RTS direction

[joe_da_cro]

I still think this is a valid question: "What is the minimum necessary to enable a satisfactory PvP experience?"

the more i read this the more i think factorio should play out like a cross between RTS and MOBAs

[piriform]

I had to look up a definition of MOBA game, and found this: http://www.urbandictionary.com/define.php?term=MOBA. If this is what you have in mind then I'm confused. MOBA focuses on Hero development via xp mechanisms and killing of minions in personal combat. Base and infrastructure building is not really a big part, whereas in Factorio, that currently is the part.

[Drury]

MOBA is also known as Action RTS, or ARTS for short, for it's high emphasis on micromanaging a single unit. This is in line with Factorio.

[piriform]

Thanks! Just goes to show you what preconceptions will get you. All along I had this notion that the engineer character would play the role of a builder King. Essentially vulnerable but able to form and direct armies. MOBA style of RTS would be quite a switch.

[joe_da_cro]

I had to look up a definition of MOBA game, and found this: http://www.urbandictionary.com/define.php?term=MOBA. If this is what you have in mind then I'm confused. MOBA focuses on Hero development via xp mechanisms and killing of minions in personal combat. Base and infrastructure building is not really a big part, whereas in Factorio, that currently is the part.

this is what factorio is about when you play with the character. the difference is xp is translated into research, and the more direct killing nests for making research packs.

This is why i keep thinking of MOBAs for PVP (if thats what the devs want). But earlier i was suggesting using strategic tools so you can almost automate combat if you take the time to set up which fills the RTS niche but more importantly fulfills factorio's automation core think of it like the player setting up the AI directly for assaulting the biters/other players. You set up the attacking task force, using interface you provide the path finding, and designate the attack area/patrol area.

The idea of it is that is if you know you are being attacked by a nest northeast of your factory you can set a factory to build and harass that base so that they are not able to send forces to your factory. the factory will keep replenishing losses until the player stops the factory from moving forces to the enemy or they forces destroy the base.

[piriform]

this is what factorio is about when you play with the character. the difference is xp is translated into research, and the more direct killing nests for making research packs.

Hmm, I can see some parallels between research and say, power armor and weapons, giving the character enhanced abilities. But I'm still struggling with the character's role. Is the character's principal job to wade in and splatter. Or is (s)he to work behind the scenes creating the infrastructure to build armies which will then do the deed.

You set up the attacking task force, using interface you provide the path finding, and designate the attack area/patrol area.

All the elements you mention: way points, patrol zones, defense zones,designated attack points, etc are good points and I personally agree. Whether they are essential is a good question to ask. Notwithstanding all that (unless I misunderstood your proposal) I'm not sure that integrating them into a fixed script is desirable. Good players will quickly learn to exploit any fixed script and negate its utility (then again I'm NOT a Good player )

[Zackreaver]

We are considering the rts direction of the game a lot.

Builder robots
Player could just order stuff to be built and where. It would be up to the builder robots, to get the needed materials from the logistic system and build it. This would be combatibile with blueprints (some kind of ctrl-c ctrl-v system, or even blueprint as item), building would take some time, and the more robots the player have, the faster it is.

Fighting units
Ordered like in rts, some kinds of robots/automated etc.
I could imagine, that these robots shouldn't be craftable like other items (I don't really like the idea of having few tanks in my pocket and placing them in front of the enemy base). Fighting units would still need ingredients and time like normal recipes, but would be only buildable in special buildings, and instead of producing items, the unit would exit the building.
This means these wouldn't be minable as well.

Throne room
Player would enter it, and he could start control the game in rts style, at this phase, he would already have robots for building stuff, and units to fight, so he would just order what needs to be done in bigger scale.

That sounds like my kind of game.

[joe_da_cro]

Hmm, I can see some parallels between research and say, power armor and weapons, giving the character enhanced abilities. But I'm still struggling with the character's role. Is the character's principal job to wade in and splatter. Or is (s)he to work behind the scenes creating the infrastructure to build armies which will then do the deed.

this is something the devs will have to work out. The need to define what they want the player character to be and equip it accordingly. AN option here which would tie into the RPG thing, would be to have choice research trees. so for the sake of the RTS discussion research which buffs armies (which makes the player behind the scenes type) or Research which buffs the player( which makes the player the wade in and splatter type). The game at the moment allows for infinite amount of tweeking because it has elements of many types of games built in already. it will really take baby steps and await feedback move from there.

