What's Next after v0.4.1?

[eleazar]

With respect to limiting runaway production growth. I think the fact that you have to continually spam ships is indicative of a need to rethink how ships are managed.

Games like this are meant to be about thinking strategically.

• What solar systems are most important?
• Where are the key weaknesses in my supply lines?
• In what direction should I expand my empire?
• What are my macro priorities (Population vs Production vs Science)?
• Do I want to have a dense highly developed empire on every solar system, or do I want to have a sparsely developed empire?

I Substantially agree.

However you are looking for a solution in the opposite direction than were we would look.
When a game mechanic leads to gameplay that's tedious and uninteresting to the player, our game philosophy (see my signature) is not to introduce an AI or management layer, but to remove, alter or streamline the game mechanics so that they are interesting and not overly burdensome.

In short we're trying to make a game, not a simulation. Not only is it IMHO a better way to make a truly high-level strategy game, but it requires a *lot* of extra work to create complicated systems that aren't fun, and then cover them up with abstracting management GUIs or AIs. That path seems to lead to failed 4X games.

[Vezzra]

When a game mechanic leads to gameplay that's tedious and uninteresting to the player, our game philosophy (see my signature) is not to introduce an AI or management layer, but to remove, alter or streamline the game mechanics so that they are interesting and not overly burdensome.

Although I agree with this design philosophy, I've been wondering for a long time now if it's possible to stick to that without any compromise ever. The problem I see is if certain game mechanics essential to a 4X game really can be designed in a way that they will smoothly scale up from early to end game.

Especially when it comes to the whole planets/colonies/fleets/ships thing. One of the fundamental aspects of 4X is that you start small and grow very, very big. The difference between what you start with and what you can have at the end game stage is tremendous. I've yet to see a 4X game that is able to handle this in a way that the game is (at least to a reasonable degree) equally interesting and fun throughout all stages. What I've seen is that game mechanics that allow for an interesting and fun early game tend to get tedious and micromanagy in later stages (as you have to manage more and more assets), and game mechanics that get interesting and fun at later stages are rather boring during the early stages.

Regarding FO we currently have case one. Early game is really fun and interesting, managing a few colonies and fleets is no problem at all, and trying to automate things would take the fun out of it. However, once your empire grows beyond a certain size, managing all the colonies, ships and fleets gets less and less fun. What's fun when you have only very few assets isn't fun anymore when you've a hundred times the assets.

It's the same difference as with running corporations. Managing a tiny family business is very different from running a big, international corporation. You simply can't take care of things like you do in a small business, because you'd be overwhelmed, that's why additional layers of organisation/administration are required you wouln't need otherwise, or which would be even detrimental for a much smaller business.

Your empire in a FO game can be compared to a corporation. You start with a small business, you end up with a big international corporation. The UI provides the player with ways and tools to manage this "corporation". And these ways and tools have to change/adapt throughout the various stages of the game, as the "corporation" gets bigger and bigger. I don't think we'll be able to entirely avoid providing at least some mechanics/tools that allow for a certain level of automation. And here is where davidescott's ideas might be worth considering, at least some of them.

The only other way would be to invent/design game mechanics in a way that enables the player to manage an empire consisting of one colony and three ships the same way/using the same set of tools as an empire consisting of 50+ systems, 100+ colonies and dozens of fleets with hundreds of ships. However, apart from requiring a fundamental redesign of FO, I doubt that this is possible at all.

[davidescott]

However you are looking for a solution in the opposite direction than were we would look.
When a game mechanic leads to gameplay that's tedious and uninteresting to the player, our game philosophy (see my signature) is not to introduce an AI or management layer, but to remove, alter or streamline the game mechanics so that they are interesting and not overly burdensome.

Two comments on this.

Part of the problem is how I play games like these. Once I have tamed a portion of space I want to settle every single planet in every single system. The term "sim-civilization" describes how I like to play pretty well. Winning against the AI and conquering the galaxy is a by product of my designing and managing the perfect galactic empire. I don't find conquering the opponent very satisfying if the only way to do it is not to settle half the habitable planets. As a result I am overly exposed to the scaling issues.

