Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Ishman]

So now that there's an over map in our hands and the potential for exploration of new systems - what systems and mechanics do we need in place for a good Economy (running on Industry) before we move on to Story?

Before getting into ideas I'm sure we all have for added value goods, manufacturing items, raw resources, property to own (Tourism for rich corporates); what framework is going to support it?

I'll get this kicking with the prices themselves - ship hulls need to be more stratified, a paragon costs 20 or so lashers - the Gerald R. Ford costs $13.5B, this is enough to eat the entire production budget costs of thousands of tanks and armored fighting vehicles, or hundreds of fighter jets (except the F-35, a billion gets you five). You should be trading the entire revenue of a year's worth of production from some ultra-high tech industry plants farming antimatter off a star for a battle-cruiser, or a few aggri-planet loads of biomass.

Economy wise though, beyond the obvious missions transporting bulk goods from here to fro through varying degrees of territory danger paying out more than the value of goods being delivered (or less, providing ramifications to those of ill-repute), I'd like to see options for exploration based income: (rechartering territories that contact has been lost / forgotten / withheld from)
(caches of odd and fabulous materials and goods on long-forgotten worlds lost to time and danger)
(uncovering new and quicker trade-lanes and selling rights to same)

[Durandal]

Mining!

I'm a mining nut in games like this.  It would be easy to set up asteroid fields kind of like a current combat set-up (and it would be fun if you were harassed by the occasional pirate in more dangerous belts).  But why stop with asteroids in fairly stable orbits?  I'd love high-value rogue rocks traveling through the system alongside comets.

These could provide all matter of basic materials (minerals, gases and even exotic elements).

This would be a great base to form the economy around.  Ships could then be produced based on time and material cost (not necessarily player-based of course, but handy for even background calculations).  Of course, this wont work for absolutely everything.  Medical supplies, food and other non-idustrial commodities would still need to be based on some kind of algorithm based on the statistics of a planet.  But for the pure industrial side it would be interesting.  If the general supply of raw materials directly effects costs of things like ships, weapons, fuel and supplies then we have the start of a rudimentary trade system.

But more than just straight up mining, I'd love it if there was room for prospecting as well.  Hunting down the paths of rogue asteroids to sell to the highest bidder.  Exploring to find a gas giant rich in a certain kind of exotic material.  Things like that.  Could be interesting either as proactive player-driven action or as contract jobs from various corporations.

I'd also love to see missions that can change the basic makeup of systems.  For instance, a string of missions that end with the creation of a new outpost or space station in a remote corner of the galaxy would be cool.  Or perhaps missions where you eventually see a colony spring up after bringing enough supplies, people and equipment to an uninhabited world.  Having missions allow for a dynamically changing galaxy would have interesting ramifications on trading as well.  That new colony would be a great source of income as they are beginning.  Requiring a lot of pre-made equipment (which you can ship in for a high cost) while possibly producing some rare good that can be traded in civilized space for a tidy profit.

Definitely more complex than you tend to see in games like this, but dear lord would it be rewarding to see the galaxy change and flourish based on your actions.

[andreialbu]

One of the things the world lacks today is a game in which the universe is based on complex economical, social, political, cultural and military interactions and which reacts to everything that happens inside it, especially the player's actions. The big companies never seem interested of churning out anything even remotely complex and interactive. Sure, one can state that Skyrim is interactive because you can kick cabbages around, and that it's complex because there are lots of them and there's some physics involved. But those are only instinct-based complexity and interactiveness. Even a chimp can obtain pleasure from playing Skyrim, given that he is sufficiently immersed so that he can basically discern what's happening. Sure, you can say that you can kill guards and the town sentinel will take notice and even bounties will be put on your head. But in essence that's just cabbage-party with some social flavouring. In fact, the game is a thin layer beneath which lies a big void. The universe seems to float around the player. Until you personally arrive somewhere, that place is dead, be it a modest farmstead or a proud castle. Even if you want to get something going, there's nothing much you can. You can start killing the people, robbing the inn-keeper or piling some cabbages for the sake of exploding them, but that gets awfully-awfully boring in a very short time.