All the elements you mention: way points, patrol zones, defense zones,designated attack points, etc are good points and I personally agree. Whether they are essential is a good question to ask. Notwithstanding all that (unless I misunderstood your proposal) I'm not sure that integrating them into a fixed script is desirable. Good players will quickly learn to exploit any fixed script and negate its utility (then again I'm NOT a Good player )

the best game i can relate to when describing it would be homeworld 2. you gave the order to build a interceptor squad, 5 ships will be built and form a squad. you can give the squad move commands waypoints patrols etc but when it came to fighting they flew around fighting other ships on their own in reality there is little interaction once fighting started. but even homeworld 2 doesn't have everything i described.

[Lallante]

I have an idea for the RTS element of Factorio that is in keeping with its existing features and (relatively) easy to bolt on.


You would construct robot factories which would be assigned recipes for particular kinds of fighting robot, and would have a "target" designated using the minimap. You would then feed these factories their recipe resources and the factory would spit out the fighting robots, which would then "attack move" (think RTS order) to the target and back. While you could redesignate the target of the factory, and any returning robots would be assigned the new target, you wouldnt be able to control the attack robots directly.

In essence you would need to build enough robot factories to create a wave of robots that could overcome the spawning of the biters.

The robot factories should be lategame, and create immense pollution so biter spawning increases to face the threat.

[MalcolmCooks]

I was thinking about this a little today, and I'm not sure that traditional style RTS would really fit well with the rest of the game. Such an idea is great for PVP but the biters aren't really an enemy worthy of sending a robot army after.
So how about this: your fighting robots are actually large and powerful, but can't act autonymously. They are controlled like vehicles. You don't get in the robot itself, but into a seperate robot control center building. This lets you act as if you are driving the robot like a vehicle, but you aren't inside it. Then you can send your robot stomping along to wipe out biter bases and (unlike using a tank) if it gets destroyed or runs out of ammo you won't die.

[piriform]

the best game i can relate to when describing it would be homeworld 2

+1

Such an idea is great for PVP but the biters aren't really an enemy worthy of sending a robot army after

+1

And this is where (my) confusion lies. Are we talking about PvP or PvE, both, neither? A clarification from the development team (or perhaps a poll) would great;y help in structuring this discussion.

[joe_da_cro]

well up until late game how do most players get rid of biter nests? turret creep is the most widely used tactic to clear large biter nests.

so the rts element would remove boring micro in order to clear nests in the mid game. in the late game they can also be used to supplement the player to clear nests.

I also like the idea posted by MalcolmCooks. having the player control vehicles like a drone would a good start.

[ProphetofEden]

Lately ive been thinking of awesome it would be to have war-machines that i could command onto patrols to guard areas or to assault areas, but i would rather big machines then little robots like we have. like you have a dedicated hanger per war-machine and each war machine is customizable, they would need to be very complex and very powerful basically customizable tanks that you could command.

[The Phoenixian]

I've been mulling over the RTS side of Factorio for a while and here's the conclusions I've made so far.

1: The logistics equipping and maintaining a fighting force, be it mobile or stationary, should be just as important as the combat itself.

This fairly simple in concept; Tanks need fuel, lasers need power, guns need ammo. It's worth noting that there is room for more convenient and easily maintained weapons in the arsenal but they should be weaker for it in general. If a laser can be plopped down and forgotten about, then it needs a disadvantage to compensate for that logistic simplicity.

To examine how that should come into balance, I think it's worth considering how four types of combat units might work. Namely, Stationary, mobile, aerial, and naval.

• Turrets.
  
  A turret is stationary. It needs to be manually setup which requires the player journey to the point of emplacement, set everything up, and deliver supply lines to it. How much supply is necessary varies: Lasers need very little in supply while a rocket battery might need quite a lot. In general however, there's a lot involved.
  
  We've all played with these but it's worth detailing explicitly as the balance of turrets forms the baseline against which other forms of weaponry should be measured.
• Mobile armor
  
  A tank is a mobile unit. It must be produced and shipped to it's area of operation, driving short ranges, but using either supply depots or being loaded onto a train or ship if the range is too long. Once in place however, it isn't bound to it's supply center and can freely move around within it's fuel range to defend outposts or attack alien encampments.
  
  Additionally, If you have an outpost using tanks and C&C style mobile harvesters one might be able to set up mines quickly from a central depot, but there's a tradeoff in that one might well need rearmament and reinforcement schedules to ensure the mines remain consistently defended. If the tanks stay too long they'll run out of ammo or fuel and need to rearm. Likewise some contingency must exist to replace combat assets lost in action and ship them to the front.
  
