what's the actual difference between 4x and grand strategy anyway? they seem like rather similar genres to me
 Shopping Cart Returner Shirt $21.68 |
 Tip Your Landlord Shirt $21.68 |
Shopping Cart Returner Shirt $21.68 |
what's the actual difference between 4x and grand strategy anyway? they seem like rather similar genres to me
>what's the actual difference between 4x and grand strategy anyway?
Autism
Grand strategy is a aimless sandbox about moving sliders and AFKing until something happens until you "win". 4x is all about dynamic, tactical decision making, navigating diplomatic challenges, building infrastructure and leaning into the strengths of your civilization to win multiple different and interesting victories.
TL;DR Paradox sucks, Firaxis fricks
4x is for games like Civilization. You start with nothing, and so do most of your opponents. First three stages are you vs your environment (Explore, Expand, Exploit), and last stage is vs your opponents (Exterminate). Can be turn based, or even an RTS.
Grand strategy is games like Hearts of Iron. Usually has an established political situation, and a focus on historical accuracy. The time simulation is also realistic, usually occurring over real days or months.
>not exterminating while exploring
Confirmed GSG virgin
Hey, no need to be snippy. If you haven't found an enemy, there's nothing to exterminate. But obviously you'll probably find at least one neighbor pretty quickly.
A 4X game will tend to just involve building up or producing the most efficient results in order to be the largest, most powerful nation. Some of the modern ones try to challenge this, but only in the most basic ways. In a good 4X game, you're spending more time rushing a specific goal like a science victory or something, rather than dealing with unpredictable shit or competing for victory conditions that you don't care about.
Grand strategy games will tend to try to get you to make best use of whatever resources available, with you having limited ability to build up said resources. New resources are generally just acquired by acquiring more land or whatever equivalent the game uses. In a good GS game, you're spending more time developing the nation by making use of whatever you have or can annex.
On the absolute most basic level, a 4X game has a win condition and a Grand Strategy doesn't
If I was going to make a more derogatory explanation, 4X is like chess where you have the specific goal of capturing the king and the challenge is getting there, and GSGs are like checkers where the end goal is just being the only one with pieces on the board
This doesn't necessarily mean 4Xs are all more complex than GSGs, but where they focus is entirely different
4x are gamey, gsgs are simulationist
A 4X has a randomly generated world, and generally follows a paradigm where players start out small, on equal footing.
A Grand Strategy game has a hand-made world, and generally follows a paradigm where the world is moderately to heavily populated by existing civilizations to reflect either historical/present accuracy, a plausible future, or a plausible fictional world, rather than symmetry or balance.
So if I put Civ on one of the pre-gen maps like Earth True Start, it's grand strategy?
It's true that grand strategy implies a full world that was deliberately made, but it hardly seems necessary. If I do random new world in EUIV and then just play a tribe for a century, is that 4x?
>So if I put Civ on one of the pre-gen maps like Earth True Start, it's grand strategy?
You have made it more like Grand Strategy, although the game's underlying mechanics are still 4x slop.
>If I do random new world in EUIV and then just play a tribe for a century, is that 4x?
You have injected 4X elements into a Grand Strategy game.
It's a continuum.
Can you see the entire map from the start? Is the entire map populated from the start? Are there more things to do than total war with major factions? That's a grand strat.
so technically, Total war games are 4X.
>are there more things to do [...]
Obviously TW is NOT a grand strat. But it's not a 4x either.
Field of Glory Empires is a grand strat
Field of Glory Empires is a bad board game which relies on arbitrary "victory points" as a crutch for the game not being fun to play as soon as you own more than two regions due to your turns taking like 20 minutes. Literally the only thing the game has going for it is the decadence/government tier thing as a unique mechanic for emergent empire stagnation. Fricking Rome 1 has better empire management
Filtered
grand strategy as defined by Paradox (so not the strategic war term but the term for a vidya genre) means pausable real-time strategy on a nation scale
rule of thumb:
mostly symmetric starts is 4x
asymmetric starts is gs
Starting small and slowly blobbing your way to being #1, taking on bigger and bigger enemies is so fun. I suppose it wouldn't work with the way most 4X victory conditions work
GSG typically has a preset map based on a historical period (or fictional history), with numerous factions of varying strength. 4X is a sandbox. Yes, this means Stellaris is 4X and not GSG.
The edge case would be Total War games, since their maps are semi-historical and the starts are assymetrical, but this side of Seleucids in RTW you still start out pretty small regardless of what you pick.