I think the foundation of any game which pretends epicness should be a seriously elaborated global economy. I think that the main factor which influences such an economy, and without which one cannot do, is population. Each planet, each moon - even each asteroid, if one considers such a thoroughly small scale worthwhile - should have a population. Each population is involved in production. Based on the nature of the product and the rentability of the production process, the population has a prosperity. Based on the prosperity, the structure of demand is determined. A prosperous planet has a high demand for both basic and luxury foodstuffs, whilst a non-prosperous one only for basic foodstuffs. The structure of supply is determined of course by the domestic industry, but it is also determined by commerce. A planet placed in a very advantageous geopolitical position may sacrifice agricultural activities and depend on imports for providing it's food. That was the case in history with countries like Flanders, is the case in the present with countries like Singapore, and will certainly be the case in the galactic era. Such a planet, nonetheless, will need very prudent leaders, because any severing of routes may result in a food shortage, which in turn increases the risk of social disorder. It will be very cool if the player has a covert ops department, so that he can incite revolts on planets facing food price spikes. He could also negotiate with the planet leaders so that he'll cease dissent, receiving instead trade concessions, political influence or direct money payments. Planets/states rival to that planet may also engage in such dirty politics. If a trade route by which a planet exports one of it's most profitable products is severed, there will also be a risk of social disorder, because the domestic industry will become incapable of paying it's workforce, ultimately resulting in unemployment. Cease of trade could be caused by many things. Except the direct blocking/hampering of the route itself by war or piracy, trade routes could be affected by the non-functioning of import/export facilities (either through acts of sabotage or simply administrative negligence), bankruptcy of trade companies, adoption of protectionist policies (tariffs and quotas) or failure of the merchant fleet (again, either through sabotage or the negligence of the ones in charge).

I think there should be a healthy number of goods, but not mind-boggingly many. The civilian goods should be: basic food, luxury food, household goods, clothing, vehicles and drugs. Then there should be the industrial raw matter: silica, carbon, common metals, uncommon metals, rare metals, common gases, rare gases, common flammables, rare flammables etc. And then the finished products serving all other purposes than civilian: integrated chips, building blocks, nanofibers, ammo missiles, hull plates etc. On each planet, for every one of these goods, there should be several key-variables, which change every two or three days: domestic production, imported quantity, exported quantity, supply, demand, cost of production (I'm talking about socially-accepted average CoP) and market price.

The player will be able to own and manage several types of enterprises: industrial, mining, commercial and even financial (I know that when we start talking about banking and insurance things get really beefy). Everything he does (as everything else that happens in the universe) affects the universe and so affects his business a little. However, only some actions affect it strongly. For instance, if you start plundering a trade route on which your own industrial enterprise exports it's output, your profit will probably lower. If you invest in the educational infrastructure of the planet on which you own a business, more educated workforce will be available, hence, the cheaper it will get and the lower the risk of an accident in your factories. And so on and so forth.

I could be rambling a lot on this subject, but I think I said enough for now. I really hope this game gets really, really interesting. Don't neglect complex interrelationships, otherwise it will end along with every other title sooner or later.

[xenoargh]

I totally disagree with the above.

1.  A complex economy is a development death-trap.  

In a game context, it either really matters, in which case we have a game like this where the combat is the core experience and all of a sudden, it's all about green-eyeshades play, if you want to do much... or it's purposefully nerfed until it's accessible, in which case enormous effort went into developing something that isn't really all that important.

Mount and Blade: Warband is a really good example of how to get it largely wrong while trying really hard to satisfy this kind of wish.  

It has an uber-complex economy under the hood, that actually deals with scarcity and adjusts prices accordingly.  Most players don't even know it, but it's actually checking out production from all the sources and many other things under the hood.

Problem is, it couldn't be made to work well with the AI at all.  At a certain level of complexity, it just becomes too hard to reconcile that kind of economy with the practical problems of having real-time AI agents.  

So the AI doesn't really interact with it much, while pretending to do so (for example, the Lords in that game raised armies, but they never really pay for them, nor do they have to get them in the same way players do- they're just magically generated out of thin air).

So the player was stuck with the (unexplained, undocumented, unpredictable) consequences of this system, that made things like trade runs less than fun because they weren't predictable, but the AI ignored it all because it wasn't possible for it to work well enough.

