>>2506165
I won't, because there is nothing wrong with snowballing. It is logical and the alternatives tend to feel bad, be gamey, and feel artificial, and would make early game decisions worthless. I am fine with mechanics that resemble empire decay or diplomatic alliances to resist a dominant faction though, as long as they are not entirely arbitrary and artificial.

I think most games underestimate how expensive it is to maintain a professional military with expeditionary capabilities
Other forms of military recruitment are harder to use for wars of conquest, or rely on factors beyond more territory = better
The level of militarization your society has should be culturally dependent and it should be hard to maintain it at a high level purely by player action and not external forces
There should also be a large variation in how much tax/resources can be extracted from conquered lands. EU5 had the right idea in building an entire mechanic around this (control) but the AI is terrible at using it and the way it interacts with the rest of the economy is nonsensical

Victoria 2 figured this out. You have to model population, their needs, an economy. Before you say this would make a game too complex, don't worry. Viccy also figured this out. Most of the stuff happens below the hood. You don't directly interact with artisans, who make wine. You only interact with them through taxes, tarriffs and technology.

>>2506165
By making the player suffer from the same as big empires suffered: over-extension. Owning more territory should either make you less efficient at owning them or at threat of losing it more easily (internal politics? Far-away regions wanting more autonomy?) It's a hard balance and I think it may be impossible without fine tuning. Take Civ V for example, every veteran player swears by 4-city tradition because of how the city scaling works: heavily towards 'tall' until a very 'wide' can catch up.

>>2506165
if you want to stop snowballing just quit the game and start another

>>2506438
Civ 5 is a bad example because it's very artificial, Civ 4's maintainence scaling with city distance and number of cities is better

>>2506165
On games where where campaign is on one map it's bound to happen, unless you at some end game threats like Stellaris or other games sometimes do, but it can also be annoying and not that interesting.
Now, for games where you fight on different maps/missions, that's not nearly as applicable. Still can be, if you can take trained up units, heroes, extra resources from past games, but still, not as impactful.

>>2506165
>How would (You) solve the Snowball Question, anon?

By playing Civ V with Vox Populi. Seriously the mod makes the AI play to win, it will change its objectives according to its situation and it will try to stop whoever is winning if it has no other choice

>>2506270
>and would make early game decisions worthless

That only happens if you are trying to halt snowballing by giving AIs bonuses, the only alternative is simply to make the AI smarter

>>2506473
calling something "artificial" in a game is meaningless

>>2506491
>>2506493
Man I wish we had the technology for smart AI in videogames. This timeline sucks.

>>2506502
you wouldn't enjoy playing against smart AI because it'd make you feel like a dumbass

>>2506165
>With so many gsg and 4X games, it's way too easy to translate an early game victory into continued success.
I think this is true for pretty much any game, getting an early lead will make it easier to close the game out unless you make a mistake or the opponent has a way to adapt and counter your lead.

Like even in a simple running race if you get ahead and you can keep up then victory is pretty likely, unless you tire yourself out before the finish line or trip or the opponents have greater reserves of strength than you then you could argue that the race has already been decided at the start. But a obviously a running race is way more complicated than this.

Early game victories should matter and matter A LOT, they are there to reward good game awareness, to keep every player on their toes, to punish carelessness and overly greedy plays. Or else the early game will be an absolute snore and arguably the outcome decided more by the circumstances the players have been given instead of the players being in control of the game.

>How would (You) solve the Snowball Question, anon?
This is usually solved by adding parallel gameplay routes. Alternate objectives and victory conditions to close the game instead of the bog standard "KILL THE ENEMY!!!" which can be accomplished while being on the defesive. Civ games (if we are going there) usually have different victory conditions, Religion, Science, Culture, Eco, Diplo or whatever.

Also this video with examples
https://www.youtube.com/watch?v=NUlOe46yz60

>>2506165
MoO2 solved this with Antaran invasions. Just have an off map threat that goes after the first place guy. Sins of a Solar Empire sorta did it with pirates, but you could pay them off and they were arguably a good thing if you had a strong fleet since they gave exp there.