The second is that I don't think you have to have add lots of burdensome management layers. The system I envison would allow existing players to play very much in the same fashion as they currently do but would allow people like myself the ability to scale up a bit more. The key tools would be the following:

• A colonization overview
  It would show all the planets I have seen. Allowing me to filter based on distance from existing supply lines, distance from specific starbases, suitability for species, specials, max population for species. I could then from that gui click "add to colonization queue and set the initial focus", and then reprioritize my queue by dragging. There would be a big "colonization" budget which I could tweak up and down.
• A planets overview (perhaps part of the colonization overview but with a "colonized" filter) that would allow me to change and manipulate focus in bulk.
• Starbases and flagships
  • The player would initially start with a single starbase and a single Mark I. Those species that start with a Mark I and a Mark V would instead start with a Mark I, Mark V, Starbase and Flagship.
  • On the "objects" screen you could click on a fleet and open a fleet management gui. Vertically would be the different designs available and you set Min/Target/Max for could drag them up and down vertically to set priority. As ships are produced they would be assigned to fleets or moved from fleet to fleet automatically in accordance with the Min/Target/Max for those ships. Retrofits could also be automatic.
    
    A few examples. Starbases A and B, designs X and Y. X is prioritized over Y.
    
    A has 5 X and 3 Y. B has 5 X and 6 Y. Min for each is set to 5 at both bases. Result, the Y at B is transferred to A and begins traveling to A, an additional Y is moved to the top of the production.
    
    X and Y are the same hull. A had a min of 3 X and a target of 5, now with Y the min for X is reduced to zero, and the target is reduced to 0. The Min for Y is set to 3 and the target set to 5 (Y replaces X). Result 3 of the X's begin their retrofit immediately, the remaining 2 X's will be retrofit according to the budget.
    
    To further manage this each starbase/flagship would have an overall production figure again three part value Emergency>Target>Optional, this might be best expressed as a percent of the overall production of the empire rather than a static value (since production ramps up). Emergency would be the budget to reach the Min values, Target to reach the target values, and Optional to reach the Max fleet size. There could be global budget modifier that allows you to tweak all starbase budgets up and down to adjust for various circumstances (I'm at war increase the military budget across the board vs I won the war decrease the budget again).
  • Ships could be sent from the Starbase/Flagship for a mission (raid this enemy colony or attack this monster that entered my space) but would not be able to travel too far from their starting point, and would then have to return once the mission was done but could otherwise stay in that system as long as desired.
  • Planets would feel unhappiness if they were not well protected by a starbase (based on a calculation of how many ships the starbase could send to protect them withing N-turns). And larger populations would demand more protection than smaller populations.

I don't see a huge difference between this and the way the game is currently played. At the moment I don't build colony ships unless I have some idea where I want to send them, so this just makes explicit the "this ship is for this planet" that I already had in mind. In my mind it simplifies ship production and makes a clearer distinction between "this is my defense fleet" and "this is my attack fleet."

In the end its your game, so its up to you how things are developed, but I always get to the point where I'm fairly confident I could win, and then things bog down as I try to expand my empire so I just give up. Which is personally frustrating because I never actually get to tech out or enjoy the proces of actually destroying an enemy empire.

The alternative tack you are taking seems to be a "balance the gameplay and build costs and scale up ship tech" to ensure that the player only ever wants to have 20 total ships which I think is hard to get right for all kinds of play. It certainly never works for how I play.

[davidescott]

One last comment. If your feeling as the game designers is that you dont want to take it in the direction I'm suggesting I will gladly leave you alone.

There is some tension between the design philosophy (there always will be). In particular between "You should be able to explain the basic rules to a reasonably clever child without difficulty" and "The player should spend his time making interesting and significant decisions: i.e. not making slight tweaks to a huge number of variables."

A large part of the game rules you have kept fairly simple. There are ships... they have weapons and defense -- bigger numbers are better. Newer hulls are better -- they can store more weapons/are faster. You build colonies with colony ships. Having more planets is good -- they make production/science.

Its all very simple and straightforward how to play. You create as many colonies as you can defend from threats, research techs, build better ships, create more colonies, and the process feeds on itself. There are lots of fun strategic decisions to make along the way.

Things start to go south when you hit the boundaries of the universe. You control your entire arm of the galaxy. You have 20 colonies and with new research can settle those inhospitable planets giving you hundreds of 100 planets withing your arm of the galaxy. You can amass a stack of doom and concentrate it on your borders.... and everything gets unbalanced.