Stellaris is a blurring of 4X and GSG. Both in the original and current versions, how you build up a planet and develop a planet is more GSG inspired. I'd say the game is something like 70% 4X.
Total War is firmly 4x in my opinion. The lines have only blurred as of late because what were once npc rebel territories that existed only to be conquered by someone else are now generic factional territories that simply are not playable.
The way anon described it, TW only recently became more 4X, if anything. Cause Warhams did away with the whole "asymetric starts" and gave everyone exactly one settlement to start with
Shogun 2 also had pretty balanced starts. You usually had 1-2 provinces and at least 1 ally and 1 enemy.
>GSG typically has a preset map based on a historical period (or fictional history), with numerous factions of varying strength. 4X is a sandbox
Rather the reverse: Grand Strategy is an historical sandbox, whereas 4x is a game with set win conditions.
But yes, Stellaris is 4x and so is masters of orion.
Old World has assymetrical starts, but it's 4x rather than grand strategy. Of the two descriptions, grand strategy is the narrower one.
There is none, "grand strategy" is just a marketing term
this
I played EU 1 when it came out. I don't remember anyone calling it a grandiose strategy game at the time
plenty of people called it a gay strategy game though
Usually the depth of abstraction.
Grand Strategy games having some actual data to illustrate, has to design in a way that makes a difference between a Trade oriented coastal merchant federation and Autarky feudal mining and field crops based hegemony actually visible in a way that doesn't just assigns income to ownership, because neither of those could benefit from each other resources in the same way, something that 4X usually hand waves, so you can enjoy playing and steering the above in the way you want.
Eu1(and 2) were less GsGs and more Tabletop strategies.
Paradox keeps trying to make their GsG's more "arcadey" and welcoming for the wider audience which I assume includes 4X fanbase, and pic related sums up opinion on the issue which I share.
As for the spreadsheets gov and econ simulator Democracy4 comes close, and I think it may become a great game once the dev let's go of his "pure country and economy simulator untouchted by filthy real world party politics, coalitions etc." autism
One is a spreadsheet pretending to be a game, the other is a game where so much is automated it needs to use a spreadsheet to convince you that you're playing it. Neither require strategy, only imply it.
Neither are 'simulations' also, they again rely heavily on marketing and shills to be thought of as such. Play any driving or flight simulator; are games by PDX even remotely equivalent in terms of simulating what they pretend to? People learn to drive and fly for real with basic competence in driving and flight sims. The supposed political, strategy and economic games claimed to be sims do not teach any practical skills or knowledge of this kind. So why are they called sims? They're role-playing games, using spreadsheets just like in the days of pen and paper, except those games used to have some meaningful design attempts at representing real ideas.
4x has base building, GS doesn't
>muh diplomatic prowess
Spoiler alert: single player diplomacy is a fricking joke because AI is still so incredibly basic. Give it another 10 years and we'll be making diplomatic plays/deals with human-level AI using actual language and discussion instead of some arbitrary checkbox that decides whether or not an AI nation wants to trade with you.
The future of AI is going to change 4X and grand strategy games for the better, by a factor of basically 1000x. For once we'll have an immersive experience and I might actually enjoy gaming again...
>Give it another 10 years and we'll be making diplomatic plays/deals with human-level AI using actual language and discussion instead of some arbitrary checkbox that decides whether or not an AI nation wants to trade with you.
That's not what customers want, so that's not what we'll get.
>That's not what customers want, so that's not what we'll get.
Who says that? Immersive AI would make literally every game genre better. RPG games will also feel 1000x more immersive as you communicate with party members and NPCs you encounter.
It's also going to open the door to custom AI driven content for every other industry as well. Why wouldn't customers want an AI to be able to create custom tailored shows/movies/video games for them?
>Who says that?
All of the data showing that people prefer to get smashed by the AI 50+ times, then win through pure trial-and-error.
In addition, the very vocal majority of players in both 4X and GSG who just want a "comfy" game with big numbers instead of "tryhard" PvP.
You're assuming that human level AI means it will always be difficult. Why wouldn't there be varying levels of skill, just like in human players?
And why not just play a stronger nation and bully the minors if you want comfy while role playing within an immersive world?
>Why wouldn't there be varying levels of skill, just like in human players?
Because you don't need a tech advancement to simulate a glue-huffing moron.
