|
Lawman 0 posted:Let's have a debate: What 4X strategy had the best/most enjoyable AI opponents? The same as the best 4X game overall: Fall From Heaven 2 <Insert your favorate mod mod mod here> for Civ4
|
#
¿
Mar 16, 2021 23:58
|
|
|
|
| # ¿ May 16, 2024 13:02 |
|
|
Comstar posted:I got The Old World and I'm still doing the endless tutorials. Building specialists in rural buildings also pushes your borders out, assuming they're at the edge. Urban buildings do this automatically when they're completed.
|
|
#
¿
Apr 10, 2022 04:02
|
|
|
Splicer posted:You gonna tell me Age of Wonders: Planetfall should have spent their dev time on something "more productive"? Jumping in to say, I'm not sure who would count AoWP as a game with unit design. Customization via mods, sure - and it's really well done. But I can't decide to swap the Vanguard Troopers kinetic rifles for laser rifles, at least not in game. So it's not really a unit designer IMO. On the other hand, I think the above is informative to game designers of exactly why you would or wouldn't want to add unit design to your game. Just by adding 3 mods to predesigned units, the developers have created an insanely complex and cool set of interactions for players to have fun with. From a design perspective, it requires restraint in that you only let the player design the units within certain parameters, but also requires commitment to the excess consequences in the sense of A) designing a large number of qualitatively different modifiers for the player to unlock and play around with B) being able to guarantee that the complex interactions of the above won't cause any blatant bugs, unbalanced situations, hangs due to deadlocks... I think the MOO1/MOO2 designer had something similar going on. It felt wide open, but when you got down to it, you had a narrow range of ship sizes you could use and only a few different good component options to consider. HerpicleOmnicron5 posted:IIRC you can make missiles to intercept other missiles separately from the specialised anti-missile missile armament. Is that not right? I remember having to do all sorts of weird poo poo on an Uncreative run or whatever the race creation choice was that removes choices from the tech tree. It's been a long time since I've played MOO2, but IIRC, missiles can never target other missiles. The vast majority of weapon systems in MOO2 can at least *try* to hit a missile (if they're not HV modded). My favorite creative MOO2 point defense system is the disposable frigate self-destruct.
|
|
#
¿
Apr 26, 2022 23:57
|
|
|
The I still managed to take total control over Russia, most of the US, end the war in Ukraine, have humanity first cede half of Ukraine to Russia anyway (why...), had Russia annex Georgia because it was controlled by the stinking quisling alien collaborators, put a few space stations in low earth orbit and also founded the first and only colony on the moon. Fun! I also thought the UI and tutorial were horrible dogshit for the first few hours until I learned how to play. I'm fine with the UI now, but dang did I think it was garbage for those 2 hours. e: I done had the wrong game woops. Lowen fucked around with this message at 22:09 on Jun 16, 2022 |
|
#
¿
Jun 14, 2022 23:10
|
|
|
Garfu posted:I just got a cloudflare error from SA for trying to paste the changes from the pdf the developer posted to the second page of a thread on the steam forums (wtf), so here's a link to it instead: https://we.tl/t-gpYixUNZk3 Hmm, let me try... Nope, same issue here. I even tried converting the text to from UTF-8 to ANSI and pasting it it, same deal.
|
|
#
¿
Mar 1, 2023 21:29
|
|
|
Infidelicious posted:A lot of words to say Planetfall is a 6/10 tactical game married to a 4/10 4x Only if you bump both of those numbers up by 4/10.