The end result was pretty lousy.  Most players ended up doing almost anything but deal with trade and the economy; the enormous effort put into those features was largely wasted.

So the game was Fun, but not because of the economy; it was Fun in spite of the economy.

X3 on the other hand made the game largely about dealing with the economy, not about flying a spacecraft.  This was a really terrible idea, and I didn't like X3 at all, because it took far too much focus away from what could have been a really core experience of being a pilot or a captain.  It was the green-eyeshades version of Freelancer, but it wasn't a better game because of all of that added complexity- Freelancer was far more fun.

Starsector is currently on the Freelancer side of that fence.  It needs to stay there, or it's going to get muddled and quit being fun.


2.  You don't need a complex economy to have vibrant agents in a simulation that are dynamic and interesting.  You're obviously new and aren't aware of what's been explored here by mods; Exerelin for 0.54a showed us a pretty reasonable dynamic system working on this engine, and it was pretty darn cool; with the relaxation of a lot of limits now in 0.6a+, it should be possible to build very decent control systems that will allow that model to become a lot better.

But the core idea that there needs to be a complex economic simulation behind AI agents' behavior is a chimera; the Lords in Mount and Blade, again, did not actually interact with the economy, yet they are fun and interesting and a challenge for players.

To have vibrant agents that are fun is really pretty straightforward; dynamic elements need to be in place where the player can make things, destroy things, and the same for the AI.  

That these activities are essentially economic doesn't mean that they must be complex to be Fun; setting up resource collection in an RTS is the creation of economic objects, and it's definitely fun and challenging without being super-complicated.

I really don't like the idea of there being 100 different economic objects that can matter, beyond simple trade runs with fixed locations of production (i.e., things that are unique to planets or unique Stations representing the last remnants of technology from before the Fall).

For the most part, it should mainly be about playing the core game- flying around in a fleet.  Giving directions to other fleets as your empire grows and a few rudimentary commands should be the extent of it- like the Combat Orders system now in place, player control should be loose, not detailed and tight, and player actions- direct interventions by using the player's Fleet in particular- should make the difference between defeat and victory.

It has to walk that line; if it starts veering towards becoming just a 4X in real-time it just won't work all that well, because the heart and soul will get lost somewhere along the way and the audience is completely different.

I mean... really. Next, we'll get invaded by no-twitch-skills 4X players who will demand that the combat core be made less and less important because they really want MOO II in real-time.  

While that's not a bad idea for a game and I loved MOO II... this game is kind of the inverse of MOO II, where the 4X was built and then the combat was added to it.  This is a game where combat is the core and the economy is being added to it.  A much different emphasis should be in the final product.  Action > economy.

[Hopelessnoob]

I want an economy that will exist without me but I can greatly influence.

If you've ever played the X series of games(if not get it on humble bundle right now) without player interaction the whole universes economy just collapses. I'd really like starsector to have an economy that changes based on what faction is winning even if i am not influencing the world. I would like to be able to buy up all the supplies from the hegemony and hurt their war effort or intercept the freighters going there and hurt the supplies in that way kill a hegemony system defense fleet without ever engaging it because it can't keep those ships well supplied.

I do not want to see things magically created would really like it to be NPC or player built.

[Silver Silence]

@Xeno and andre

Fun's a subjective thing, y'know. Your fun is not my fun is not the next poster's fun is not the poster-after-them's fun. Some might like blowing stuff up in a big stompy capital ship like myself. Others like going David vs Goliath with frigates vs the world. Perhaps some people like X3 for being able to set up a working self-sustained chain of stations that can pump out ships and all the equipment needed for those ships.

I think X3 did it's economy fine, I just wish it was possible to bring the economy to it's knees by destroying all the npc solar farms and thereby monopolize it with your own solar farms.

[xenoargh]

It is subjective, yes... but what becomes a game rule is not.

I didn't like X3; I thought it made the game far too much about economic tinkering and not nearly enough about flying around in spacecraft blowing stuff up.  

There is plenty of wiggle-room between those two likes.  I'm not saying that I don't want any economy at all, I just don't want one where that is the real game- I want blowing stuff up in my fleet to be the real game.  Again, look at Mount and Blade- they got it right, despite having an economy that was if anything more complicated than X3's under the hood.  Mount and Blade mainly comes down to player skill in real-time, not number-crunching.