The worst solution is making AIs form retarded coalitions or just absolutely castrate themselves to attack the player, fuck that.

you could make a lot of easily acquireable territory appear halfway through the game

for example if player A has 70 land and player B 30, and another 100 land suddenly appears and is split between them evenly, then the new score would be 120 to 80. you just have to make it so that the land isn't acquired by the players in proportion to the land that they already have.

>>2506502
>Man I wish we had the technology for smart AI in videogames. This timeline sucks.
>This timeline sucks.
Because video games can be the way you imagined? You talk about some epic first world problems dude.

>>2506508
Forming coalitions is actually the most logical choice since they would also happen in real life wars, whoever hates that concept probably adores blobbing and thus dislikes challenge.
The worst solutions are giving bonuses to AI or making the AI play a completely different game compared to the player, like it happens in titles where the AI ignores entire mechanics because they don't affect it

>>2506512
Player A would probably just conquer that land faster and gain an additional advantage, unless of course there are mechanics like overextension or rebelllions

>>2506165
I'm not sure if it can be fixed in a matter that's fun, because curbing it would come at the expense of player agency. Snowballing is the result of two problems compounding:
>the player typically has meta knowledge and employs more optimal strategies during all stages of the game and the player's skill competency is consistent
Historically the competency of a nation's ruler, statesmen and generals would vary greatly but in video games it is consistent with player skill
>the player gets to rule as a near-omniscient dictator with immediate access to information that's presented in an incredible elegant way
Fixing this by adding arbitrary province modifiers of bureaucratic corruption or denying players access to information would just be frustrating. Games don't typically model the upkeep of empires very well, there's a token period of unrest after conquering a province but after a few turns/years you've pacified it and it turns into a net cash contributor that adds to your snowball.

>>2506165
Internal politics, autonomy/control, and coalitions. For the first you see in pretty much every strategy game that the feature is lackluster and easily circumventable or otherwise rendered mute, a country in EU4 or 5 or Stellaris or whatever can start with a specific religion or government, dynasty etc and just keep it up with no issues whatsoever for centuries unless the player intentionally moves towards an outcome, there are no agents in the country with their own goals and ideals so every country is permanently in stasis.
For control and the contribution of land to the country I think EU5 did a very good job popularizing the idea but way overshot in its implementation by being too simulationist leading to the player having to care about every road or the proximity and control of individual locations and also causing shit like pseudo-feudalism via vassal spam.
In the case of coalitions it has every single time been designed as just a fun little feature where a bunch of guys jump into a rapevan and organize an expedition to fuck somebody over, with seemingly nobody having any reason to fight at all or any particular desires and of course only the leader being negotiable. I think the potential of coalitions is completely squandered with how nobody has goals or reasons to fight and therefore no dynamism, you won't see countries switching sides every summer because their priorities changed or because they desire to see the fighting remain in deadlock for as long as possible or anything. I'm not sure how I would implement this myself since I think the way PDX designs their gsgs would make things like constant sideswitching cancer for the player and that the value system( +70 -67 = 3 acceptance = yes) is inherently really bad at simulation and making the AI act reasonably, hence also why in so many PDX games you have dozens of interactions hidden behind -200, -1000 base value so the player can never pick them.

>>2506513
What else should I complain about on a video game board? Being hungry, dirty and forced to live in a slum more your speed?

>>2506473
A lot of things in Civ V are very artificial (gamey), but that's beside my point. I just wanted to exemplify that systems that try to prevent snowball by territory need to be fine tuned to prevent 'pigeonholing' the gameplay. In Civ V you either go very wide or 4 cities.
By what I can gather Civ IV incentivizes lots of cities so the scaling doesn't prevent snowballing effectively. Expanding is (almost) always rewarded.