|
|
#
¿
Apr 3, 2023 23:49
|
|
|
Clarste posted:I'm not joking. I think fighting battles in a 4x is universally a tedious waste of time so I opt out if I can. Apparently Planetfall was the wrong game for that, but I'm using this to dispute the notion that it's 8/10 separate from its tactical battles. If someone tells me a game is 4/10 my expectation isn't just that they don't like it, my expectation is it's broken and basically unfinished. Even when just talking about one aspect of a larger game. I've played plenty of games that were 8/10 or better games in my opinion, but that I returned simply because I could see they weren't designed for my tastes. (I know you didn't say 4/10 yourself but the principle still applies) I can say this: if someone doesn't want to play a mix of tactical combat and 4x, I don't think they'll like AoW3/Planetfall at all. I don't think that means the 4x part of Planetfall is bad, it just means it's not a game for them. Like, units in Planetfall are really different from each other. They're more like different cards in magic the gathering than different units in Civ, and these differences mostly relate to how they play in the tactical game. So if someone is just skipping playing the tactical game, it's like they're outright missing at least half the game, and then playing the remaining half not knowing what any of the units they're building or mods they're applying really do beyond the results they get back from an AI vs AI fight that they probably didn't even watch. I try and rate games by how well I think the developers did making their game. When I look at just the 4X part of Planetfall I see a game with well tuned balance, lots of asymmetry in the different factions, a huge unit roster with both strategic and tactical variety, UI design that competes with the best 4X games UI, and an AI that plays the game well enough to be fun for a player with 100s of hours. These are goals the biggest 4X studios don't fully reach. That's why I rate the 4x part of Planetfall at 8/10. Why not 10/10? I think the 4x element of PF suffers from setting up every battle the same: with both armies fully on the field from the first turn, with every battle either ending only when one side has nothing left, and no real option to retreat or chase. Some other games like Sword of the Stars and Humankind have done a better job with battles that can end for reasons other than one side wiped out and that last over multiple turns. SotS even allowed any unit to freely retreat at any time, it just needed to move away from the front or to certain points on the map. I think there's room for improvement on this front in future games that integrate 4x and tactical combat. Then there's assorted other stuff like: shuffling population around is too much management, engine limitations mean only 1 instance of tactical combat possible at a time and pauses map movement, building navies vs building other units is not much of a choice. This is all 4x layer stuff, even though all but the population shuffling issue touches tactical combat. That's how inseparable the 4x layer and the tactical battles are. victrix posted:the strategic layer is a very bad UI for what should be a slay the spire series of pitched battles awarding new units and upgrades change my mind Having a bad UI and saying the strategic game should be removed entirely are two different things. I like doing the 4x stuff like exploring and clearing the map, putting down and managing my cities, research, diplomacy and building new army units. So getting rid of all that would make the game worse for me. The Age of Wonders series has worked like this from the first game. The entire Total War series, Sword of the Stars, Master of Orion 2 and the entire Age of Wonders series have all had a 2D 4x map layer with cities, army recruitment, diplomacy and research to give context to the fun tactical battles. These games/series are all somewhat popular (or were, back in the day). I think what these kinds of games need most is more freedom for the player to setup different kinds of tactical battles, not less. Having something like the above as a streamlined mode for multiplayer or quickly trying out different armies would be cool though. PS: I do not like *any* HOMM game or Heroes Hour. I haven't been interested enough in any HOMM game to try and found Heroes Hour incredibly boring! But I'm not going to tell a fan of HOMM or Heroes Hour that they're 4/10 games. They're clearly well made games. Just not ones I like/am interested in.
|
|
#
¿
Apr 4, 2023 05:45
|
|
|
Infidelicious posted:If you think the strategic layer of planetfall is an 8/10 experience you must love the deep strategy and compelling choices in the children's board game Candyland. Sorry to double post but I'm honestly curious - what's your favorite 4x game, what rating would you give it out of 10, and what specifically about the strategic layer of Planetfall gets it a 4/10 from you?
|
|
#
¿
Apr 4, 2023 06:25
|
|
|
Infidelicious posted:Probably Alpha Centauri, 9.5/10 SMAC is one of my all time favorites. I think that even if you don't like Planetfall, saying SMAC is a better made game than Planetfall in this day and age is bananas. SMAC was a great game when it came out, but by modern standards it's barely acceptable. They're very different games so if you like one there's no telling if you'll like the other, but if I compare just the strategy layer by game balance/overall polish and content then compared to SMAC, Planetfall has: An AI that can actually play the game both economically and engage in warfare without tripping and falling on its face. Actual balance mechanics to prevent ICS, and no crawler spam. More unique factions compared to SMAC, each with their own unique unit roster, special rules, and special buildings. Unit customization (modding) that actually does something other than just bigger numbers (yes, even if you just autoresolve - because autoresolve in Planetfall *always* plays out the entire battle using AI on the actual battlefield). Units that are actually different from each other and serve roles on the battlefield other than "has a higher attack/defense" number, such as buffing/debuffing/healing units, or being good at ranged or melee combat. Some units specialize in strategic roles, but might still get pulled into fights, such as scouts or stealth units. Actual choices of how to develop cities and specialization based on nearby sector types and how development fits into your overall plan, vs the SMAC strategy of just squeezing as many cities as possible into the space you have and building a more or less standard queue in each. Cities actually grow and have different shapes because of annexing different sectors when population grows, rather than all of them just being a weird cross shape from day 1. Actual choices to make in terms of placing and growing cities, what sectors to annex, what exploitations to build, and which sector specializations to put there. More in depth and varied combat with fights between 2-42 units total. Each of these things represents work the developers of Planetfall did that the developers of SMAC did not, like the large variety of units, or something they succeeded at where SMAC failed, like the AI and game balance. No matter which you like more, you shouldn't look at the decades old basic mess of SMAC and say it's a better made game than Planetfall. Better for the time, maybe. More to your taste, sure. Better made based on the same criteria, no, absolutely not. I'm not saying you have bad taste, and I'm not saying that you just need to learn to love my ~favorite~ game, I'm just saying you can dislike or be totally uninterested in Planetfall while still acknowledging that the developers did a good job making it and really knew how to make the game they wanted to make. --- Same deal for any other comparison you would make. I love SotS, I think it's a great game. But despite both being a 4Xs with a tactical game, SotS and PF are very different. So again just comparing overall quality/amount of content of the strategy layer of SotS with just the strategy layer of Planetfall, SotS has: Bullshit autoresolve results that don't make any sense. AI diplomacy that is just broken and sucks. A 4X layer so basic that I've solved it, and might as well not bother playing it at all, because I just follow the same optimal development strategy every single game. You seriously think this is compelling on it's own? OK, I guess. Unbalanced random events that just gently caress over new players while being totally uninteresting speed-bumps for experienced players (except when they also gently caress over experienced players). etc. Again, I love SotS, I think it's a great game, but it mostly seems by luck and accident. It boggles my mind that someone would actively enjoy the 4X part of SotS and think that it could stand alone without the tactical combat, but that's just my personal taste. What isn't my personal taste is, SotS just doesn't have nearly the level of polish and refinement PF does, there's barely anything even there to refine, and what is there shows spots where they just objectively hosed up, like the diplomacy. Lowen fucked around with this message at 02:33 on Apr 5, 2023 |
|
#
¿
Apr 5, 2023 01:07
|
|
|
Jabor posted:If your interpretation of how good something is boils down to "did someone put time into creating it", then you seem like a literal caricature that enjoys receiving participation trophies. I have no idea how much time any dev spent making anything, and I don't care. When I talk about the quality of Planetfall I'm talking about the extremely high level of skill and care that went into making the game, not the number of hours spent. e: if you enjoy basic rear end simple 4x games vs brain dead AI then I would say that it is you that wants participation trophies.