>And why not just play a stronger nation and bully the minors if you want comfy while role playing within an immersive world?
Who knows. They just don't.
>Because you don't need a tech advancement to simulate a glue-huffing moron.
A glue-huffing moron with personality is literally still a vast improvement over current AI and still decades away.
Yeah moron, current AI advancements are nothing special and literally no different from your gay chatbots of 1995. Holy shit imagine being this stupid lmao
>AI hasn't improved since Master of Orion
>But it literally improved enough to beat the world's best chess and Go grandmasters while downsizing its hardware by a factor of 1000
Why am I wasting time on a mongrel who contradicts himself in the very same fricking sentence lmao
Congratulations, nothing you just said contradicted his point, he's still right and you're still wrong. There will be no """advanced""" AI in strategy games in the future. You don't know how AI is made and you don't even have a reason for talking about this subject. You don't even want """advanced""" AI in games, you wouldn't play a game that does have it.
This is the dumbest fricking thing I've read on /vst/ yet. Just another moronic troglodyte who thinks that given an infinite amount of time for technological advancement there will never be human-level AI because of some mystical/magical bullshit reasons. Do you think god put a hard cap on AI or something? kek
No shit they don't use the same kind of AI you moronic subhuman. Now why is it we've been able to create AI that beat pros at games like StarCraft? But you think a turnbased strategy is more complex? hahahaha imagine being this fricking stupid
>Computer games, ironically enough, aren't.
lmao
It's actually embarrassing how fricking stupid you kids are. You morons actually think AI just stopped progressing with your retro 4x games hahaha
Stable Diffusion and similar AIs are nothing special but for some weird reason it never existed until. AlphaGo is also nothing special. All those deepmind RTS AIs that beat pros are nothing special. Holy shit you're frickin stupid
>blah blah blah
I'll reiterate, because it's still exactly as true.
Congratulations, nothing you just said contradicted his point, he's still right and you're still wrong. There will be no """advanced""" AI in strategy games in the future. You don't know how AI is made and you don't even have a reason for talking about this subject. You don't even want """advanced""" AI in games, you wouldn't play a game that does have it.
>"IT'S TRUE BECAUSE I SAY SO, IGNORE ALL THE AI DEVELOPERS SAYING OTHERWISE REEEE"
Okay dipshit. Imagine actually being so fricking stupid that you're confident in the fact that AI has made 0 progress since fricking Master of Orion lmao
Now go back to play against easy AI in your shitty 4x games because you apparently can't stand the thought of someone beating your moronic ass without crying like a b***h.
>blah blah blah
I'll reiterate, because it's still exactly as true.
Congratulations, nothing you just said contradicted his point, he's still right and you're still wrong. There will be no """advanced""" AI in strategy games in the future. You don't know how AI is made and you don't even have a reason for talking about this subject. You don't even want """advanced""" AI in games, you wouldn't play a game that does have it.
>Now why is it we've been able to create AI that beat pros at games like StarCraft?
We haven't though, imbecile, not without so many handicaps and cheats they make Deep Blue original victory over Kasparov look complety uncontroversial. Literal fricking maphacks were involved in that "victory", which brings me to my previous point about perfect information.
Read what you just wrote moron.
>we haven't
>but we actually did, while also handicapping the AI
How many frickin times were you tards dropped on your heads as children?
If you can't understand why being able to converse with human-level AI in strategy or RPG games might create a more immersive experience then maybe you should stick with the easy level bots for the rest of your life
More cope from a brainlet who literally contradicted himself in the same fricking sentence lmao
>"you don't know how AI is made"
>t. literal brainlet who doesn't know a damned thing about AI and thinks there's been 0 progress in 20 years
Impossible to talk sense into an actual moron
>blah blah blah
I'll reiterate, because it's still exactly as true.
Congratulations, nothing you just said contradicted his point, he's still right and you're still wrong. There will be no """advanced""" AI in strategy games in the future. You don't know how AI is made and you don't even have a reason for talking about this subject. You don't even want """advanced""" AI in games, you wouldn't play a game that does have it.
>If you can't understand why being able to converse with human-level AI in strategy or RPG games might create a more immersive experience
I don't think you understand chatbot AI or game AI. Especially for civ-style games. That's really not how they work.
I'm talking about human level AI or post-singularity AI that can provide a fully immersive experience. Like playing against human players head-to-head but it's an AI and you can pause/resume the game whenever you want.