  Because tanks are more versatile than turrets and require only a single dedicated depot for a large number, they should have a comparative disadvantage to any turrets with the same their resource cost and weapon type.
• Air power
  
  An aerial combat unit is similar to a mobile unit but it's likely that its range and response times will be vastly greater, in addition to being unimpeded by terrain features. It seems plausible that a bomber would be produced at base, and sit at a local air field until ordered to attack or respond to a situation and generally be expected to reach all but the furthest situations even on the most explored and developed maps.
  
  With the extreme range of such a unit even from a central location, it seems likely that it should come with a significant disadvantage. Simple loss of combat effectiveness might work for a tank, but for a plane the possibility of rapid centralized spam means that even more disadvantages should be needed.
  
  I think that here traditional RTSes have a few pointers. Aircraft often require more in the way of logistics or even permanent base emplacements. For a bomber to require a 8x5 sized supply and landing zone that cannot be shared with other units would quickly come mean that any large fleet of aircraft would need significant space provided for it at the home base. Likewise higher fuel, construction, and/or ammo costs would give less responsive units like tanks a comparative advantage.
  
  Additionally, two different kinds of disadvantage for two different kinds of units help to make them distinct.
• Naval units.
  
  Boats come into a rather different place than tanks. I imagine that it's obvious that combat boats would come into play alongside water based enemies and cargo ships but while there may be seafaring enemies water resources aren't certain. As such while tanks have both a potential offensive and defensive roles, a gunboat would likely serve almost exclusively to defend with the possibility of the rare beachhead. I imagine most of it's time would be spent patrolling near cargo ships in order to protect them.
  
  Logistically, setting and supplying a gunboat means it would be produced at or near a dock and as a shoreline is a requirement for machines to interact with it, a boat could only be resupplied at a dock as well. Thus most of a gunboat's logistic needs can be made similar to that of a cargo ship. The two types can have similar fuel storage to consumption ratios.
  
  That said both the fact that boats can only be used on water and the fact that, absent hovercraft, aircraft will be the only other unit able to do so together mean that boats already have a potent disadvantage. As such their combat capabilities can be made far closer to parity with stationary emplacements, or even exceed them, depending on how naval combat works.

2: The combat balance of Factorio will likely be similar to that of Tower Defense games.

This, I think, is far more obvious than the previous point. Single player Factorio combat is very similar to tower defense and it's only natural that it have a similar balance.

As has been mentioned before, different aliens should be resistant to different weapon types. The mass of meat that is the larger biters may boast high laser resistance while a spitter might outrange a gun turret. Additionally, burrowing aliens could easily bypass walls and only be damaged by the concussive force of explosions, as a mirror to the flying enemies of other games.

Where I think Factorio could come into it's own in this vein is that these resistances could be less of a factor in weaker enemies and become more pronounced as blue and green alien variants become common. If all biters have laser resistance but larger biters have a much greater proportional laser resistance while spitters have longer and longer ranges as they reach their peak. (Which was originally proposed by the devs, but shot down due to players wanting gun turrets to remain viable. As long as they have a place this should be fine)


3: PvP would need to be fundamentally different from traditional RTS games.

This is just a note, but I feel that while PvP is an eventual possibility, making it work in Factorio seems like something that requires drastically reevaluating the genre. This is a many tens of hours long game and with a traditional RTS style it would be too easy to simply cripple the opponent's infrastructure early on.

Certainly it's worthwhile to think about how various combat situations should go. The central factory for example should be extremely hard to crack at any level of development, while a supply line can be much more vulnerable. As part of that, a stationary emplacement should get the best of any mobile unit when trading in equal weights, possibly requiring several times more mobile units to destroy than it requires turrets to defend.

That said, quite a lot of work is needed to have a way to keep conflict interesting in a long game without having a 1v1 becoming a race to destroy each other as soon as possible. (For example, it's easy to imagine the game becoming a micromanagement fest that's ten times worse than most as players strive to build their factories as optimally as possible for half an hour before the fighting starts.)

Given the design work needed for PvP, while consideration can be given to it, single player and cooperative play should be the standard everything is built for, at least for the time being.

[Tobie]

I think a good and (relatively) simple intermediate step from Factorio as it exists now to a Factorio with more RTS elements would be something like this:

1) Add unit production facilities
- Constructs units in much the same way that assembling machines create items.
- When the unit is constructed it is spawned at a designated area adjacent to the facility (maybe using an arrow like inserters)

2) Add rally points
- Add the ability to designate a rally point on the map.
- When a unit production facility has a rally point, all units produced at said facility will automatically move to that location when they spawn.