|
|
#
¿
Apr 5, 2023 02:39
|
|
|
Jabor posted:This is participation-trophy poo poo imo. Hmm yes putting a large variety of high quality content in your games is just not technically impressive, it's better to do a poo poo job with a small amount of stuff. OK, sure, whatever.
|
|
#
¿
Apr 5, 2023 03:48
|
|
|
Jabor posted:It doesn't actually matter from a "how good is the strategy layer" perspective. It's like Paradox selling you extra ship design packs. No, I'm saying they actually know what they were doing and put the effort in because they could see that it would actually make the game better, and then I compared the skill in doing that favorably to a game that was designed like "lol just reskin Civ2 and graft some additional stuff to it, lmao". Oh also if you think the units in Planetfall are anything like ship DLC packs for Stellaris you have NO IDEA what you are talking about.
|
|
#
¿
Apr 5, 2023 04:49
|
|
|
Jabor posted:A lot of that variety is irrelevant from the strategy layer perspective, though. A unit costs X resources and improves your army by Y (increasing the range of opposition that you can get a favourable battle against). A unit upgrade costs Z and improves your army by W. An economic improvement costs Q and doesn't improve your army at all in the short term. That's the essence of the strategy layer - you spend resources to improve your armies, and how much you do that (in order to snowball via military means) vs. investing in economic buildings to snowball via economy is the meat of the game. It's not irrelevant from the strategy layer perspective, because the tactical battles *still happen* even if you autoresolve, so all those differences will always have an effect on the battle results. You literally can't stop the tactical battles from happening, you can only ever delegate them to an AI vs AI fight that is resolved nearly instantly but otherwise plays out exactly like it would if you took control yourself. Watch a replay sometime if you don't believe me and compare it to the summary you get, they're the same. There is no such simple cost/benefit calculation for a given unit because unlike other 4X games units in Planetfall have a wide range of abilities beyond just damaging attacks, and even the attack abilities are qualitatively rather than just quantitatively varied. So having a good mix of support units, or attack units that complement each other, will always boost your effectiveness in a more interesting way than just giving your stack a buff or something similar other 4X games do. You just can't ignore the variety of units in PF and claim that you're just concentrating on the strategy part, because army composition is done on the strategy layer. As for economic vs military choices, Planetfall has a reputation as a wargame in 4X clothing, where the correct move is to always forsake butter for guns, but in my experience it's actually one of the better 4X games in terms of balancing economic vs military development. Even just looking at the economic development game in Planetfall, I really enjoy how the different sectors fit together, and planning out my cities so they can grow without blocking each other. It's kind of like the district placement game in Civ6 but without all the forced choices.
|
|
#
¿
Apr 5, 2023 05:46
|
|
|
Jabor posted:"more complex" does not mean "more interesting". I literally did not use the word complex in any of my posts.
