• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

Stellaris Dev Diary #111 - Anomaly Rework & Expanded Exploration

Hello everyone and welcome to another Stellaris development diary. Today, we're going to start talking about the 2.1 'Niven' update, which will be the next major update after 2.0. At this point I cannot give you any details on the exact nature of the update or when it's arriving, but I *can* talk about some changes we're making and new features we're introducing in regards to exploration, galaxy generation and anomalies.

Anomaly Changes
In 2.1, we're changing the way anomalies work in a few ways. First and foremost, we are removing the concept of failure risk - we found that the possibility to fail on anomalies added little to the game in terms of interesting choices, and mostly frustrated players or made them wait with researching said anomalies until their chance of success was maximized. As such, instead of making it so that anomalies have a failure risk based on scientist skill level, we've instead made it so that the time it takes to research an anomaly is heavily dependent on the scientist skill versus the level of the anomaly - researching a level 2 anomaly with a level 2 scientist will be a comparatively quick affair, while attempting a level 10 anomaly with the same scientist can take a very, very long time, and might mean that it is better to return to it later with a more skilled scientist, so not to hold up your early exploration.
2018_04_19_2.png

(Note: Not final numbers, etc)

As part of this we've also added an anomaly tracker tab to the situation log. The anomaly tracker will keep track of anomalies that you have discovered but not yet researched and easily let find and you return to them.
2018_04_19_1.png


Hyperlane Generation
Another thing that is changing in 2.1 is the way the Hyperlane network is generated. Rather than simply attempting to connect stars to nearby stars, we've created a new generation algorithm that builds up 'clusters' of stars with a high degree of internal connectivity, that are connected to each other by thinner 'highways' which form natural chokepoints. These chokepoints are also registered as such by the game, allowing us to find actual chokepoint systems and avoid placing Leviathans and other powerful space monsters there, as well as improving the AI's ability to detect suitable spots for defensive starbses. The hyperlane connectivity setting will determine the level of connectivity between clusters, and thus how frequent and easily circumvented chokepoints are.
2018_04_19_3.png

(Note: Image is not final. We're still working on the algorithm)

As part of this it will now also be possible for modders to easily generate systems and clusters of systems that are not connected to the main hyperlane network.

New Stars & Systems
Lastly for today, we've added a bunch of new system and star types to the game. First out are binary and trinary star systems - systems containing more than a single star. These systems come in a variety of configurations, and will often contain more planets and resources than conventional, single-star systems. We've also added some new star types to the game in the form of Brown Dwarves (not technically stars, I know) and Class M red super-giants. We've also made it possible to generate more than a single asteroid belt in a system, and created some new mineral-rich asteroid-heavy systems. Finally, there are some new unique systems to find with large amounts of resources in them, guarded by powerful space creatures.
2018_04_19_5.png

2018_04_19_6.png

2018_04_19_4.png


That's all for today! Next week we're going to be talking about something just a little bit mysterious called the L-Cluster... see you then!
 
just expressing some random thoughts

a lot of people are of the opinion that anomaly fail risk is not interesting,
but to me the fail risk is an important part of how i percieve anomalies, the fail risk is its identity
the only thing thats sets it apart from all the other projects that are solely time-based.
while interconnected projects that are important for a bigger picture are the projects without fail risk.

i came up with a midway solution, have the research time be based around scientist skill-level like planned at the moment,
but have a constant failure risk that is tied to the anomaly itself, in other words eacht anomaly has a standard fail risk that doesnt change over the course of the game.

it might satisfy both sides of the conflict
Then it's not RNG, it's just stupid.
 
just expressing some random thoughts

a lot of people are of the opinion that anomaly fail risk is not interesting,
but to me the fail risk is an important part of how i percieve anomalies, the fail risk is its identity
the only thing thats sets it apart from all the other projects that are solely time-based.
while interconnected projects that are important for a bigger picture are the projects without fail risk.

i came up with a midway solution, have the research time be based around scientist skill-level like planned at the moment,
but have a constant failure risk that is tied to the anomaly itself, in other words eacht anomaly has a standard fail risk that doesnt change over the course of the game.

it might satisfy both sides of the conflict

That's even worse. High level anomalies would be a 50/50 chance or worse. I don't think either side would be happy with that solution.

Everyone who isn't being reckless to "roleplay" waits for the 5%-10% anomaly risk chance anyways. Anyone who disagrees with that is simply being dishonest to push a point they themselves don't fully experience.
 
I'd like to go so far as to ask if the Hyperlane Registrar Starbase Module could be used to affect a number of systems within range of the starbase, and not JUST it's system, so we can manufacture highways? Maybe something equal to the Sensor Range of the station, so it improves with time, and a Listening Post would let you boost it even further. Plus, tack on a second module that would link the starbases together that have the prerequisites for it, so it doesn't just boost it out in ALL directions from a given system. That way you have value in making a string of starbases across your space, letting you redeploy much faster to a flank in a defensive war. Just a way to build infrastructure in the earlier game before you can start dropping Gateways all over the place, which have their own inherit flaws when dealing with endgame crisis.

Actually, speaking of that... could we get a rare tech, once you can build Gateways, to install some kind of inhibitor on them, to make it so said crisis either cannot enter our space through them, or at least gets penalized for doing so? I mean, if the Eldar were able to cut parts of the Webway off from the Necron using Dolan Gates to hop into it, why can't we do the same?
Wait what? I thought hostiles couldn't travel through gateways in your space at all...
 
Love the idea of long starlanes and some systems or clusters being off on their own.

But having this as a general map feature, isn't there a worry that this will pull every war back to doomstacks? Judging from the image posted (which, 100% acknowledged is not final), every empire on the map can defend themselves with one or two systems.

