozen fighters based at a colony. The fighters were able to destroy the raider, though he did lose about 10% of the fighters. The third probe went badly because he had no deep–space tracking systems in place, so an unknown number of raiders simply flew undetected up to his fighter base and destroyed it. Oops.
The final probe, however, revealed the real flaw in his plan. More raiders showed up back in the same system where he easily destroyed a lone raider ship, but this time it was a sizable force. Multiple groups of ships, some with civilian engines and all of them escorted. His fighter squadron was quickly destroyed fighting a group of three ships. Troop transports arrived a few days later, and his colony was invaded. His 50kt garrison was lost in the first combat cycle and the colony fell.
So what went wrong?
His forces were too concentrated. He had a handful of carriers each armed with dozens of fighters, but they were mostly stationed in Sol at the time. That put them practically right next door to the fighting, but also 33 days away. They simply could not arrive at any of his extra–solar colonies in time to prevent an invasion; they would have to already be in place when that main fleet arrives in order to do any good. It probably would have taken at least half of his carriers to be certain of success against the three or four raider groups that we saw, and while you can put half of your forces in two places at once, you can’t put them in four places.
He also didn’t expand his ground forces very much. He had a lot more of them on Earth, but hadn’t moved any more out to the colonies. Given the raider’s tech advantages he might not have been able to save his colony, but he might have had enough troops to get a kill or two. Worse, he had no STOs. I think all of his weapons are railguns, and he was perhaps waiting until they had more range and stopping power, but some fortified STOs might have put some holes in some ships.
I think the obvious conclusion is that he should have limited himself to just one extra–solar colony. His defense stations should have been larger, even though that means needing a larger population to support them. 36 of his fighters would fare a lot better against a raider group than the 18 he was using, so that is the minimum he should use. He would have been able to split his carriers between Sol and his colony, to drive off any large force that attacked either one. After killing some raiders he would be able to start a third colony.
Changing the game settings might also be a good idea. He could start with 4× population, like Steve did in his most recent AAR. That would give him more forces at the start of the game, and more exploration ships to survey those first 10 systems more quickly. I don’t know if he had any NPRs enabled at game start, that might have helped as punchkid surmises. He could have set the raiders to attack after the 5th system is explored instead of the 10th; they would have had half the forces and he would have had half as many colonies to protect.
What lessons can I learn for my game? Well, I have colonies in both Proxima and Alpha Centauri, and although I was planning to start another that is probably plenty for now. My ground forces are larger than Defran’s, but I too need to actually move more significant forces to my colonies. I have gotten a few weapon and armor techs and my latest armor designs have significantly more stopping power than my initial forces. My anchorages are already twice as large as his. I made them as large as my carriers, with 43 fighters each plus an assault shuttle full of very bored Space Marines and a scout. Unfortunately I actually have _two_ colonies in Alpha Centauri; Alpha Centauri-B II’s gallicite deposits were just too tempting. I have no anchorage there yet, but even worse I have no fighters to put in it and gallicite is very tight. All of my existing fighters are at other anchorages or in carriers. And with just barely enough gallicite coming in for making maintenance supplies, I haven’t upgraded many of my existing fighters. I have half a squadron of my third–generation fighter (50% faster than the first generation), two squadrons of the second generation (40% faster than the original), and 7 squadrons of the originals. It seems a bit thin when divided amongst three systems. Maybe I shouldn’t have upgraded any of them, and instead used that gallicite to build new fighters and carriers to carry them. I guess we will see. Also, I see that Earth is about to have 78 million very bored miners unless I find some other place to put all these mines. And the miners.
That turned out to be rather longer than I expected. Once sentence summary, in case your eyes glazed over: there are obvious ways to get ready for the raiders, but it is also very easy to spread yourself too thin.
-
Another variable to this might be how the raiders interact with NPRs.
Does the ships built up get divide among all systems with active NPRs/players?
If it does, having more NPRs in the game will make the raiders considerably easier to deal with.
In my current game I started with 5 NPRs, and in 25 years (130 systems explored) have only seen smaller raider forces in a few systems.