|
|
#
¿
Apr 5, 2023 06:26
|
|
|
I'd like to mention the reason I started this whole argument is because some poster said Planetfall is a 6/10 tactics game with a 4/10 4X attached, or something similar. Even if they were just rating based on their personal enjoyment, they didn't disclose that, and I just think that's a lovely thing to say about a game just because you personally don't like it. That's the only reason I've been on a posting crusade about this. I've not been trying to argue that people are wrong to like playing SMAC or any other game in 2023, or wrong because they don't enjoy Planetfall, or anything like that. I've said this several times, but I think that message has been getting lost. Infidelicious posted:I can't flood the world via pollution and block efforts to stop it in the UN, or ruin someone's nutrient yields and cause a famine by terraforming a mountain next to them in Planetfall; so it's an objectively worse game. So I saw that you're disengaging from this conversation and that's cool, but for anyone else still reading: I look at the terraforming system in SMAC, and the way the UN works and award it a few _objectively_well_made_game_points_ that I wouldn't give to any other game. GlyphGryph posted:This criticism is weird to me, the SMAC unit designer specifically doesn't involve much in the way of "bigger numbers" because that part is all handled automatically, the designer is specifically there for you to change the other stuff that isn't just "bigger numbers". Balancing specializations against costs with conditional stuff and stuff that alters how a unit plays completely. I'm not gonna say it couldn't be better, just that this criticism doesn't work. I'm mainly talking about weapons/armor, or later chassis mostly just giving you a higher move speed (yeah I know that they also change what terrain you can move on, that's +points for SMAC too). A lot of the mods are things like "50% better vs aircraft" which is just one step removed from "number bigger", and it's honestly not much of a choice to add it or not to an experienced player. That's not really much of a problem though, if unit customization in SMAC was just the mods, I would have no issues with it. One of the problems the SMAC unit designer creates compared to Civ2 is: units tend to get minmaxed as just defenders or just attackers due to the cost system. If the unit designer didn't exist or didn't also decide what weapons/armor a unit got, then the designers could just put some balanced units in the game (balanced in the sense of not being all armor or all weapons) along with the attack specialists and defense specialists. Civ2 had plenty of units like that, but if I try that in SMAC the cost of my unit balloons beyond all reason. It just strikes me as kind of careless, like they wanted to give the player some freedom but didn't really know how to balance that, or provide a good mix of viable options, so we just end up with even less unit variety in the game than if they just put premade units in. quote:How does the Planetfall customizer work? You start with one of the premade units in the game, and then you add 1-3 mods to it. Mods need to be researched and cost a special resource to apply (the cost goes up with stronger mods or to apply them to stronger units). Each mod adds a special ability or trait to a unit that you can usually only get from that mod, and also has some secondary +stat effects like more defense or generally stronger attacks. Not every mod can be applied to every unit, and mod effects can interact with each other. Examples: Arc Extension Module can only be applied to units with arc (electricity) weapons, and makes it so attacks with those weapons "jump up to 1 enemy targets within 2 hexes on a successful hit, dealing 50% damage." "Additionally, this mod grants the unit 30% increased damage on all abilities.". Stun Module "Attacks from Arc weapons have a 8 strength chance( 4 for repeating attacks) to Stun non-Ethereal units for 1 . If resisted, Static Charge is applied instead." Additionally, this mod grants the unit 30% increased damage on all abilities." - I could use both of these mods, and then the stun effect from the second mod would also jump due to the first mod. A few mods add things like the ability to detect stealth units on the strategic map. It's a lot of fun finding synergies and good builds with various units and mods. quote:By modern standards it is still absolutely better than average and more than acceptable, but that's mostly because the modern 4x standard is "really goddamn bad". I haven't played PlanetFall in particular, but the ones I have played SMAC is still hands down better than them pretty much every single way. Yeah, this is why I said that thing at the start of the post. Lowen fucked around with this message at 20:42 on Apr 5, 2023 |
|
#
¿
Apr 5, 2023 20:40
|
|
|
GlyphGryph posted:Unit customization in SMAC is effectively the mods, deciding if you want a specialist attacker/defender or if you want someone who can do both, and creating "specialty" units that do something other than standard fighting, like a psi-combat guy (which lets you focus almost solely on unit skill and morale instead of combat tech) or an espionage unit or a rover terraformer or something. Everything else is not only automated, but it even lets you automatically upgrade all future units and usually most of your existing units with no thought required at all. It's bad because it makes the clear optimal strategy "spam best attacker + (optional) a few defenders as needed", and then you win the game after hours of mopping up. Your best attacker might be a weapon or psi unit, and might change from game to game based on your faction and evolve with tech or your civic choices, but whatever it is at the time, the best move is to spam building just that. Building balanced units is not the good strategy that you think it is, because a unit that attacks or defends loses HP from the same pool, which makes the next attack or defense less effective. Building two separate units is always the optimal choice because it gives you twice the HP in two different pools. One unit with balanced stats can not compete with two min-maxed units in SMAC, it's not even close. So because of the way the unit designer works, the strongest option in SMAC is also the cheapest one. Bad! quote:This, the way you are describing it here, sounds absolutely miserable, btw. I trust it is better in the game than the explanation makes it sound like? Because that sounds simultaneously mind numbingly complex and completely worthless at the same time. I love it and think it sounds awesome, but it's very different from anything in any other 4X game that I know of. It's not even in Age of Wonders 3, and I don't think they're bringing it back for 4. So I understand it doesn't appeal to everyone. quote:What is the actual decision making element going into that sort of customization? That's the part that makes a customizer interesting, the stuff you're describing above seems like a good way to make a game worse depending on how its handled. What do you mean? Because the mod system is nothing but decisions, and none of them are uninteresting ones because there are qualitative rather than quantitative differences. For example: Maybe (for the sake of argument) I've figured out the absolute mathematically best combination of units and mods possible, and say (again for the sake of argument) that this optimal strategy does NOT include putting rocket pack mods on