The economy should be there to provide a framework for building my fleet, buffing my fleet and building NPC fleets that help me... but if I can't really expect to play beyond the basics without getting deep into economic-sim stuff, that would be very disappointing.  

The game's core Fun is what's happening in battles; anything that drags the focus too far away from that is to the game's detriment, imo.  I'd much rather see more focus put onto Story- big bad Bosses to kill, gadgets to acquire, stuff to do, than into economic stuff.

Part of it's because I'm a min-max player; if there's an economy and I'm not utterly bored by the whole thing, I'm going to figure out how to abuse it.  But I really hate having to play that way to have fun.  I'd rather that that part was simpler and my K/D ratio mattered a lot more than ROI calcs and figuring out how to corner the markets to screw up the AI.

The AI is the main problem here; AIs that can actually do much with a really complex economy are very non-trivial undertakings.  It's no big deal to make a resource-free AI that can provide a reasonable challenge, but one that cannot be economically defeated and plays without cheating is a whole 'nother story.  

Since making the AI actually able to play things well is such a major problem and it's so incidental to whether the core's fun, I'm saying that I don't think it should be emphasized much if at all.  

It's one thing, if the AI cheats up new resources when it needs them but can be defeated by manipulating economic rules designed for the players... but if the AI has to follow all the same rules, it's very unlikely to provide a sufficient challenge and defeating it will be mainly about manipulating numbers, not about skill.  There just isn't any satisfaction in defeating an AI that way, by starving it of stuff until it can't even fight back, so tbh I'd rather it wasn't even a serious option.

But if it's going to cheat, which it almost certainly is, then there's no point in pretending it's a real economy, is there?  

So just build a player-side system that provides a challenge to players, but let the AI do things that provide a challenge.  To the player, the economy will be "real"- it will have real rules that provide a decent challenge to get enough resources to defeat the AIs.  But don't bother dragging the AIs into that mess- it's not necessary, they're just there to give the player a challenging experience, after all.  

This isn't a simulation, it's a video game.  It needs to stay focused on that.

__________________________________________________

Since all this stuff may be a bit abstract for a lot of you theory-crafters, let's make it simple and concrete.

Let's say that, at optimal production times, it takes five Systems of average resources and fully-upgraded AutoFacs about 3 months to assemble all of the ships needed for one Hegemony SDF.

So, I as the player can kill one of those a week, easily.  With just one fleet under my direct control.

How, exactly, is a non-cheating AI going to even come close to dealing with that reality?  Answer is... it can't.  Period.  So it's going to have to cheat.  And we're right back to square one, where the AI's reality is different than the player's reality.

Since the above situation, with players kicking the AI's butt in actual combat if they're skilled and buffed correctly is likely to remain the norm, because it's already been battle-tested quite a bit and we know it's fun... I really can't reconcile that with the "real economy" stuff at all.  

It just can't be made to work that way- even if the Hegemony has 100 Systems, one player fleet is basically going to wreck it (60 Systems will be doing nothing but building replacement SDFs to replace what I've killed off, which in a "real" economy would totally derange everything), and if I can destroy the means of production and all that implies... there is just no way any AI will be able to deal with it.

[Megas]

Economy should do one of or both things:
* A breather between non-stop fighting.
* A viable, non-combat way for pacifists to win the game.

Me, if I want to play a warmonger, I do not want to get dragged down by microeconomics, especially when my ultimate goal is to destroy it (by destroying all factions).  I want to spend most of my time in Starsector blasting ships.  Just like if I play a fantasy game, I want to slay monsters, not bake bread or weave baskets.

[Alfalfa]

Mount & Blade's economy was not an 'enormous effort.'  It was piles of stuff whose value was based on how big the pile was.