I think in a decade or two we'll get a basic version of it but obviously it's gonna take a lot longer for the fully immersive version.
It baffles me that people sit on /vst/ and criticize modern grand strategy/4x games as being cookie clicker bullshit and yet also can't seem to fathom how much gameplay would improve with AI personalities that provide an immersive experience, on top of improving other aspects of the game. i.e. being able to control your nation building via commands to an advisor instead of just clicking buttons all day.
>I'm talking about human level AI
still many light years away as far as we're concerned, at best we can make ai that can fool humans into thinking its humanlike, like text transformer chatbots, human-like ai is still way beyond the horizon. not to mention that you don't need human-like ai for a fricking videogame anyway. that's like trying to build a supercomputer capable of roughly simulating the development of hyperclusters across the cosmic web across the 13 billion years that the universe has existed just so you can play half-life 1 on it. its fricking overkill. at best, you just want a responsive ai that can flexibly adapt to your actions as well as the actions of other actors while also making sure it's not so good as to make it unwinnable. you understand neither ai nor game design
>post-singularity AI
you can't be serious. even if such were to happen, at that point humans would have been obsolete and you wouldn't exist anymore to play videogames
>not to mention that you don't need human-like ai for a fricking videogame anyway
This argument makes 0 sense.
Just look at RPGs for instance. You'd rather NPCs have a few scripted options for responses rather than an AI that is actually capable of having real discussions with you (in character)?
Have you never seen the movie eXistenZ? You guys really don't think there's a market for immersive gaming? lol
People have been dreaming about realism in video games since the dawn of gaming. People wanna see their little soldiers fighting in real cinematic battles and living real lives in their cities. Even the fricking Simpsons had a halloween episode where Bart and Lisa were gods of their own little worlds. I am not sure how you can honestly post on /vst/ yet still be baffled at the idea that there are people out there who want to witness these sorts of full immersion scenarios.
>This argument makes 0 sense.
no, you are just moronic
>Just look at RPGs for instance. You'd rather NPCs have a few scripted options for responses rather than an AI that is actually capable of having real discussions with you (in character)?
anon you don't fricking need human-level ai to do that. you just need ais good enough to fool people into thinking that they're more intelligent than they actually are. shit, text transformers aren't sentient nor are they actually capable of thinking, but they're good enough in producing outputs that can feasibly pass for an actual human response, the entire point of the turing test is not to gauge whether or not the machine is sentient, just that it's good enough to fool actual humans into believing that it's a human
tl;dr: you just need good-enough ai to fool people into assuming intelligence, not actual human-level ais
>You guys really don't think there's a market for immersive gaming?
read the fricking post again. not once did i say that there is no market for it or that it'd make for a shit game, its just that you actually don't fricking understand how ais work, and thus have a fricking moronic idea as to how to implement it. i mean, given that people are literally paying for what basically amounts to really advanced text transformers (cough cough AI Dungeon cough NovelAI cough cough), there is no doubt that a market for such games can exist in the future. the thing is that you don't want to simulate everything from the ground up and you don't have to do it either when you can achieve the goal of 'immersive and interactive ai that you can talk to' using far simpler methods.that costs far less calculation power. again, you just have to trick the person into thinking that your AI is more intelligent than it actually is.
>get proven wrong
>"noooo you're heckin moronic"
lol
Okay dumbass, if you ever step outside of your basement for once in your life you might come to realize that people want full immersion.
If it were up to morons like you we'd still be playing Pong.
>you just need ais good enough to fool people into thinking that they're more intelligent than they actually are
Imagine being dumb enough to concede this point but denying that we'd ever want it to get more advanced in the future. The history of gaming development alone proves your dumbass point wrong. You troglodytes really are single digit IQ tards.
>'immersive and interactive ai that you can talk to' using far simpler methods.that costs far less calculation power. again, you just have to trick the person into thinking that your AI is more intelligent than it actually is.
lmao
"noooo you don't need human level AI, you just need AI that's good enough to essentially be human-level AI"
Imagine being this utterly fricking stupid hahahaha
You literally just agreed with me.
>Nobody with a brain wants this
Says the moron who can't cope with the fact that he's literally wrong about everything? Full immersion fantasies are everywhere in media. There's no fricking sense in arguing with a dipshit who is completely incapable of being rational.
>they're looking for events and interactions the dev hand crafted.
Why the frick are you assuming the AI is going to just be some random box of personalities you brainlet? It's still going to behave according to the developer's build for it. It's just going to have more than a couple scripted dialogue options.
