Fractal Softworks Forum

Please login or register.

Login with username, password and session length
Advanced search  


Starsector 0.97a is out! (02/02/24); New blog post: New music for Galatia Academy (06/12/24)

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - UrbanGiraffe

Pages: [1] 2 3 ... 5
The problem:

One problem with the current officer system is that officer and player-piloted ship skills suffer a bit from combinatorial explosion. To some extent every officer's personality and skills need to be considered for placement on every ship (often scrolling through lists in two GUIs where every officer and every ship cannot be seen/compared at once), while also considering how every permutation of these assignments compares to every other permutation as a whole. This creeps into other areas of fleet customization, such as contemplating alterations to the hull designs to fit possible new officer configurations, with a high time cost to testing these out and undoing/redoing the changes. Obviously players can weed out many of these possibilities, but at higher fleet sizes and officer counts it becomes noticeably cumbersome.

Players can resolve this by:

- gradually homogenizing their officer skills (and personalities, and elite skills...) to only the ones considered universally useful (rather than attempting to work with what they randomly acquired)
- by giving up and ignoring most of the nuances of their officers' varied skills
- by giving up on making major changes to their fleet's structure past a certain point
- or by wasting possibly large amounts of time revising these fiddly little details repeatedly

This gets particularly bad when it's considered that the distribution needs to be at least partially re-evaluated every time a major ship is added or removed from the fleet, or sometimes even just have its weapons/mods changed. I suspect that by the mid-to-late game most players will either work to homogenize their officers or stop engaging with the mechanic in depth, eventually defeating the impact of having randomized officers at all.

Parallel to this, a conflict exists for the player between the more exciting "piloted-ship" skills and the more boring (but effective) fleet-wide skills. It's a conflict that I don't think is resolved easily, as "balancing" the two trends toward making the piloted-ship skills overly powerful (possibly to the point of disrupting the vitally important core combat system) or making the fleet-wide skills increasingly boring and unattractive. (The latest blog post discusses buffing piloted-ship skills to shift the current balance).

A solution:

One starting point is to ask "why do officers exist in the first place?" in military organizations (not that realism matters, but I think following this line happens to work). One answer to that question is pretty straightforward: officers exist to implement the policies or "doctrine" decided upon by the military leadership.

Imagine each ship has a "leadership" rating: an individual ship can be poorly-led or well-led.

Poorly-led ships can be imagined as timid, unable to endure a long engagement, and unable to use the complicated doctrines/procedures/tactics/whatever the player character has developed in their career as a fleet commander. These aspects are currently implemented as officer personality, maximum combat readiness, and player/officer skills.

Well-led ships can be imagined as less resistant to risky/dangerous orders, better prepared for long fights, and capable of complicated doctrines/tactics/whatever that require greater coordination by the crew.

Under this possible new system, when the player levels up they pick between either the ordinary fleet-wide/economic bonuses or unlocking a doctrine that can then be added to any number of ships in the fleet (these "doctrines" can just be the current "piloted-ship" skills). When the player revises a ship's design, they can select within the same screen which of these doctrines its officers will attempt to implement and what its desired aggression level will be (with the maximum number of doctrines and the maximum aggression levels limited by the ship's leadership rating).

Where does leadership rating come from? Some options are:
- A new, expensive, and possibly rare personnel type alongside crew and marines, called "officers" (only "sold" by military markets?)
- Recruited commanders assigned to a ship, as in the current system (but adding only leadership rating or maybe other economic benefits to a ship, not the skills directly)
- Possibly passive player skills that boost fleet-wide leadership in various ways, though these meta-skills risk being particularly boring-but-necessary ways of spending points
- Reintroducing the old varied crew skill levels, but that likely isn't worth the complication

(This hypothetical system could handle fluctuations in leadership rating smoothly. If the desired doctrines/skills for each ship are put in a list that can re-ordered by the player, low priority doctrines can be deactivated first if that ship's leadership rating dips for whatever reason. Similarly, higher leadership enables higher maximum levels of aggressive "personality", but can smoothly fall back to less aggressive settings than the player selected if the rating drops. One simple algorithm for allocating the crew-officers would be to borrow from the old varied crew system, with ships ordered earlier in the player's fleet list being prioritized for crew-officers.)