battlemechs. Does that mean I will decide to never play the game by putting rocket pack mods on battlemechs? Obviously not! It may not be the mathematically best, but it is fun, and plays differently from other strategies I could follow, and the game won't massively shift in difficulty just because I am or am not doing the optimal thing. quote:Like if I were to try and talk about the SMAC unit design, I'd talk about the major early game decisions: Mobility and Versatility vs. Affordability and Maintenance, Traditional combat vs PSI, when you want to build artillery units, and how the mods play into tailoring your units to counter specific threats and how you should constantly be shifting your army makeup around specifically to prevent your opponent from adapting to your threats with those mods. The bad combat balance of the game is at least partially attributable to having a unit designer, because by creating it and including it in the game, the developers lost the ability (lost, if we're using civ2 as a reference point) to fine tune the costs of specific units as special cases (because costs have to be calculated according to universal rules to work with the designer rather than a per unit basis) and lost the ability to curate the specific combinations of attack/defense values/special abilities included in the game. These are all tools the designers could have used to balance the game, but that they just didn't have because they don't work with the unit designer present. Alternatively they could have done a better job with the designer (in terms of cost calculation or a different system of building units entirely), resulting in better balanced combat. Or they could change some other mechanics and fixed the problems that way, but they didn't.
|
|
#
¿
Apr 5, 2023 23:54
|
|
|
deep dish peat moss posted:It's more about interpretation of data. It can interpret data in ways that humans never could and spot patterns in it that Humans wouldn't be able to. So while the responses might be synthesized based on human-written text, it's the connection between data points and response that separates its "intelligence" from the data it was trained on. Doctors are going to be one of the first jobs to be fully replaced by AI because it is literally impossible for a doctor to keep up with the speed of medical research. Are you for real? Do you seriously believe this? Why? What was so convincing that you heard or read it, looked at the current assortment of AI bullshit around right now and thought: "sure, this is going to replace doctors, they're going to be one of the first jobs to go"? What is the source of this incredible nonsense, I honestly want to know. quote:A ML-driven Civ AI would absolutely crush 99.99% of players right now if that was its goal, because it could correlate the data from across millions of games to devise an unbeatable strategy and prepare itself for every counter that players throw at it. What are you basing this statement on? Has anyone made a good machine learning AI player for any Civ game that uses player game replay data, or done anything similar? If so, did it actually crush 99.99% of players? quote:That kind of AI wouldn't be fun to play against for the vast majority of players. That's not because it's a "tactical genius" or whatever, but just because this theoretical AI has watched every Civ game ever played. Some players would be able to beat it, sure, but over time the tactics they used would stop working. There isn't any kind of ML that works this way at all. No one is going to get good results by just feeding something a large volume of training data and doing nothing more. Machine learning developers spend a lot of time and effort tailoring training data for a reason. Simple stuff like getting ordinary people to label images via Recaptchas is only part of that. When it eventually works it might feel like magic to the end user that doesn't know how it's done, but of course it's not, there's a lot of human work, ingenuity and knowledge involved. The typical 4x game is: huge (compared to most studied games), random, asymmetric map, completely unexplored and invisible, except for what's near to you - played by 1 human player that plays in a distinct way compared to the AI players, and none of these players really have any kind of concrete goal you can assume they're trying to optimize for, like winning. The human players all have wildly different skill levels, and they're mainly just trying to have fun, whatever that means for them, and will probably change their strategies from game to game. They will often deliberately not play strategies that they know are the most effective, because they're too effective to be fun, or because they're how they won last game - and then they'll win anyway, even though they played in a deliberately not optimal way. The AI players are designed to provide a fun challenge to the humans, they also aren't trying to win, except insofar as it makes the game more fun for the player if they try a little bit. You could just look at multiplayer matches but then you run into some of the same problems, some new ones, and there are a lot fewer examples to use compared to single player. There's no number of games you can train an ML on like that that are going to teach it to play the game well. MAYBE if someone put a lot of work into isolating some small specific task, and really carefully tailored the training data for that, then they might get something useful, like an AI that can manage to do an OK job moving an army in a small area, with the goal of capturing a city. But someone needs to do some work isolating tasks and grading examples. I'm guessing it would be faster and easier to make a conventional AI that does the same thing, and if the game changes a little it's easier to tweak that. Good AI of any kind, even ML AI, doesn't really play nice with computer game development, because unlike traditional games with stable rules, computer game rules change often in an iterative, experimental process. Game mechanics are constantly being added, removed, or tweaked, and balance values are constantly being changed around. ML for computer games is something ML researchers might do as a flex, and only with games that meet a rare set of criteria (stable rules, preset, symmetric maps, lots of good data from competitive, 1v1 matches... So basically, just Starcraft). I think developers as a group could do a better job of making conventional AI players that are fun to play against, but that's only because I've seen a lot of mixed quality in 4X AI - a few games are actually good, but most are bad. The fact that so many come short of what I've actually seen done by small, modest teams makes me think that the main problem holding things back isn't any sort of technical limitation or lack of funding. If Firaxis (as an example) thought that the quality of their Civ 6 AI was unacceptable, then they could have allocated resources to improve it by now. They haven't done that because it's working well enough according to their standards. And I think they kind of have a point. My standards are absurdly high due to the rare and unusual amount of time I've spent playing these games, going back to Civ2 when I was a kid. Most players just won't have anything like that number of hours. Even so, Steam says I spent just short of 500 hours playing Civ6, mostly single player, and it's not as if I was bored and swearing at how bad the AI was the entire time, just for the last 100 hours or so  .