>Okay dumbass, if you ever step outside of your basement for once in your life you might come to realize that people want full immersion.
fricking moron, i didn't say people don't want full immersion. in fact, i fricking want my immersive games. its just that you misunderstand what ai really is and how it currently compares to actual organic intelligence.
>"noooo you don't need human level AI, you just need AI that's good enough to essentially be human-level AI"
'human-level ai' is not equivalent to 'ai capable of fooling humans' you dumbfrick. for one, human-level ai implies an ai that has the cognitive capacity of an adult human. 'ai capable of fooling humans' doesn't require human-level cognitive capacity to trick humans into thinking that it's intelligent. as i told you previously, and example of this would be text transformers. they're not fricking sentient, they're incapable of thought, and only generates speech by predicting what arrangement of words and letters follow with high probability, and they're good enough that they even fooled some morons into thinking that it's sentient
ergo, you don't need human-level intelligence to produce results that seem intelligent. you don't need human-level ai to beat grandmasters at chess, thus it follows that you do not need human-level intelligence to create an immersive and responsive vidya ai you fricking dimwit.
>Just look at RPGs for instance. You'd rather NPCs have a few scripted options for responses rather than an AI that is actually capable of having real discussions with you (in character)?
Nobody with a brain wants this. When people play Gothic, they're looking for events and interactions the dev hand crafted. They're not looking for whatever the frick you're handwaving.
>I'm talking about human level AI or post-singularity AI that can provide a fully immersive experience. Like playing against human players head-to-head but it's an AI and you can pause/resume the game whenever you want.
And you think they can artificially simulate an idiot's mindset and playstyle, in precisely a way that it makes the game easy, and not just unpredictable?
We literally already have that today. And why wouldn't it be possible, given a million years of technological advancement, to eventually create an AI that could simulate any human intellect? What is with all you tards and this belief that there's some sort of hard cap on AI development that will be impossible to surpass?
Listen moron, I'm gonna make this simple for you:
There is no difference between an AI that can artificially simulate human level of intellect and personality, and an AI that can genuinely/organically do it. You are arguing for the sake of arguing because you're a little pissant who can't stand being wrong.
I am not the fricking demiurge. I don't want actual AIs to suffer as individual units in my games so I can experience immersion. I want them to be playing as actors as per the centralized AI entity in the whole game.
>'human-level ai' is not equivalent to 'ai capable of fooling humans' you dumbfrick
No one said that, moron. Imagine being this utterly fricking moronic lmao
>ergo, you don't need human-level intelligence to produce results that seem intelligent
Again repeating the same moronic fricking argument that makes no sense. It doesn't need to be organic. It just needs to provide an immersive experience. Literally what the frick are you even arguing, shit-for-brains? Learn to fricking read you stupid frick hahahaha holy shit
How many decades of virgin rage causes someone to be such a sad sack of moronic shit?
>We literally already have that today.
We do not. We have AI with one priority list that just happens to suck, and then various bonuses applied to that AI to make them a challenge.
>And why wouldn't it be possible, given a million years of technological advancement
Why would I give a shit what happens a million years from now?
>There is no difference between an AI that can artificially simulate human level of intellect and personality, and an AI that can genuinely/organically do it
Fricking brainlet, that's not the fricking point. you seem to have a difficult time grasping this. Again, I will say this, you do not need AI that can simulate human level of intellect and personality for a fricking strategy game for it to be an actual interesting opponent. You do not need AI that is humanly intelligent for it to appear intelligent and capable in a game. CICERO exists, it can play Diplomacy and interact with humans, but it's not as smart as an actual human. It's just smart enough to trick people into thinking that it is within the environment it's contained in.
>You are arguing for the sake of arguing because you're a little pissant who can't stand being wrong.
nice projection lmao. perfectly describes what you're doing so far
>I don't want actual AIs to suffer as individual units in my games so I can experience immersion. I want them to be playing as actors as per the centralized AI entity in the whole game.
nobody fricking said that each and every individual unit in the game has to be an actual fricking sentient ais with their own little sentient minds you disingenuous mongoloid, and the fact that you even actually interpreted my statements as such just shows how lacking you are in the cognitive department. not to mention that ais cannot feel either pain or fear anyway unless you specifically program it inside them, which opens up and entire can of worms regarding ethical ai, but i digress
>It doesn't need to be organic. It just needs to provide an immersive experience
literally what i've been telling you, you dimwitted cretin, the whole problem is that you do not even seem to know how AIs actually work and function, or even basic game design for that matter, yet here you are throwing a shitfit when people point out what you got wrong, you fart-huffing imbecile
You still don't fricking get it.