- All mechanics related to configuring an individual ship are handled in one straightforward place with minimal need to consider the effects of the officer pool on other ships (and no incentive to waste time hunting for "optimal" officers for a configuration the player might abandon later)
- The incentive for the player to avoid the more exciting piloted-ship skills in favor of the more effective fleet-wide skills is removed, without making the piloted-ship skills overly powerful or the fleet-wide skills overly boring
- The only truly new mechanic required is per-ship (or at least fleet-wide) leadership rating, and perhaps the new personnel type
- Crew-officers would allow players to control the cost of their fleet over time though a new factor, rather than only considering each hull's relatively static supply cost
- A new economic tuning parameter to incentivize/disincentivize certain hulls in an interesting way
- A new way to disincentivize crew casualties, if needed
- A thematic and easy way for the player to tweak their AI aggression, possibly avoiding frustration from the AI mishandling increasingly complicated configurations


- Making commanders more interchangeable to avoid micromanagement also makes them less individually interesting (but in my opinion the current system strongly incentivizes players to minimize their variance or ignore it)
- Little thought is given here to integrating the 'elite' skill system
- What does this add that hull mods etc do not? (A possibly valid criticism, but one that also applies to the current system, and one answer is that it's more closely linked to the player's experience-point progression than other ship customization systems in the game)
- The skill system has already been though many major revisions and the game is excellent as-is (though the many and continuing revisions might suggest a flaw from the beginning)

I set up a bare bones colony in the middle of the sector without any resources, and amusingly it still ended up getting targeted by an expedition to disrupt the black market production of drugs from the basic "Population & Infrastructure" building:



To be fair, it does have the second highest "market share":

I know this is a weird edge case for a couple reasons, but it doesn't feel WAD for a colony this unoffensive to be targeted for this reason (and it does show that it's actually completely impossible to have a colony that isn't targeted by expeditions, unless its growth or accessibility is deliberately stunted).

General Discussion / Re: What planets are best for colonization?
« on: December 27, 2018, 02:54:46 PM »
The market system is a bit unintuitive, and I think there are a few things worth noting (but I think it's hard in the current version to do badly with colonies).

With the way that demand and market share works, it doesn't seem worth it to produce commodities on more than one planet (except for food). It's easy to quickly get a massive colony going (even with mediocre hazard rating) that will dominate every industry it has the resources to produce. Having other planets on top of that producing the same goods will just spread the sector-wide market value of the commodities thinner, providing an increase in income that diminishes on top of the already diminishing market share increase. In fact, the additional income from producing goods on multiple planets could rapidly flip negative for most commodities, when you consider the reduction in income on your other colonies. Food is an exception, since it has large sector-wide value and demand while the infrastructure itself for food production has very low monthly upkeep. Drugs and organs are possibly another exception that could be worth it in multiple locations, and maybe fuel.

In this respect, I think the "optimal" setup that will usually shake out is to have one open-port colony (for growth and massive organ/drug revenue) in an attractive location, with however many other planets you'd like as a single source of goods not available on that planet (if only because it's "cozy" to produce everything in-faction), plus as many low-hazard rating farm planets you can manage (whose purpose is to produce nothing but food and population income while increasing sector-wide market value for goods you produce).

Where you pick the main colony doesn't actually matter much (although hazard rating is the most important factor). Things like a cryosleeper, a slot for a comm relay, accessibility bonuses, local resources, ruins, etc are nice, but not essential. It could probably be in the middle of the sector on one of the "featureless" planets and get by producing nothing but drugs/organs and heavy industry assuming they are valid for colonies, but I haven't tried that. The "feeder" planets producing nothing but food and without defensive upkeep will be easily profitable, but will also increase sector-wide demand in drugs, organs, ships, organics, domestic/luxury goods, supplies, fuel, and machinery which should indirectly boost income in the main colonies substantially if they grow to a large population.