    Item.economicValue = Item.value / Item.stackSize   <--- M&B's economy

The problem with Mount & Blade is that they started off with a really fun combat system, then started adding overmap features piecemeal and never managed to get them all working together properly.  So you had this rudimentary economy, but the interface was too janky to make trading very fun, and Lords didn't touch it, so it didn't offer any advantages in war.  Whether a country had a 100 fiefs or one poor village, it would have exactly the same number of full-strength Lords' armies marching around.  Did they provide a challenge?  Sure, but everything in the overmap was pointless except the ability to run around and find fights, so why did it exist?  They made all these countries, all of these Lords and cities.  They gave the ability to lay siege and gain territory; to get married and become a noble.  None of it was tied together though, so it all just served to highlight that the entire world was just a hollow playground for a bunch of immortal pheonix-like warlords.

Tying opponents into the economy would open up many options: raiding supply routes to weaken systems; taking over strategic systems like factory-worlds to boost your economy while crippling theirs.  Yes, the strategic AI could be outwitted.  So what?  The combat AI can be outwitted, is that bad?  As you said, the purpose of an AI is to provide a challenge, not an invincible barrier.  Want to defeat the Hegemony through overt combat?  Go ahead.  Others want to defeat the Hegemony via economic pressure and sabotage?  Let them.  It's a single-player game, no onrush of economic min-maxers is going to obsolete your playstyle.

Also, for those of us who have been playing this game for a long time, keep in mind that you're probably really good at it.  For many new players in the future casually ROFLstomping the defense fleet on a regular basis is not going to be an option, and I think Alex wants to cast his net a little wider than 'Top-down space shooter uber-elite pilots.'

[xenoargh]

Mount & Blade's economy was not an 'enormous effort.'  It was piles of stuff whose value was based on how big the pile was.

Not true- you're confusing the results with the system.  Under the hood, the economy is actually checking out supply and demand chains and a bunch of other stuff. 

Hence why values kept  changing constantly and why if you played it well and used the Enterprises you could corner the markets and manipulate prices.  If you never figured out how to do that, that's not the same as saying that it couldn't be done.

But the AI wasn't tied into it at all, so it kept it from being a dire problem.  The AI was originally tied into it, in the source, but they ran into exactly the same problems I pointed out above 

When you get past theorycrafting and you see what happens when high-skill players can demolish an economic output that might seem "fair" for players but is woefully inadequate for the AI, it's basically impossible to balance while keeping the core gameplay fun (i.e., not making combat so hard that people aren't roflstomping the AI a lot, which really turns newbies away).

The problem with Mount & Blade is that they started off with a really fun combat system, then started adding overmap features piecemeal and never managed to get them all working together properly.

No, the problem is that you can't have both.

If you have a game whose underlying rules are based on real economics, and it's not based on expansion curves that are really fast and easy... then destruction of anything, whether it's the products, the sources or the supply chain is a very big deal. 

Take that example of killing off the SDF from earlier.

If just kill one SDF per month, that's 5% of the total economic output gone for 3 months.  It would completely derange any sane economy to sustain those kinds of losses.

But what about killing the equivalent value in Lashers and Hounds?  What if I can build my own fleets and I've tuned them to kill those fleets at a good attrition ratio while being fast enough to avoid a SDF? 

There are just so many things that are so difficult to prevent a player from doing, and that's purely on the destruction-of-product side, let alone if I am able to destroy the means of production or actively interfere with the supply chain or, even worse, actual supply and demand are factors and I can corner markets. 

Moreover, think about how complex the AI would have to get, to put up much of a fight.  It's just barely possible to write RTS AIs that can sometimes beat a medium-skill human player, with a very limited map and very few economic variables; you're basically asking Alex to build something even better than that, because if it's not playing out a credible game, it's going to get hammered the minute the flaws in the analysis are found.  At best, the only way to fix it would be to buff the AI's results artificially, and even then, if there are markets that players can corner, we'll have to play with self-imposed rules to give the AI a reasonable chance to thrive, which I think is lame.

Anyhow... sorry if any of this is put a little too roughly.  I just think that this is one of the major areas where it'd be easy to screw up the game design while chasing fools gold.  This kind of "I want it to simulate everything" urge is a very common one in game-design discussions, but it's largely made by people who haven't had to actually build those kinds of systems and haven't seen the pitfalls yet.  I trust Alex will find a good middle ground that's fun, reasonably accessible and that provides an appropriate challenge.

[Alfalfa]

Mount & Blade's economy was not an 'enormous effort.'  It was piles of stuff whose value was based on how big the pile was.