I want to have virtual war rooms and face-to-face communications with AI rulers and commanders.
>you do not need AI that can simulate human level of intellect and personality for a fricking strategy game for it to be an actual interesting opponent
This does not solve the issue for people like me who want more immersion you moron. The point is to get as close to realistic diplomacy and war as possible, not another bullshit excuse of "well it's interesting enough, why do we need to improve anything? Let's just settle for cookie clicker bullshit despite the fact that we constantly complain about it on /vst/" you fricking dolt holy shit lmao
I'm done arguing with brain damaged morons. All I did was offer an opinion of how I wanted more immersion in my games and you lost your fricking mind and unleashed your turbo virgin rage all over the thread. Frickin delete yourself IRL if you can't handle discussion.
>This does not solve the issue for people like me who want more immersion you moron.
The point is that it can absolutely do exactly as you fricking wanted without needing human-level ai you fricking mentally-challenged bonobo. You don't need human-level AI because for our purposes we can create something approximating human behaviour even with algorithms that are less complicated than even insects. That's why I said having human-level AI is overkill and inefficient, because human-level AI implies that the AI has a full processing capacity of an organic human mind. It's like asking to recreate and simulate the entire universe just so we can better predict weather patterns on the next week or so, we do not need to overcomplicate shit to do the thing you want, especially if its for a fricking videogame. How do you still not understand this?
>"well it's interesting enough, why do we need to improve anything? Let's just settle for cookie clicker bullshit despite the fact that we constantly complain about it on /vst/"
And what you don't understand is that you're basically just daydreaming as to how it can be without actually understanding the logistics as to how it could be implemented, while absolutely not taking even the slightest bit of pushback from your idea like the fricking sensitive homosexual that you are.
>Frickin delete yourself IRL if you can't handle discussion.
What an absolute fricking clown, you keep arguing and arguing without actually having an idea as to how it can be implemented, let alone any idea how the thing you're talking about even works, and you're surprised that people keep telling you that you're fricking wrong, and then you b***h and moan like a child when you receive even the slightest bit of insult. Not only do you refuse to understand what i've been arguing, you deliberately avoid doing so while crying about how 'we can't handle discussion' when you're the one avoiding it in the first place, you delusional sperglord
Anon, you are arguing with a child. Ignore him and move on with your day.
Nice samegayging.
Or were you the moron who claimed AI never progressed past MoO then in the very same sentence conceded that AI had since developed to the point of beating chess and GO grandmasters while having the hardware the size of a pocketwatch?
What's the point of arguing with you negative IQ morons when you literally contradict yourselves in the same fricking sentence lmao
You realize that chess and 4X games don't use the same kind of AI, dipshit? That's why one progressed but the other didn't. Chess is a game for computers. Computer games, ironically enough, aren't. Chess is a perfect information game that can be beaten by searching a game tree. There is no game tree for a 4X game to bruteforce.
I'd tell you in 10 years we might have an AI as dumb as you at best, but we already had that 20 years ago as well.
Wrong.
Computer games can be solved via AI.
Open AI beat a bunch of Dota 2 world champions a few years back.
https://openai.com/blog/openai-five-defeats-dota-2-world-champions/#:~:text=OpenAI%20Five%20is%20the%20first,OG%2C%20at%20Finals%20this%20weekend.
In total, the current version of OpenAI Five has consumed 800 petaflop/s-days and experienced about 45,000 years of Dota self-play over 10 realtime months (up from about 10,000 years over 1.5 realtime months as of The International), for an average of 250 years of simulated experience per day. The Finals version of OpenAI Five has a 99.9% winrate versus the TI version.
https://openai.com/blog/vpt/
>crafting diamond tools in minecraft
OpenAI costs unironically billions more to develop than DOTA2. IBM's chess AI didn't cost that much to handle the actual AI, ignoring hardware development. You're unironically advocating for AI to be developed during game development using shitloads of money, probably running on hardware that is unreasonable for even a well moneyed dev to rent.
>unironically
>unironically
>unironically
First of all, shut the frick up. Second of all, it cost billions to develop OpenAI, not to train it on Dota2. Actually, it must have cost pennies, since they just did it for marketing.