Another note: there's a chance when raiding a planet with a nanoforge that the nanoforge will be looted, permanently reducing their production substantially and boosting your own market share in the heavy industry commodities or fuel production. It's also an alternative route to acquire these things (plus blueprints!) without extensive exploration. It's more tempting than the temporary disruptions, anyway, since it's permanent.

And one more note: the decrease in demand from AI cores should be a strictly bad thing under most circumstances, since it'll decrease sector-wide market value for that commodity. Unless it's needed to prevent a shortage, it's probably best to never use the lowest tier cores and to be cautious about using the middle tier. However, the top tier cores provide special bonuses to certain buildings, which might be worth checking out.

General Discussion / Re: Legion-class Battlecarrier
« on: December 18, 2018, 06:41:47 PM »
If you can find one you might want to consider pulling the Mark IXs and putting on a storm needler instead.

While you lose relatively greater weapons arc coverage and 320 range you gain 8 OP, 150 flux/second and a weapon that is harder to avoid.

Storm Needlers are more flux efficient than railguns and also provide more DPS per OP. The two railguns you have in the front plus the Mark IXs are a total of 25 OP 515 DPS, and 550 Flux/second.

For only 3 OP more a storm needler is 750 DPS and 650 Flux a second. Even if you can’t fire it all the time you’re coming out ahead.

I would use lower tech bombers and eschew the longbows. The legion tends to get to the front when it’s on your side and the large kinetic pressure you can put down with the storm needlers will negate the need for the longbows

If those are sparks then definitely drop them for something else. Claws are OK for the ion damage. But I think you might do better with perditions/khopeshs/piranhas and Thunders.

You also probably don’t need 4 flacks. You could drop to 2 and put IPDAI on to make them effectively 3 and save a bit of flux.

Yeah, Claws are much better than Sparks, I think for that screenshot I had them switched out to save on pilot deaths and didn't notice. My main issue with the Thunder is the high replacement time, and that they tend to get obliterated by big groups of Remnants with their burst PD. The extra OP for Daggers over Perditions is usually worth it IMO, since Perditions are pretty useless against frigates. The longbows can definitely be dropped with the addition of the Storm Needler.

And one Storm Needler does make the ship way better. Since it also acts as anti-fighter PD, it's possible to drop the flaks (one dual flak in the other large turret seems like enough), and the freed-up OP can be better spent on more heavy maulers or luxurious hull mods.

General Discussion / Re: Legion-class Battlecarrier
« on: December 18, 2018, 02:39:41 PM »
This is the setup I ultimately settled on for use by an AI officer:


If the officer in this case had missile spec, I would've used medium sabots, but here I wanted something that could survive sustained fights without missiles. If the officer had Ordinance Expertise (faster projectiles) I would've used autocannons and other cheap weapons instead of railguns. I used simulation fights against fleets with many fighters, frigates, and a few destroyers to gauge how well it does when totally overwhelmed, since the ship has a tendency to use the burn ability semi-suicidally. Ultimately, I used this to win a hopeless Red Planet fight alongside only 4 other ships.

The ship has poor flux dissipation that limits what sort of weapons can be used in the large slots. Those railguns in the back might seem weird, but they're there to fight off frigates that can very easily hang out behind the ship with impunity. The heavy mortar offers the bare minimum of high explosive damage necessary to kill things if given the chance, while the kinetic emphasis is there mostly to force ships to back off.

Suggestions / Re: The weekly teaser
« on: December 11, 2012, 12:50:03 AM »
I think a comparable game in this instance would have to be Dwarf Fortress... while updates can be many months in the making, frequent development blurbs on the site make the waiting much less dull. I definitely think it'd help Starfarer keep up its momentum to try something similar, but it all comes down the dev's personal preference.

General Discussion / Re: That Moment
« on: December 02, 2012, 02:52:47 PM »
Getting both torpedo hits with a Cyclone Torpedo Launcher feels pretty great. I once hit two Hounds consecutively with one volley, and it was quite the spiritual experience.