Won't every every empire just stack maximum defenses in those systems, leading every war to become a question of who can break through the other guy's wall? Chokepoints can be a great element to strategy, but I worry that this doubles down on them as essentially the entire game in and of itself. Without other viable options we're basically back where we were before. Fortifying a choke point, then throwing the biggest possible fleet at the enemy's chokepoint will become the dominant (if not only) move.

Also, not to be snarky, but isn't this almost exactly what Endless Space 2 did with its constellations?
If you intentionally limit your empire to one of those clusters, you will be able to defend only two systems. If you try and expand in any direction from those clusters you will rapidly find yourself with much more terrain to defend.
 
a lot of people are of the opinion that anomaly fail risk is not interesting,
but to me the fail risk is an important part of how i percieve anomalies, the fail risk is its identity

The fail risk is really just an extra timer before you have enough skill levels and -risk modifiers to get it down to an acceptable level.

It's not an "identity" at all, because you can reduce them all to 5% anyway without even a lot of effort.
 
Fine.... AND it's mostly pointless.

Edit: Wait a minute... You only quoted the part of my comment that let you do that. Read the whole thing.
 
Last edited:
Everyone who isn't being reckless to "roleplay" waits for the 5%-10% anomaly risk chance anyways. Anyone who disagrees with that is simply being dishonest to push a point they themselves don't fully experience.
I don't worry about RP at all, and I just try every anomaly I stumble across regardless of odds because I think it's fun to do them as I discover them.
 
well.... I like the new star systems thing and anomaly tracker.... not a fan of the changes to anomalies or the inane insistence on 'muh choke points' but I still despise the hyperlane only change as well
 
Cheers for the DD Wiz :). Lots of interesting changes, and like the sound of the 'clusters' and chokepoints. Removing the possibility of getting the same anomaly twice for a particular empire, particularly given the changes, a very good move :). Some thoughts:

First and foremost, we are removing the concept of failure risk - we found that the possibility to fail on anomalies added little to the game in terms of interesting choices

Randomization has no inherent value. It only has value if it creates interesting gameplay outcomes or choices. Anomaly fail risk did not.

These anomaly changes, though, don't actually change the decision being made in most cases (high level anomaly will still mean 'wait until I have a high-level scientist' to research). What they do is remove the variation in gameplay outcomes (instead of a chance anomalies could go one way or another (failure/success), leading to a subtly different narrative, the variation now is in how long one waits before something happens - but what happens is no longer varied. The anomaly tracker is a great addition, but it would be just as great an addition for the pre-2.1 system, because the impact on the actual choice is negligible.

In that context, these changes in and of themselves, as best I can tell (and apologies if I have this wrong), leave choices relatively unchanged but reduce the range of interesting gameplay outcomes (including, presumably, the catastrophic failures that lead to scientists getting knocked off - which was usually an interesting gameplay event in my playthroughs at least).

If the goal is to avoid 'unlucky negative results' (noting that players had perfect information as to their chance of being unlucky, so it's not particularly harsh), one way around this without reducing the range of gameplay outcomes might be to have, instead of 'fails', a 'range of anomaly outcomes' - where the 'low end' outcome replaced failure, and didn't result in as negative outcomes? Would involve a bunch of content work at the very least though.

It's not a biggy - but the 'no more anomaly fails' does feel at first glance to 'remove life' from the Stellaris universe, but with a negligible-to-nil improvement in gameplay. Not trying to have a go - am a big fan of the game :). Just throwing ideas around in case they help.

avoid placing Leviathans and other powerful space monsters there, as well as improving the AI's ability to detect suitable spots for defensive starbses.

Having some choke points with monsters/Leviathans could be fun and add colour and variation to a galaxy/playthrough. I'd probably be happier with a few 'interestingly blocked' chokepoints than not - noting I could be a freak so my thoughts here may be silly :).
 
New Stars & Systems
Lastly for today, we've added a bunch of new system and star types to the game. First out are binary and trinary star systems - systems containing more than a single star. These systems come in a variety of configurations, and will often contain more planets and resources than conventional, single-star systems. We've also added some new star types to the game in the form of Brown Dwarves (not technically stars, I know) and Class M red super-giants.
At the moment, in common\random_names\base\00_random_names.txt, you have a specific category for black_hole_names but all star names are in one big list. Could you please make categories for each class of stars, red giants, etc.? Then when the game names a star, it might name it from the name list for stars of that class or from a generic star name list, but we wouldn't get G class stars named after RL red giants, etc.
 
Random and unavoidable 5% chance to fail is not fun. Having your first scientist randomly die the first anomaly your research is not fun. Having a level 5 scientist die while looking at a shiny rock because they pressed the self-destruct button by accident is not compelling gameplay.

However, what might be fun is anomalies that have multiple possible outcomes and different options for how to handle them. Low level anomalies should mostly be pretty predictable modest bonuses to the object they arise from. I like the idea that higher level anomalies might be more varied, more dangerous and possibly with options for how to handle them similar to abandoned terraforming equipment. Take the risk for a chance of something really good, or take the safer option for something handy but unremarkable.
 
Are multiple star systems going to occur in "realistic" ratios (i.e. the most common type of system w/single star systems being less common), or are they going to be a special thing that only shows up uncommonly/rarely?
 
like the direction Anomaly change is going, risk chance was very.....min/max boring, and well losing a scientist to a low chance was annoying, but that should be rare in the first place i think.
As for the hyperlaning, I know you guys picked hyperlane travel as the default for the purpose of being able to choke point, which is a good choice with how the game funcitions and with how you wanted to go with fleets. So great idea, would like to see it come to fruition,
 
Hopefully your anomaly changes help with exploration being gutted from mid game on with the latest version.
More chokepoints, and more stagnation, just what we need!
New star system layouts that are implementations of what people have included from mods for years.