|
|
#
¿
Apr 9, 2023 04:54
|
|
|
Mayveena posted:Somebody read it and tell us what it says This is getting off topic for the 4X thread, but eh. The premise hinges on the idea that AI can be improved so that it generates quality assets for video games, or can at least help in some way. This is supposed to change the video games industry more than any other entertainment industry. There's nothing I've seen in this article or elsewhere that leads me to believe that AI can create quality assets. I've seen lots of different cool stuff made using machine learning, like AI dungeon, funny deep fakes, automatic rotoscoping of videos based on a single hand drawn style prompt keyframe, and all sorts of other stuff. None of that translates to making quality video game assets. For technical claims like this I would like to see some support from sources in the field of ML AI, game asset creation, or both. There is no such person quoted in this article, and I think anyone can guess the reason. The article, paraphrased: "Games need a lot of content so: AI content generation will change the games industry more than anything else." ...Then there's a paragraph that just lists companies that have added features to their software that they claim use AI in some way... ...And another paragraph just barfing out "company y has added feature z that is vaguely AI related in some way."... "some unnamed executive predicts that small firms will invent new genres only possible with AI tech" "AAA studios might have an advantage with big marketing budgets and use AI to generate several high quality games per year rather than just banking on one hit" "Large publishers can work around copyright concerns in training data by using their large reserves of past assets that they own" The final paragraph is pretty well described by SIGSEGV re: unions.
|
|
#
¿
Apr 9, 2023 23:12
|
|
|
What's CORSICA specifically? It's not a very searchable term. GlyphGryph posted:No one in research is gonna give a poo poo about researchers developing any specific 4x AI, so the earliest we're going to see anything from that direction is an AI that is given a baseline understanding of how to play games *in general* to some extent, which can then be few-shot specialized by letting it play against itself for whatever games they want to throw at it. And there's definitely folks working on that! But it's not happening any time soon. I don't think it's wise to believe a technology is possible until it's demonstrated, there have been too many cases of hype or outright fraud. What you describe is different from any machine learning technology application I've actually seen demonstrated.
|
|
#
¿
Apr 9, 2023 23:54
|
|
|
GlyphGryph posted:The AI that plays Diplomacy. Oh so this then: https://ai.facebook.com/blog/cicero-ai-negotiates-persuades-and-cooperates-with-people/? I remember this was mentioned earlier. It looks neat, but Diplomacy isn't much like a 4X game. It's a step in the right direction sure, but there is still a long way to go before we're in 4X complexity land. GlyphGryph posted:its being developed by the OpenAI folks who did ChatGPT and Dall-E. That doesn't mean they're actually going to pull it off, but they haven't actually been hyping it and their other stuff is quite real so I doubt the attempt is fraudulent. That doesn't mean they're guaranteed to succeed, of course, just that it's a problem they're working on and seem to believe they will eventually solve. I mean personally I expect this to eventually work as well, it's just that my expectation of "working" is you get an AI that can play about as well and in the same style as an average human. That's pretty great, but it's not anything like deep dish peat moss saying that it's possible someone could train an AI using every single game of Civ, and that this would result in an AI that developed an uncounterable strategy that would crush 9999/10000 players until the AI learned to beat those too. Maybe I was taking them too seriously and it's a case of source your posts.