Another question a bit off topic, will swarm, invaders and raiders fight each other?
The raiders will pick any system and don't differentiate between players and NPRs (assuming that starting option is selected), so more NPRs will generally make raiders less of a threat in general. The main issue is slow starts with minimal NPRs.
-
When a new group is formed, the raiders choose the next from a list of operational group types. When the list expires, they start it again.
Lone Raider
Salvage Group (Salvage, 2-4 raiders, 1-3 escorts)
Slave Transport Group (Transport, 1 escort)
Raider Squadron (2-3 raiders, 1 escort)
Troop Transport Group (Troop Transport, 1 raider, 1 escort)
Lone Raider
Salvage Group
Slave Transport Group
Lone Raider
Troop Transport Group
Raider Squadron
Lone Raider
Slave Transport Group
Raiding Fleet (1-2 large raiders, 2-4 raiders, 1-2 escorts)
I wonder if the unhappiness over raider hordes could be alleviated by tying the size of the raider operational groups to raider wealth (similar to civilian shipping lines). This would create a feedback loop: reward players for destroying raids and punish them for letting raids succeed.
- The raiders maintain a pool of ships.
- Instead of a fixed number of ships, raider squadrons/fleets and salvage/slave/troop groups take 20% of the ships available in the pool (alternatively, each ship has a 20% chance of being selected to join the expedition). If there are no ships available, then the expedition is downsized or cancelled.
- If a raid succeeds, then the loot is spent building new ships and adding them to the pool. If the raid was particularly profitable, then the raiders build a battleship (https://wh40k.lexicanum.com/wiki/Dark_Eldar_Space_Fleet#Battleships) to protect their larger fleets.
- The raiders periodically get "free money" (at a rate depending on game time). This prevents players from completely crippling the raiders.
- Each ship in the pool has a 5% chance of retirement each year (an expected service life of 20 years). This prevents excessive raider snowballing.
-
I generally agree with the people suggesting that the raiders should begin scaling up once the number-of-systems threshold is crossed, rather than from game start.
The main reason for this is user experience - basically, the setting is not behaving in the way that players are expecting it to, so there is a strong argument that it should be changed to function as players expect it to. I would note, for any other spoilers it works in this way, albeit accidentally (Swarm and Invaders do not 'develop' or grow prior to being discovered as they generate on discovery, and Precursors remain completely static AFAIK), so perhaps some of the confusion is that Raiders behave very differently from other spoilers with respect to the similar game setting.
Right now, the way to get this "expected" behavior is to turn the Raiders off at game start and turn them on in the game settings at a desired point in time. This is a bit unintuitive, and I thought was the reason why the number-of-systems trigger was added was to avoid having to do this, e.g. for Invaders to prevent discovering an Aether Rift one jump from Sol and dying in five years.
The one thing I think is left uncertain is how to preserve the current behavior, i.e., the Raiders grow from game start but will not appear until the player has expanded a bit. However, I think it is an open question if this behavior is the one many players want to use, I think most players want the "expected" behavior described above (for slow/conventional starts) or they want 100% risk from the start of the game (for traditional/high-pop TN starts), not so many seem to want the in-between?
Changing the game settings might also be a good idea. He could start with 4× population, like Steve did in his most recent AAR. That would give him more forces at the start of the game, and more exploration ships to survey those first 10 systems more quickly. I don’t know if he had any NPRs enabled at game start, that might have helped as punchkid surmises. He could have set the raiders to attack after the 5th system is explored instead of the 10th; they would have had half the forces and he would have had half as many colonies to protect.
I really do think, on a side note, that the default population for game setup should be 1b. When I started playing I quickly found 500m to be very slow and limiting, and with the new spoilers I think this is even more of the case. Of course the player can and usually will set it to whatever they want, but I think it is good to have a more reasonable default to set the expectation for new players.
-
I also second the idea of raiders starting to build up from the moment they are triggered, not from game start. This is a good quick fix.
-
Opening this thread up to comments regarding the spawn rate and how it could be improved - or is it fine now?