General Discussion / Re: Stuck and out of ideas...
« on: October 23, 2012, 02:48:19 PM »
Awesome, you also use the apogee! It's definitely my favorite ship to pilot. A fun configuration for it is to use a plasma cannon and two antimatter blasters on the front, and use it as a strike cruiser. If you really want to go to town, you can also put a reaper launcher on the large missile mount instead of a hurricane. It's quite a versatile ship, especially since the new drones can essentially replace PD (against fighters/bombers) allowing for more points elsewhere. Going toe-to-toe with onslaughts in that setup is surprisingly effective, especially if you use hardened shields and focus on vents.

General Discussion / Re: Why the Aurora nerf?
« on: May 08, 2012, 09:55:04 PM »
Eh, missile spam definitely has its merits.  :)

Plasma cannon spam is in an entirely different stratum of fun, though.

General Discussion / Re: Why the Aurora nerf?
« on: May 08, 2012, 09:42:08 PM »
The Apogee's always been my favorite ship, by far. I use a plasma cannon, two antimatter blasters in the small hardpoints, and a hurricane MIRV. PD isn't really necessary with 360 shields and a massive flux reserve. It'll instagib anything smaller than a cruiser, and overload any shields in a couple salvos.

I've always disliked the Aurora. It only really functions well with everything (aside from missiles) on autofire, so there's no real combat involved. Floating around spamming missiles and tanking isn't all that enjoyable.

Discussions / Re: Dwarf Fortress
« on: May 01, 2012, 02:45:18 PM »
I would recommend quarantining fresh immigrants after the fortress reaches a decent population. I usually atom smash the useless cheesemakers and such, and then interrogate each dwarf for signs of them being a vampire. If they're a tramp who's been to 74 other forts before immigrating, that's a sure sign. They also carry with them cute tokens of their kills, so if one has 600 +Dwarf Bone Earrings+, they're a vampire. If it's a recently cursed vampire, they may not yet follow these, though, so a starvation test usually settles it.

Fortresses with an overenthusiastic secret police have the tendency to collapse, especially if the vampire manages to frame someone else, such as Urist McLegendaryCookWho'sFriendsWithEveryone. I typically only keep a wimpy hammerer equipped with the junkiest weapon possible, as otherwise random craftsmen who fail mandates get their skulls liquefied.

General Discussion / Re: Added a simple spoiler tag
« on: April 19, 2012, 09:56:49 PM »
A choose-your-own-spoiler.

Starting ship?

None. Sells weapons, buys candy instead.
A lone fighter wing attacks. You barely make it out alive... smart plan was smart.

Head back to Hegemony station to a life of shame and prostitution.
You stumble across a decimated Tri-Tachyon fleet, with just a Dagger wing remaining to escort an insufficiently crewed Paragon. You still die.
Game over.

Attack a crew-deficient pirate fleet, hoping to capture some ships.
After kiting for 30 minutes, they retreat, and you capture 4 ships. In your excitement, you knock a glass of orange juice on your keyboard.
Game over.

Use candy as ammunition in ship mounted slingshots.
Oh, it's that kind of candy...
Rename ship the URST Armok, conquer galaxy.
The end.


Killed by a shuttle.
Game over.

You sell the equipped weapons, and then realize all they have at the Hegemony station are light machine guns and mortars. The selling penalty is such that you can't afford to buy your stuff back, by a margin of 10 credits.
Uh, ok, I still deal kinetic, right?
You run into another Hound.
Game over.
Buys candy instead.
See above.
Sell hound, buy shuttle.
Capture the remainder of an insufficiently crewed defense fleet, including two fully equipped Onslaughts, without doing anything.
The end.


You and your funny jokes.

You get kited to death by a Hound, which was engaged with 20% health remaining.
Game over.


Discussions / Re: Dwarf Fortress
« on: April 19, 2012, 02:36:39 PM »
Urist McApogeeCaptain cancels shoot pirate: Interrupted by unexpected DF thread.

General Discussion / Re: shuttle love pirate bomber hate
« on: April 18, 2012, 10:11:24 PM »
Now I totally want to mass 60 shuttles, and unleash them on a defense fleet.

Pages: [1] 2 3 ... 5