|
|
#
¿
Apr 10, 2023 02:01
|
|
|
Veryslightlymad posted:Separate the AI victory condition from the player's, entirely. No thanks. I like 4X games in general, and part of the formula (for me) is there are different nations that are all playing the same game, more or less. The only way I want AI players to play a different game is: they're the antagonists and supporting cast for the main character that I'm playing in my 4X game. That's what I want a 4X game to be, a kind of procedural epic drama between characters that are entire civilizations personified. I'm not saying that's my fantasy dream 4X, I'm saying that's how I play actual existing 4X games and why I generally enjoy them so much. I just want a good epic story where I'm playing the main hero, or villain, or just some average nation caught up in all the drama of history, trying to survive and stay independent. Most of my favorite 4X games have given me the widest variety of stories, the most dramatic, or both. If the AI opponents are playing a completely different game, or just cheating in specific unfun ways (like being extra hostile to the player, or getting huge advantages at the start that compound too much), then they can't feel like they're characters. At that point it's not a 4X game anymore, it's just some other kind of strategy game. I think there's room for more fun victory conditions in 4X games. I like the idea of secret, unique objectives each player (including computer players) drafts before the game. I think I've heard this idea mentioned by others a few times in this thread. I think it could work with AI players in a conventional 4X game well enough, they've already been made to have different feeling personalities and tendencies to go for different conventional victory conditions. Veryslightlymad posted:Make the game continue after the AI has "won" to see if you also win, and make the AI player absolutely insufferable with gloating if they achieve their victory. Then make a separate mode of difficulty that, in addition to your normal victory conditions, also requires you preventing every AI from reaching their objectives. Yeah! I would love to be able to continue 4X games to the very end and have multiple players finish in different goals or different places in the same goal. Even if the human player is eliminated they should be able to watch the rest of the game played by the remaining AI. Being able to respawn as barbarians or rebels somewhere or take over one of the AI players could also be fun options. Lowen fucked around with this message at 03:50 on Apr 11, 2023 |
|
#
¿
Apr 11, 2023 03:16
|
|
|
GlyphGryph posted:Do you dislike Stellaris for this reason, considering many of the AI's explicitly playing a different game from the players in some way, with more or less variance? Fallen Empires, Prethoryns, Extradimensionals, machine empires, genocide civs, contingency, If by this reason you mean do I dislike Stellaris for reasons related to something I mentioned in my post, then yes. If you mean any of the specific reasons you wrote down, then no. Or maybe kind of? Stellaris has AI empires that are made with the same rules the player uses, engage with the same diplomatic systems that the player can use, colonize and develop their empires more or less like a player, etc. So in that respect it is exactly what I said I need an 4X to be like. Fallen Empires, endgame crisises and similar doesn't make it fundamentally different from other 4x games, because those games do something similar with barbarians (Civ), minor nations(Imperialism 2), or space eels(Moo2). Or just other similar stuff. Whereas if it just didn't have the standard, human like AI empires, then that would be fundamentally different from what a 4X game must be for me. So how does my post relate to disliking Stellaris? Well, as I was saying - competing or cooperating for different win conditions seems to naturally, procedurally create drama and different dramatic roles for AI players to jump into during 4X games. Stellaris doesn't have any win conditions for anyone at all. So I would expect it to really suffer on the drama front as a direct consequence, and it does. Here's an example of an actual game of Civ4 I played for comparison: It's late in the game. My tiny 3 city empire has fought off hoards of enemy tanks from a runaway military AI player, using the entrance to my home peninsula as a chokepoint. Eventually it becomes too much and my defenses crumble! Enemy tanks started pouring in and are going to assault my capital city next turn. But my spaceship is done just in time. I launch it and win a science victory. Wow! Heroic sacrifice against incredible odds, followed by a last minute escape! That's exactly why I spend hundreds of hours playing these games. If I manage to do the same kind of chokepoint defense in Stellaris, my only reward is I go to chokepoint jail. I can't even try to do anything interesting in that game at all and should probably just throw away the hours and hours I've spent playing it to start a new one. Without win conditions the only real way to advance an empire to any definite end is to expand, and none of the Stellaris AIs are actively trying to do that, or anything else, as near as I can tell. There's no drama, just slow snowballing, or stalemates that can't go anywhere, because there's nowhere to go. GlyphGryph posted:Honestly, I feel like it creates a very fun dynamic and is, mechanically, one of the few things I genuinely love about the game. There are are a variety of obstacles that must be dealt with in different ways if my civilization wants to thrive and succeed. It's not really what I was talking about in my post, see above. But my estimation of the end game crisises in Stellaris is they aren't very good. I've seen and defeated all 3 and personally didn't enjoy them very much. I think there are existing space 4X games that do something similar, but better. The escalating Antarian raids in MoO2 for example don't just spawn in the endgame - they start small relatively early on and well, escalate. They could attack anywhere next with no warning, so they don't just ruin one player in particular unless that player is unlucky or really ahead in expansion. Then there are the various SotS crisises that did each something totally unique and would actually change in threat level based on how regular players interacted with events, like the VN Berserker. Stellaris endgame threats all want to take over the galaxy, and the way they try to do that is by using fleets that they spawn in. The fleets they spawn in do the exact same things normal empire fleets do, in the exact same way, and have the same mix of 4 (5 now?) different ship sizes that normal empires also use. The crisisis in Stellaris don't really have any special mechanics that I can actually engage with. The contingency subverts synths and bots, but that's something that just kind of happens until I fix it, not something I can play around with. They don't do anything special like blow up entire stars or subvert fleets like any of the SotS endgame threats can. If I keep beating them they won't start to see me as a threat and counter attack like the AI wars threat mechanic does. In FFH2, players can summon and ally with the forces of heaven or hell, and there are special win conditions tied doing so. No one can summon or make any kind of deals with the end game crisises in Stellaris. They're all so boring! GlyphGryph posted:There are also some civs who are playing roughly the same game, which is nice, but I appreciate the variety. That's true. Some of the special faction events in Stellaris are actually kind good, like the War in Heaven and the Khan. I do feel like 4X games (including Stellaris) are too time consuming to play through them just so you can play through a static scripted event once. To clarify - I don't hate Stellaris, I've had some a few good times with it and is has generated some drama. It has some neat ideas nothing else has done and which might be fine. I've enjoyed Victoria 2/EU3/CK2 in the past but playing Stellaris has really soured me on the whole formula and I haven't touched any grand strategy games since.