If this system works out well it could be expanded upon to include things like rally point chains, patrol paths, or conditional rally points (ex: units only move to rally point B if there are 10 or more units at rally point A).
For an example of this type of game play you can see videos of supreme commander or planetary annihilation on YouTube.

I think that this type of system would lend itself well to the automated nature of Factorio.

[cosss]

I vote for this direction! I don't see any fun running around the buildings. I'm buidling a system here. Like in TTD.
Also it would be really cool to have ability to automate military units. Like in Screeps but in a more casual fashion.

[jarcionek]

I have an idea which might be relatively simple to implement and unlock a lot of possibilities.

Let's add two new items: a humanoid robot and a command center.

The robot is effectively a player - it has its inventory, tool belt, can be equipped with armor, weapon etc. Robot would have some simple AI - similarly to robots in logistic network - and perform simple tasks such as repairing, rebuilding, transporting items between logistic chests and additionally - shooting nearby aliens.

The command center is a building which the player can enter and take direct control of any robot. Controlling a robot this way would look exactly like the game looks now - the only difference would be that the robot can be killed in single player with no consequences (i.e. no game over).

At the very basic implementation, it effectively offers a teleportation. I can leave a robot in every mine that I build and instead of travelling between mines I just switch to control different robot. It could be also used as alternative to logistic robots to transport large amount of resources between chests.

We can expand this idea further by upgrades/technologies. One type of upgrades would increase the potential of single robot (e.g. let it use the weapon, usable items like grenades, or increase its operational range). Another type of upgrades would go into RTS direction and allow to give simple orders to robots which are not directly controlled, such as "follow and protect player or another robot". Imagine playing single player and attacking enemy base in a group of 10 robots, each equipped with modular armor, throwing grenades etc. It can be expanded even further into more complex orders such as "capture and fortify a territory" or "claim ore deposit, build mine, build train station and connect it to the rail network".

[joe_da_cro]

I vote for this direction! I don't see any fun running around the buildings. I'm buidling a system here. Like in TTD.
Also it would be really cool to have ability to automate military units. Like in Screeps but in a more casual fashion.

you can currently play in sandbox mode and you do not have to run around. But i definitely like to have systems where biters can be cleared with decent effectiveness without the player/character allowing for sandbox to be more viable for marathon games.

[Krayzie6682]

YES YES YES!!!!

[kyranzor]

I add my support for an RTS aspect to be added to Factorio.

Cheap shitty mass-produce-able droids footsoldiers could be manufacturable and given simple orders such as patrol from A->B, or move to X location, defend X location (only chasing a short distance if any).

droids should be manufactured in a dormant state though, which can be transported by trains/other vehicles in the same way as items, and then by inserting into some kind of building, they become activated troops and are ready for orders.

The different weapons, equipment, ammo, unit types, vehicles and a more integrated "faction" system would make the game quite awesome - there could be some form of population cap (even though I really hate pop caps) to avoid the issue of infinite troop production - have some simple building which needs a lot of power that acts as a support structure to increase how many droids can be active at once.

The player can fight the biters by sending out a never ending wall of droids to clear the death-world swarms, while running along with the droids to help out and have some fun.

In multiplayer with PvP, the game would become a battle of grand-strategy - a battle of attrition, where the ability of the opposing player/s to expand their resource outposts and keep their assets protected is paramount. Their ability to manufacture troops fast enough to fight back, to coordinate defence strategies and coordinate pushes against enemy outposts, crippling ore supply to the automation beast that is their main base.

Sneaky tactics like a distributed base network with no particular large/important base would help avoid a fatal blow if the large base was to fall.. delivery of train-loads of droids to the battlefront, setting up robot-spammed turret and wall networks..

It would truly be amazing, and be a whole new level of factory management to prepare and feed the hungry war machines. It just has to be done well and I believe Wube Software can do this without ruining the factorio feel. The robot soldiers would have to really gel well with the rest of the Factorio game. The player's character should always be a major part I think - either they should be in the battlefield winning ground, or in the factory expanding production to win the war of attrition. Of course all of this only matters for PvP, if it's just against the biters there may need to be some improvements to the biters to support/encourage robot army production/coordination.

If Wube doesn't implement an RTS style game mode, then at least enough of the mod API should be implemented that modders can fill in the features that many are asking for.