I want to say that the raiders raiding is EXTREMELY FUN. They are a great addition to the game. It's like adding sieges to the dwarf fortress. With them the game feels more lively and the universe more alive.
However, the random nature of the spawn means that the player will lose colonies if the dice rolls so. For example in my game let's say I had enough ships to defend Sol. However I can't defend every colony against 400k of raiders. That makes it frustrating as there's no tools to predict their size and where they'll be.
It also makes the game unpredictable -- many people feel the raiders are fine because they got a much smaller raider spawn. So for example I checked reddit/discord and saw maybe 10 people who got obliterated (or a colony) early on (<20 yrs) by large fleets. However, most people don't have that experience on their first gameplay (50+ ppl).
I think the problem is not the average difficulty of the raiders, but rather the outliers. Having too few or too many raiders is not as fun as having average number of raiders. Maybe spawn once a year to reduce variance?
The other thing that would be really cool is objective based spawn. Did they kill a juice freighter? Spawn a bit more. Did they lose their entire group? Spawn less. That would give direct control\destiny to the player.
Another option would be to have them attack\harass with medium strength. Currently they harass with 1-2 ships and then BOOM here's giant army. If they instead harassed 3 years earlier with smaller numbers on some colony that would give a nice transition of what to expect.
-
For v2.1.1, I've changed the raiders to only start generating additional ships after the player reaches the trigger for the number of systems.
The time multiple will still be from game start for now as that would require a DB change. I could tackle that in the next DB update.
I've reduced the disproportionate escort group for salvage ships, but that will not take effect until the next DB update.
As suggested I have also changed default starting population to one billion for players.
-
For v2.1.1, I've changed the raiders to only start generating additional ships after the player reaches the trigger for the number of systems.
The time multiple will still be from game start for now as that would require a DB change. I could tackle that in the next DB update.
I've reduced the disproportionate escort group for salvage ships, but that will not take effect until the next DB update.
As suggested I have also changed default starting population to one billion for players.
Since you've got inbound DB changes maybe it would be a good idea to release 2.1.1 on account the number of fixes and especially the raider teleport fix which I would argue is a pretty big deal.
-
For v2.1.1, I've changed the raiders to only start generating additional ships after the player reaches the trigger for the number of systems.
The time multiple will still be from game start for now as that would require a DB change. I could tackle that in the next DB update.
I've reduced the disproportionate escort group for salvage ships, but that will not take effect until the next DB update.
As suggested I have also changed default starting population to one billion for players.
Since you've got inbound DB changes maybe it would be a good idea to release 2.1.1 on account the number of fixes and especially the raider teleport fix which I would argue is a pretty big deal.
Yes, I am trying to get through the bug list now. Should be out by tomorrow.
-
:o
Did you just… announce a release date‽
surprised pikachu!!!!!1
-
It took 4 tries for the raiders to not show up and exterminate my population before I even got TN technology though admittedly on the 3rd try I managed to build some defences, but then got hit with a raider fleet which dropped 2000 elite infantry on me. None of my weapons could penetrate their armour so that was a painful experience. The first ship showing up blew up my space yard, then a second raider visit destroyed my ground combat facility even though I managed to destroy it with STO fire. But against that fleet even 26 15 cm plasma carronades wasn't anywhere near sufficient. I killed 2 more raider light ships but the heavier ones were too tough to affect before their firepower destroyed the STOs. The 3rd attempt I had just got TN tech but had not really gotten anywhere with it.
My suggestion is that the raiders actually behave like raiders. They land steal wealth, minerals and population (if so inclined) but leave the bulk of the colony intact so they can come back for more. If they encounter say a lone refinery ship or mining ship they make it refuel them or else abscond with the minerals. They pirate civilian ships or take them as prizes etc. As it is now they are exterminating the goose that lays the eggs. This would make them less of a game over event to slower/conventional starts.
My fourth attempt I set them to 20 systems as opposed to 10 as I have frankly nearly explored 10 and am still barely space capable (starting: population 250 m/4 research/1 shipyard/256 CI). At least my new land forces could engage the raider infantry with some chance of success.