>A glue-huffing moron with personality is...
Not desired by anyone right now.
Kek, the AI in 4X games hasn't improved since MoO, probably earlier. In the same timespan we went from needing a supercomputer to maybe beat a chess grandmaster to a phone being able to beat any human on the planet at any time. You're the same kind of moron who thinks stable diffusion is anything more than a glorified redraw tool and art is over, or that chatbots didn't exist 20 years ago.
Sorry, but no, not any time soon. Maybe in 25 or 50 years, not 10.
They do have great AI now, like Meta's Cicero, that can play Diplomacy. Still, it's one of the most basic social-strategy games around and is limited to 7 players.
Imagine trying to run a separate Cicero mind for every landed char in CK2, or even just for every nation in EUIV? And instead of just relaying moves or messages, it has to be continuous and always adapting? The AI in some of these strategy games actually isn't terrible. The issue is getting it all to process properly and efficiently. In order to do that, they have to strip down the number of AI "minds", which severely limits the playstyle of the AI.
And remember, you still wouldn't get to where you want to be. You want to reach a point where the AI plays like a dynamic, human player and you can send it unique messages and have it respond dynamically. We want the AI to literally play as if they were human.
>4X
Explore, Expand, Exploit, Exterminate
>Grand Strategy
Strategy on a grand scale
Simple as
Guys we'll have photorealistic graphics by 2000, just watch. Toy Story proves this.
Grand strategy games have asymmetric gameplay and 4X have procedurally generated maps, none of the other comparisons matter.
GSG is essentially PDX specific marketing gimmick. No no, you are NOT playing "just" a strategy game. You are playing "GRAND" strategy game. Same shit as with MOBA being coined specifically for just one fricking game and to rub it in the face of similar games and companies behind them.
https://en.wikipedia.org/wiki/Supreme_Ruler_2020
Keep in mind that there is a spectrum between GSG and 4X, much like there is spectrum between male and female. Some games just don't fit a binary GSG-or-4x paradigm, and that's okay.
name five examples
>in grand strategy games players have to micro-manage every aspect of the country
What gsg actually does this? Because games like EU4 and CK3 don't even let you control your tax rate.
gsg is made up term by parawieners so its shit by definition, they also call stellaris gsg while its pretty much civ tier 4x game
I think 4X games are boring and simplistic and rely too much on meta gaming and convention, what you can actually do and should do in a 4X, which is also the case for GSGs but GSGs I feel are the middle ground between 4X strategy and simulators (the same way 4X is the middle ground between RTS and GSGs).
Simulations are too close to spreadsheets for me to enjoy, and everything turn based, if not an RPG or obviously going for gameplay>realism, turns me off right away.
The only simulators I ever got into were DF (overflavoured and inferior to RimWorld, unironically) and Aurora (I could get into that because space age stuff). COM and all the tank simulator and German eastern front stuff always was too autistic for me.
I imagine 4X people think the same way about GSG.
I think I could get into a proper government and economics simulator, because that's where I expect spreadsheets, but a game like that doesn't exist.
>mfw reading screenshot of some bullshit site where someone posted a strawman between two imaginary sides of a moronic argument
DIDNT READ LOL
>didn't look at the filename and see its an ai-generated text
baka
4X games have the catan problem where (assuming everyone is competent) the starting position basically seals your fate.
yea unlike gsg where OPM has the same chances like Spain, oh wait.
4X map generators give 100x more balanced position starts than any gsg game.
Now re-read my post and point out where I said GSGs don't have this problem.
>is in 4x vs gsg thread
>posts jab at 4X in a way it implys gsg lacks of that trait
>surprised he gets misunderstood
Map generators can sometimes screw you but most of the time they just influence your game plan and expansion direction (which just adds to skill expression and gameplay variety), civ games arent solved nor perfect information game like chess where loss of few turns of tempo will pretty much lose you the game.
besides you can always use premade symmetrical map
You are in a thread where the OP is asking about the DIFFERENCE between GSG and 4x. When you say that 4x games have some problem, you are necessarily saying that GSGs do not have that problem, otherwise they would not be different in that way. God, you're such a gaytard.
Grand Strategy is 3X
That picture just paints 4xgays as whiny onions guzzling cucks.
don't blame me, blame the ai
Sasuga algorithm-dono
>4x
>doesn't require micro
Maybe if you're playing TRASH like most of the board.
gsg is for thinking people, 4x is a watered down version of gsg for smooth brains
then why is it that most of the time in a gsg game you're just sitting there with nothing happening while in 4X games you're almost always doing something?
its the other way around
4x games use hexagon tiles, grand strategy games dont.