It takes 15 minutes per turn to deal with settlement colonies because you need to build individual ships and tell them where to go and what to settle. It takes 30 minutes per turn to move ships around because where one planet could build a battleship in 5 turns then 20 colonies build 4 per turn and 100 planets would build 20 per turn. You end up making one maybe two strategic decisions every half hour and the game stops being fun.

So the rules get "adjusted" to prevent things. Where the first hull of the battleship costs 100 production units, the second costs 110, and the third 130. Where the first 5 colonies have such and such effect the next 5 have a reduced effect. Where the first ten scientific discoveries cost X the next ten have r*X. You end up fighting the exponential growth in output of the empire with exponential curves in all the other attributes, and all those carefully calibrated curves are only going to intersect and be properly balanced for a particular size galaxy with a particular density of planets/specials/etc. I think this is very anti-KISS.

For most games those curves will be reasonably close enough to intersecting that you get a well balanced (if not ideally balanced) game, and most strategies are reasonable for most maps, but it is clear that some strategies work best for certain map sizes. For smaller galaxies you probably want to be more defensive and slower growing (but if you are skilled you might be able to grow quickly), for larger galaxies growing more aggressively is good.

The problem becomes apparent when you take a parameter to extremes. A wide open galaxy with 10000 stars, 4 races, low monsters, high specials and planets. The game will completely break around turn 100. I face this all the time with Civilization type games when I want to build 20+ cities... the "unhappiness" curves meant to offset the exponential output growth break the 4X part of the game. It is no longer beneficial to eXpand. There is no good that comes from eXploring. One might as well hole-up and tech-out in their own little corner of the universe. The game was very KISS, because it was just 4X. Now its 4X up to some point then its 2X, maybe 1X.

I think this is a case where some degree of "realism" and simulation can help. A massive galaxy spanning empire is not going to be managed in the same way as a small 5 solar system group. Yes there is exponential growth in output as you grow unrestricted. I say let it happen. Don't balance the cost curves against the exponential growth, just figure out a way to manage things at a higher level.

[Vezzra]

I think you're making some valid points here. While I think it's perfectly fine to implement certain game mechanics (like some kind of "maintenance" mechanic, whatever form that may take) to counter unrestricted exponential growth to a certain extend (otherwise you'll wind up with every game just turning into a who can churn out bigger doomstacks faster), I also think that this can't be taken so far as to ensure the player will never have to handle more assets as can be easily managed with the tools sufficient for a five colonies empire (for the reasons you cited).

Currently FO works like that: it has some kind of "maintenance" implemented that counters the exponential growth of industrial capacity, but not to the extend that you never can have more ships/fleets so that things won't get too tedious to manage mid and late game. In my last test game I epxerienced exactly the same things you mentioned when my empire grow beyond a certain point. Industrial and research capacities began to rise exponentially, consequently also my fleets (not as much due to the "maintenance cost", but still), resulting in turns taking much longer, up to an hour or more before I quit. It simply became increasingly difficult to manage all the colonies and fleets. Late game stage as it is now is neither interesting nor fun (at least with larger galaxies, a very small galaxy with 30 systems might be different in that regard).

I think this is something we definitely need to do something about.

[Sloth]

My two cents about the topic brought up by davidescott:
I'm on eleazar's side in this regard and i firmly believe in the approach taken by FO. The single production queue of all supply line connected planets is just genius and goes a long way solving problems of other 4X games.

I agree that at the moment, building warships is the only thing you can do with industry at most points of the game. But there is no reason that it has to stay this way. Content can be added and adjusted, PP costs can be raised. Feeding the player with awe-inspiring projects (buildings and ships) for hundreds of turns is possible, it will just be a lot of work and will take time to develop. Some content can be so expensive that it's virtually restricted to huge galaxies, it's just a matter of balancing (and quantity of content).

Things start to go south when you hit the boundaries of the universe. You control your entire arm of the galaxy. You have 20 colonies and with new research can settle those inhospitable planets giving you hundreds of 100 planets withing your arm of the galaxy.