Not true- you're confusing the results with the system.  Under the hood, the economy is actually checking out supply and demand chains and a bunch of other stuff. 

Hence why values kept  changing constantly and why if you played it well and used the Enterprises you could corner the markets and manipulate prices.  If you never figured out how to do that, that's not the same as saying that it couldn't be done.

I never looked at the code or tried much merchant-style play, so I can't refute that.

The problem with Mount & Blade is that they started off with a really fun combat system, then started adding overmap features piecemeal and never managed to get them all working together properly.

No, the problem is that you can't have both.

If you have a game whose underlying rules are based on real economics, and it's not based on expansion curves that are really fast and easy... then destruction of anything, whether it's the products, the sources or the supply chain is a very big deal. 

Take that example of killing off the SDF from earlier.

If just kill one SDF per month, that's 5% of the total economic output gone for 3 months.  It would completely derange any sane economy to sustain those kinds of losses.

But what about killing the equivalent value in Lashers and Hounds?  What if I can build my own fleets and I've tuned them to kill those fleets at a good attrition ratio while being fast enough to avoid a SDF?

Hmmm.  What if the Hegemony and perhaps other factions existed beyond the scope of the campaign map?  That way the factions could get 'infusions' from HQ if the player FUBARed them, making them robust without being obvious cheaters.  You wouldn't defeat them so much as convince them operations in the sector were no longer profitable, at least temporarily.

There are just so many things that are so difficult to prevent a player from doing, and that's purely on the destruction-of-product side, let alone if I am able to destroy the means of production or actively interfere with the supply chain or, even worse, actual supply and demand are factors and I can corner markets. 

Moreover, think about how complex the AI would have to get, to put up much of a fight.  It's just barely possible to write RTS AIs that can sometimes beat a medium-skill human player, with a very limited map and very few economic variables; you're basically asking Alex to build something even better than that, because if it's not playing out a credible game, it's going to get hammered the minute the flaws in the analysis are found.  At best, the only way to fix it would be to buff the AI's results artificially, and even then, if there are markets that players can corner, we'll have to play with self-imposed rules to give the AI a reasonable chance to thrive, which I think is lame.

I understand that AI usually needs some help to deal with skilled players.  I'm less concerned with absolute fairness and advanced economic simulation (though having something like that, divorced from the main strategic layer, would be nice for merchant players) than I am with feeling that I am capable of affecting AI entities in a variety of ways.  Games are rarely concerned with genuine realism, a believable facsimile will almost always suffice.  To take as an example another game that attempts to blend strategy and tactics, the Total War games allow you to overcome disadvantageous battles through superior tactics, or to stack the odds in your favour via strategic domination.  The AI, depending on the difficulty, uses a variety of cheats, yet victories are victories, and repeatedly defeating the enemy, taking his territory, and sabotaging his facilities will noticeably impact his ability to retaliate. 

This is where M&B falls on its face.  The AI is painfully, agonizingly obvious about how it is completely and utterly unable to be affected by you in any way, ever (slight exaggeration).  Countries actually get stronger as you chip away at them, because they have exactly the same amount of forces, but concentrated over a smaller area, whereas more and more of your resources are expended defending your increasingly widespread holdings.

Anyhow... sorry if any of this is put a little too roughly.  I just think that this is one of the major areas where it'd be easy to screw up the game design while chasing fools gold.  This kind of "I want it to simulate everything" urge is a very common one in game-design discussions, but it's largely made by people who haven't had to actually build those kinds of systems and haven't seen the pitfalls yet.  I trust Alex will find a good middle ground that's fun, reasonably accessible and that provides an appropriate challenge.

No worries, I haven't been all sugar and spice myself.   Heated words are often just the sign of a worthwhile debate.  Mount & Blade is one my favourite games of all time and one to which I have devoted 100s, if not 1000s, of hours, but it is one whose virtues and achievements are matched by its many flaws and unfulfilled potential.  It is also the most analogous game to Starsector's present and purportedly intended state, and as such I fervently hope that Alex achieves what TaleWorlds could not.  I don't know which route to this goal is the correct one, but Alex is the most thoughtful and realistic indie developer I've encountered, so I'm sure he will find it.