>>2506682
wide is better overall in civ IV but you can't do infinite city spawl, you have to build up your cities for them to pay for their maintenance

because its not fucking FUN for the gamer to just get a game over when they are 4 hours in on their campaign and rebellion + war happens

>>2506165
put the win condition slightly after the peak of the snowball, not at like 65% map control

>>2506730
>peak
or brink or whatever you want to call it. the point of "can't lose this"

Nobody said ck2 which I think does it the best, expanding always means you have to make a vassal stronger . If you combined that with Victoria pops it would be pretty good at stopping snowball.

>>2506802
Vassal management in CK2 is really easy though and by the time you're emperor you should have so many retinues that your vassals will never rebel because your army ratios are so skewed

>>2506493
AI-only mechanics are an entirely different barrel of fish and even more cancerous.

>>2506270
The problem is the degree of it. Obviously, good play and decisions early in the game should be rewarded later, but most games in the genre are placed at an extreme where only the first like 20-30 turns actually matter and everything afterwards is a waste of time because the winner has already been determined.

>>2506845
Mods fix this with factions mostly but in the late game I could see that

>>2506165
Kinda like real life. Accumulation of Advantage.

>>2506165
How to prevent snowballing? Easy. Make it real time. Let's see you try and manage 100 cities at the same time as you're fighting a war, managing diplomacy, etc. With this, the AI doesn't even have to be halfway intelligent. It can just swamp you with infinite APM and you've got yourself a good challenge.

Distance based penalties to everything.

File: Steam Achievements.jpg (110 KB, 1920x1080)

110 KB JPG

>>2508196

Add in Achievements and meta progression that only unlock while you aren't in 1st place, so you always have incentive to keep playing.

>but that won't fix snowballing
It'll fix ragequitting when you AREN'T snowballing.

>>2508153
this is actually what passes for """""game design"""" these days, is there any wonder new games aren't even worth getting hyped about?

>>2506165
Attrition. In most game it is too easy to kill the enemy without losses or recover any damage received. Other games have victory be the necessary outcome of a battle- two factions enter one faction leaves, which is also bad. It is fast but it promotes unstoppable victories whereas retreats allow falling back and regrouping/reinforcing. Having lots of undecided battles would gradually grind down either army. Paradoxically, units should be more deadly in the sense that there are no units that exist to plink each other and be mowed over by stronger units. Even some shitty guy with a sling or scout craft should be able to damage some enemies and make you wonder if it's worth taking the damage to defeat them.

Economy wise the problem is that science is invariably tied to industry/money. Separating production limits, science speed, and income would more or less end snowballing too. The essence of the problem is more area = more everything. If you wanna be radical, make large amounts of the map completely useless. For space games, it's best when factions have different kinds of planets that are useful to them, but terraforming is too much of a workaround for this. A better thing would be considering just how much output a whole united would could produce and let a properly built world be able to take on a small empire by itself so if you or the opponent finds themselves with a handful of worlds they can just go all in on military and defense. Going from all econ or sci to military should require time and effort alongside the attrition so defending enough would get them a break as their opponent had to fix their armies.

Mercenaries and temporary alliances against a strong enemy should be a part of the game too. I liked how galciv 2 would erupt into multi party wars to keep something like a statue quo.

There really isn't an answer bro, if you've been outplaying someone all game, it massively sucks to get fucked over late game because of anti snowball mechanics.
Being able to snowball is the reward for consistently making good decisions, you can't remove it without creating kusoge

>>2508602
>Economy wise the problem is that science is invariably tied to industry/money. Separating production limits, science speed, and income would more or less end snowballing too. The essence of the problem is more area = more everything.
Despite being a blobber, EU3 actually had a decent answer to this. But Vicky 2's (and Imperator's) tech system is the best example.

>>2506802
>Nobody said ck2
Because the thread is about 4x games and not grandiose strategy games?

>>2506508
>MoO2 solved this with Antaran invasions
that's essentially just giving the computer player free units to attack you. if the anties aren't attacking your planet, you're probably already losing the game.