This is one of the things that desperately needs balancing. Making inhospitable planets hospitable is far too easy currently. At the moment you just need to research some techs which you want to research anyway (because some of them even enhance the population of your good planets). A lot more dedication has to be required (dedicated techs and buildings) and the result should be less rewarding before the late game.

The problem becomes apparent when you take a parameter to extremes. A wide open galaxy with 10000 stars, 4 races, low monsters, high specials and planets. The game will completely break around turn 100. I face this all the time with Civilization type games when I want to build 20+ cities... the "unhappiness" curves meant to offset the exponential output growth break the 4X part of the game. It is no longer beneficial to eXpand. There is no good that comes from eXploring. One might as well hole-up and tech-out in their own little corner of the universe. The game was very KISS, because it was just 4X. Now its 4X up to some point then its 2X, maybe 1X.

eleazar had some nice concepts for happiness, which included the distance to the homeworld of the resident species in its calculation. 10000 stars can't be balanced, but i'm sure a satisfying number of stars can be.
Also eXploration in FO doesn't always have to stay linked to eXpansion.

[Burkey]

Sorry for jumping in on the post and possibly changing it's direction, but is there a list anywhere of the differences (maybe major if there's alot) between 0.4.1 and the new 0.4.2 (test release?)?

[Vezzra]

Sorry for jumping in on the post and possibly changing it's direction, but is there a list anywhere of the differences (maybe major if there's alot) between 0.4.1 and the new 0.4.2 (test release?)?

Changelog.txt in your FO installation folder (if you're on Win), on OSX you need to peek inside the app package, it's located in Contents/Resources.

[Burkey]

Changelog.txt in your FO installation folder [/quote]

Holy p1$$1ng chr1$t!!!! Thats alot of changes

Shall read on my lunch at work tomorrow.... lol

Thanks, and feel free to delete my posts if they obscure they thread view

[davidescott]

The single production queue of all supply line connected planets is just genius and goes a long way solving problems of other 4X games.

Agreed. One of the nicest things I've ever seen.

at the moment, building warships is the only thing you can do with industry at most points of the game.

I don't see that as an awful thing. The awful thing is having to manually queue all those ships and then manually move them to the correct place. I don't see anything wrong with having 1000 ships vs having 20 ships, so long as they are not individual managed.

Feeding the player with awe-inspiring projects (buildings and ships) for hundreds of turns is possible, it will just be a lot of work and will take time to develop.

I don't think more buildings/more stuff to build is the answer. Player policy is what matters. One player may have a lot of planets with no buildings and no development and push all their PP into shipbuilding. Another might prioritize terraforming, or another transport/defense buildings (like lighthouses). More buildings runs the risk of being more things to have to build on each planet (and the current game logic doesn't prevent you from building multiple lighthouses in a single system which is annoying).

Making inhospitable planets hospitable is far too easy currently.

Agreed with reservations. Without the early techs to make those at least habitable the initial layout of the local stars will begin to dominate the game. At present you can choose to delay researching the "green" techs a turn or two if you get lucky with a couple good planets nearby, but if there aren't any good planets and the cost of green techs goes up those players will immediately resign.

The other concern is that by making that harder you encourage a mixed species empire. On its face that sounds good, but it complicates the growth management aspect of the civilization as you have to pick what planet to build your colony ship on. Perhaps the species could be selected at colonization time from any of the species that the empire currently has as members. Going further there could perhaps been something built into the evacuation process to repopulate planets and shift population around in order to get as many species on a good planet.

10000 stars can't be balanced, but i'm sure a satisfying number of stars can be.
Also eXploration in FO doesn't always have to stay linked to eXpansion.

I don't think 10000 can't be balanced, it would just have to be managed at a much higher level. I think that can be done in such a way that the higher level management would not play very differently from the current low level management of the game.

[eleazar]

When a game mechanic leads to gameplay that's tedious and uninteresting to the player, our game philosophy (see my signature) is not to introduce an AI or management layer, but to remove, alter or streamline the game mechanics so that they are interesting and not overly burdensome.

Although I agree with this design philosophy, I've been wondering for a long time now if it's possible to stick to that without any compromise ever. The problem I see is if certain game mechanics essential to a 4X game really can be designed in a way that they will smoothly scale up from early to end game.