-
I think the biggest thing with Raiders is that they are just randomized Invaders. The way the build up and interact with the player and npr races are near identical except Raiders are more likely to run if they are outnumbered and can spawn anywhere and everywhere rather than be limited to Aether Rifts. They are just as capable of wiping out the player as the Invaders, Swarm, or a game start NPR. The Raiders imo should be closer in capabilities and function as Precursors and Rahkas, more of a road block rather than an existential threat.
I would suggest that they don't build up over time at all, but rather every build cycle have a % chance of spawning a ship, few ships, fleet, or multiple fleets. These ships would focus on destroying or capturing (they really should have boarding craft imo) civilian shipping or attacking undefended colonies to steal population or mineral stockpiles. They should avoid contact with warships unless they feel they are significantly stronger than the player's ship(s) or planetary defenses (they really need a speed bump to run more effectively). But most importantly I think that once their ships run out of fuel they move to their initial jump point and get out, closing the door behind them. Whatever they steal in BP value could then be added to their next incursion.
This I think would better represent them as pirates and raiders and keep them in that roadblock level of difficulty as Precursors and Rahkas
-
I was also pretty surprised by the raiders in that I thought they would just be a couple pirate ships that show up rarely, but then people started talking about how their whole homeworld was destroyed by hundreds of raider ships. They really are a lot more like the invaders than just simple pirates. Not that that's a bad thing, since that's also cool to have in game, but it's different.
We'll see how much different they feel now in 2.1.1.
-
Something else about spawn rates which might be worth mentioning is the idea of modifying them by the research multiplier, which also applies to the invader timers which I remember being dicussed elsewhere.
Sometimes people like to play with longer research eras but if they do that then if the scaling spoilers occur at a fixed level of ferocity then they may find they are squashed like a bug.
If the raider spawn rate is X*((1+years)/20) then you could also add a term for the research % i.e. X*((1+years)/20)*research%term
So if the research rate is 20% then that would mean 0.2 times as ferocious a build up. Likewise with the invader timer, it could be divided by the research multiplier so it takes 5x longer.
The only problem with that is you can change the research multiplier on the fly, if you did that after 200 years you might get a nasty surprise! But it would give players a way of regulating the spoiler, with a balancing research penalty built in.
EDIT to fix brackets
-
Hey so I've never said anything on this forum yet, but I wanted to say something about the raiders here.
First I wanted to thank Mr Walmsey for making this game, which scratches a very particular itch of mine, I love it so much.
That said, I do agree the raiders are a little broken.
At the moment, the practical difficulty of the raiders is, really, tied to the number of NPRs in your game (assuming you set it so that NPRs encounter them too, which I imagine most do). The more NPRs, the more places for them to spawn (which aren't in your home system), and the less likely you'll encounter them.
I'm not exactly aware of how difficult this would be, but would it honestly be that hard to just include a difficulty setting during game parameters to control the % chance of a raider spawning?
In a game which already has severe problems with accessibility to new players, I'd imagine something like this would be very important
-
At the moment, the practical difficulty of the raiders is, really, tied to the number of NPRs in your game (assuming you set it so that NPRs encounter them too, which I imagine most do).
I do not, and since it is not the default setting I assume most don't either, though I know Steve does sometimes. Personally I avoid enabling NPR-spoiler encounters as it ends up being another way to slow down the game in a hail of short increments.
Maybe one solution would be to have a number box in the game start settings for a number of raider factions to spawn in. If there are, say, 3 Raider factions instead of 1, Raiders are 3x as likely to spawn, and having them in several factions would mean each set of Raiders has different weapons and ship designs, adding a bit of unpredictability.
That said, this is properly something to post in the Suggestions Thread (https://aurora2.pentarch.org/index.php?topic=13404.msg170088#new) rather than resurrecting a 2-year-old development thread. :)
-
will swarm, invaders and raiders fight each other?
I have seen the wreck remnants of raiders that attacked a precursor planet, including a commercial ship.