4x is easier.
4x requires more thinking, gsg """gameplay"" is just learning convoluted UI and functions of gorrilion pointless buttons, wide as ocean and deep as a puddle
4x is Low IQ, high execution. gsg is high IQ, low execution
gsg is literally 4X on shitty premade map you braindead paradrone homosexual
I don’t know how anyone can say with a straight face that GSGs are harder. Even a decently challenging 4X like Civ 4 is far more difficult than any Paradox game. On the higher levels especially, you need careful planning, decision making, and, well, grand strategy to win. Choices feel impactful, particularly those in the early game. In Paradox games you can just spend the whole time RPing, not even trying to play optimally, and you can effortlessly paint the map.
Nobody says they are harder. They are more involved. There is no real idea of "win", it's arbitrary in a gsg. It is about the roleplay. Decisions not in a 4v4 Forest Nothing, but the world you might be familiar with.
Unless it's Hearts of Iron (any of them), then it's back to micro skills, and it shits on any 4X in terms of difficulty.
all grand strategy games are meme games unless played MP which barely anyone does. you could teach a monkey to play them once you get past the ui.
>all grand strategy games are meme games unless played MP which barely anyone does
When you get a large enough group together, they're the best games in existence.
4x is for fake autistics.
I'm sure 4x fans will be devastated to hear that they are not in fact autistic
Dilate.
In most grand strategy, there is no EXPLORATION. EXTERMINATION is very abstracted. EXPLOITATION can be basic or detailed. And all EXPANSION is pretty much through war.
grand strategy is paradox-like
4X and Grand Strategy have significant overlap, so often, a game will be both. The primary distinguisher is that a 4X game must have all 4 aspects of the "ex"s in its name, where as grand strategy does not.
Grand Strategy will often have exterminate, as most of the time you are fighting a war.
Grant Strategy does not need exploration, at least not in the sense that most 4x's have. For example hearts of iron 4 is a popular example of Grand strategy, you do not really "explore" since you know the whole map, at best you gather information in certain regions but that's not really exploring so much as just expanding. For example, Diplomacy is not a 4x game, but is grand strategy, there is no exploring and you see the whole map.
Grand Strategy does tend to have expansion, however there are some games where iirc you do not actually increase territory, rather you have a set limited amount and you use them to eliminate the opposition. Technically you do not expand, as your forces only exterminate and you gain nothing other then that.
Grand Strategy tends to have exploit, but there are cases where you could not have it, see the example given in expansion. The difference here being instead of claiming territory, there are no resources to utilize and develope your forces, so you can claim more land but that may be the objective of the game without you being able to use it for anything.
If a game is lacking in any single one of the 4, it can't be called a 4x game. However if it does still have the preqreuisites of Grand Strategy, it can still be one and not the other.
Please note this applies the other way around as well, as some games don't qualify as a grand strategy. For example, a game with all 4 requirements of a 4x that has players starting on exact equal footing on a symmetric map, would be considered 4x, but not grand strategy.
In short: they are usually similar but not always.
grand strategy: all land is under someone's control
4x: most of the land is terra incognita (this is one of the founding four X's: eXploration)
unlike the stupid parts of europa universalis
i'm not a grand startegy knower, so i didn't know there's grand strategies where half the map is no man's land
however there's still minimal exploration since you know the americas look like the americas, meanwhile 4x games usually randomize the map so you don't know if the land you're moving into is a fertile field or an empty desert
>i'm not a grand startegy knower, so i didn't know there's grand strategies where half the map is no man's land
So why bother making an uninformed post? EU4 allows you to randomize the shape of the Americas. Also, if you are not a girl, you have a very feminine writing style.
>feminine writing style
frick does that even mean?
wrong thread?
No capitalization, missing punctuation. That's how women type.
i'm not a woman, i'm just a lazy ESL
Look at AI roguelike. It uses text and imagegen to make a semi-coherent thing.
AI also won diplomacy recently, https://www.science.org/content/article/ai-learns-art-diplomacy-game
A tactical roguelike with controls similar to Baldur's Gate where you have full control over the party would be nice.
In most roguelikes, even if you can have a party, the party members are strictly AI.
Grand strategy games have a huge learning curve.