This approach may not always work, (or we may not always be smart enough to figure out how to make it work), but it is what we should try until it becomes clear that it won't work.

So the rules get "adjusted" to prevent things. Where the first hull of the battleship costs 100 production units, the second costs 110, and the third 130. Where the first 5 colonies have such and such effect the next 5 have a reduced effect. Where the first ten scientific discoveries cost X the next ten have r*X. You end up fighting the exponential growth in output of the empire with exponential curves in all the other attributes, and all those carefully calibrated curves are only going to intersect and be properly balanced for a particular size galaxy with a particular density of planets/specials/etc. I think this is very anti-KISS.

Granted the current implementation is a little clunky, but some sort negative feedback loops and increased inefficiency with increased scale is realistic and useful. And it increases the significance of your decisions.

The problem becomes apparent when you take a parameter to extremes. A wide open galaxy with 10000 stars, 4 races, low monsters, high specials and planets. The game will completely break around turn 100.

Yeah, that's one reason why we don't offer 10,000 stars. I tend to think that 500 is excessive. If the carefully calibrated curves no longer intersect after X# of planets, then we can reduce the max number of planets to X + a little margin. Or just warn players that galaxies larger than X may kill playability. Currently i'm targeting 50-150 star galaxies as "standard".

I don't think it is at all practical, especially with the man-hours we have, to try to make a game that works well with any possible number of stars.

[Bigjoe5]

at the moment, building warships is the only thing you can do with industry at most points of the game.

I don't see that as an awful thing. The awful thing is having to manually queue all those ships and then manually move them to the correct place. I don't see anything wrong with having 1000 ships vs having 20 ships, so long as they are not individual managed.

In the latest version, you can queue up to 99 ships in a single production item, and they will all be produced at the same time, in a single fleet. A single build order can also be set to repeat a certain number of times.

Feeding the player with awe-inspiring projects (buildings and ships) for hundreds of turns is possible, it will just be a lot of work and will take time to develop.

I don't think more buildings/more stuff to build is the answer. Player policy is what matters. One player may have a lot of planets with no buildings and no development and push all their PP into shipbuilding. Another might prioritize terraforming, or another transport/defense buildings (like lighthouses). More buildings runs the risk of being more things to have to build on each planet (and the current game logic doesn't prevent you from building multiple lighthouses in a single system which is annoying).

If there are any buildings you find yourself building on every planet, that's a game design defect that needs to be addressed. I am however, all for streamlining the process of setting planetary focus en masse, selecting planets (perhaps also en masse) to be colonized without explicitly choosing a colony ship (even MoO2 had this feature), etc.

Making inhospitable planets hospitable is far too easy currently.

Agreed with reservations. Without the early techs to make those at least habitable the initial layout of the local stars will begin to dominate the game. At present you can choose to delay researching the "green" techs a turn or two if you get lucky with a couple good planets nearby, but if there aren't any good planets and the cost of green techs goes up those players will immediately resign.

The other concern is that by making that harder you encourage a mixed species empire. On its face that sounds good, but it complicates the growth management aspect of the civilization as you have to pick what planet to build your colony ship on. Perhaps the species could be selected at colonization time from any of the species that the empire currently has as members. Going further there could perhaps been something built into the evacuation process to repopulate planets and shift population around in order to get as many species on a good planet.

Some new controls for colonizing planets are definitely in order at some point. Eventually, I'd like to see a way to just select a planet (or a range of planets) and a species with which to colonize that planet (or range of planets) and/or possibly a location at which to build the colony ships, and then have those colony ships be enqueued and go colonize those planets without any additional user interaction.

[Vezzra]

I am however, all for streamlining the process of setting planetary focus en masse, selecting planets (perhaps also en masse) to be colonized without explicitly choosing a colony ship (even MoO2 had this feature), etc.

I want to second this...

Some new controls for colonizing planets are definitely in order at some point. Eventually, I'd like to see a way to just select a planet (or a range of planets) and a species with which to colonize that planet (or range of planets) and/or possibly a location at which to build the colony ships, and then have those colony ships be enqueued and go colonize those planets without any additional user interaction.

...and this. Unless we want to give up the concept of empires consisting of multiple species that thrive in different environments - which I hope we don't!

[davidescott]

In the latest version, you can queue up to 99 ships in a single production item, and they will all be produced at the same time, in a single fleet. A single build order can also be set to repeat a certain number of times.

And I use those heavily when I play, which is exactly what I think is wrong about them.

What do I want to have: I want to have a battle fleet with a particular set of ships 10 Attack ships, 5 fast intercept ships, a couple scouts...

What do I actually do: Either (a) produce 10x attack ships, 5x intercepts, 1x scout with repeat 10 times for each, and then move them to where they need to be. The problem here being that my production comes in spurts. I get 10 ships every 5 turns instead of getting a more consistent one ship every turn.
or (b) have a bunch of queues and then move individual ships around.

I'm saying it would make more sense to explicitly state the target, and the relative priorities of ships in that target, and then dedicate a percentage of production and just have it happen.

If there are any buildings you find yourself building on every planet, that's a game design defect that needs to be addressed.

Terraforming/Gaia, Gas Giant Generator and Lighthouses, basically all the late game buildings.

For Terraforming/Gaia the problem is that I really don't understand what benefit I will get from terraforming, but presumably having higher population is better so I just dedicate spare production to this.

Gas Giant I think this is a silly one... why else would someone put an outpost on a gas giant unless the eventual purpose was to build the building. Same for Asteroid fields... I would suggest the effect is free the moment you put the outpost on the building and have the required technology.

Lighthouses theoretically one only needs to build this on planets near the border or on the path from your central production to the border. In practice I end up building it in lots of places to deal with chasing down monsters and the like which are behind the lines.

... and then have those colony ships be enqueued and go colonize those planets without any additional user interaction.

The problem with this is that the ships will need escorts and protection, but the player is removed from the process and does not know when those escorts will be called for. I would dispense with the ships entirely and instead have a budget that goes to colonization vs development. Any planet within supply range that does not have any an enemy (monster or other playery) presence for the last 5 turns gets put in the queue... and when the budget reaches the next "colonization" level that planet is auto-colonized.

A similar thing could be done for invasions. If you can park your battle fleet in orbit over an enemy planet for 5 turns and it is within your supply range then auto-invasion should just happen and the enemy troups should start to drop until you take control. The process of building and escorting troop 20x troop ships, and then manually selecting the correct number is a real PITA.

Make it so that the only ships I have to think about are military units, and instead of producing them let me set my target numbers to manage. In my ideal game there is no production queue for anything. You can select systems for projects (like this blue/white star needs to have the shipyard project) so that I can make energy hulls, and this empty space area needs to have the starbase project so that I can repair ships there and defend this critical node.

[Dilvish]

I'm saying it would make more sense to explicitly state the target, and the relative priorities of ships in that target, and then dedicate a percentage of production and just have it happen....
[re colonization] I would dispense with the ships entirely and instead have a budget that goes to colonization vs development. Any planet within supply range that does not have any an enemy (monster or other playery) presence for the last 5 turns gets put in the queue... and when the budget reaches the next "colonization" level that planet is auto-colonized. A similar thing could be done for invasions. If you can park your battle fleet in orbit over an enemy planet for 5 turns and it is within your supply range then auto-invasion should just happen and the enemy troups should start to drop until you take control. The process of building and escorting troop 20x troop ships, and then manually selecting the correct number is a real PITA.

relative to this last bit, please note that with current test builds an adequate number of local troopships can be auto-selected for you to one-click-invade with. Regarding the rest, it sounds like you're wiling to throw out a great deal of tactics in order to avoid a modest bit of administration. Auto-colonize? Auto-invade? These sound both boring to me, and handicapping -- Choosing when and how to do these is often (though granted, not always) an enjoyable tactical challenge; particularly the combination of when and where makes these extremely significant decisions; I cringe at the thought of throwing that away.

Make it so that the only ships I have to think about are military units, and instead of producing them let me set my target numbers to manage. In my ideal game there is no production queue for anything. You can select systems for projects (like this blue/white star needs to have the shipyard project) so that I can make energy hulls, and this empty space area needs to have the starbase project so that I can repair ships there and defend this critical node.

Your vision really starts to sound like sim-space to me, with only a modest bit of significant decision making. Do you even plan to choose targets for the ships that you don't manage production of, or is that auto-selected along with the invasion targets?