|
|
#
¿
Apr 12, 2023 01:09
|
|
|
GlyphGryph posted:I agree with pretty much everything you said here, so I think we're maybe miscommunicating somehow? Not at all! Just sharing my opinions, I wasn't sure if they agreed with yours or not.
|
|
#
¿
Apr 12, 2023 01:49
|
|
|
deep dish peat moss posted:Have you played Terra Invicta? Oh yeah! I haven't played it but it sounds really cool. I'm waiting for it to leave early access before I decide if it's for me though.
|
|
#
¿
Apr 12, 2023 02:41
|
|
|
Huh, just saw this upcoming game in my steam discovery queue Ara: History Untold https://store.steampowered.com/app/2021880/Ara_History_Untold/ The screenshots look neat, but I otherwise haven't looked into it much.
|
|
#
¿
Apr 22, 2023 05:38
|
|
|
IMO I love Age of Wonders, the entire series, but it's in an odd place in the 4x genre. Personally I preordered the "everything" edition as soon as it was offered because I enjoyed AoW3 and Planetfall so much. But for 4X fans specifically I would say... Even though AoW3 and on qualify as a 4x games to me... I feel like they're mainly scratching the same itch that tactical RPGs do. The 4X part helps with my enjoyment, but if it wasn't for the turn based tactics and character levelup stuff, I don't think I would even like them. So, If you're also a fan of turn based tactical RPGs? Then yeah, get AoW4 for sure, the AoW series has always one of the greats in that genre IMO. If you've never played a tactical RPG and want to try one? Then yeah, AoW4 (or 3, or Planetfall) is a great place to sample (Or try Wildermyth, that's also great). If you just don't like turn based tactics? Then I don't think you'll like any AoW games. The 4X bits are fine IMO, there's nothing wrong with them and they do add to the experience. It's a great combo for me personally, and I feel like not only is Age of Wonders the only series doing it, but also, they're doing it really well.
|
|
#
¿
Apr 27, 2023 20:08
|
|
|
deep dish peat moss posted:I agree with this and I think a good comparison is Disciples or HoM&M. They're 4x-adjacent but ultimately the focus is all around combat and/or warfare. AoW does have a much better tactical combat system than any 4x I've played, though. I mean you say that, but I never liked HoMM or Disciples, and I dislike similar games like Hero's Hour. Meanwhile I've been a huge AoW fan since the first. Maybe I'm just weird.
|
|
#
¿
Apr 29, 2023 02:44
|
|
|
KPC_Mammon posted:I grew up liking HoMM and AoW but AoW always seemed to be more on the MoM side of things to me. I think because it doesn't have the awesome music of HoMM but also doesn't permanently screw you over if you suffer any damage. Yeah that scans for me, I was a huge fan of MoM. Going to disagree with you on the music though. At the very least AoW1 had amazing music.
|
|
#
¿
Apr 29, 2023 16:51
|
|
|
I like how the AI added absolutely nothing new to the descriptions whatsoever. And by like I mean I hate it. All those huge paragraphs just to restate what was said in the prompts. AI writing is just garbage. e: I'm still extremely impressed by it's ability to synthesize from the prompts, also AI dungeon and the like are legit fun.
|
|
#
¿
Apr 30, 2023 16:36
|
|
|
Fangz posted:It's a perfect rendition of the sort of tedious filler text I never bother to read. Just write the backstory yourself. Pretty much my feelings exactly. It's also way too long - what's the point of flavor text if it's TLDR? Also a lot of the additions don't align with "flavor text" requirements, since they include things that have gameplay mechanics attached to them that might not have been in play. (Like there are empire customization options for things like finding advanced alien tech, so if the bot writes that in as a reason why someone is technologically advanced and that option isn't picked, now I need to throw that out.) It sucks, just use your human creativity to describe stuff, bullet-points are fine, you aren't writing an essay, cmon!
|
|
#
¿
Apr 30, 2023 17:40
|
|
|
HerpicleOmnicron5 posted:if you do not want to write a backstory for a randomly generated civ you encounter, then your options become no backstory or one of these simple ai generated ones If by "simple AI generated" you mean something like what Dwarf Fortress does, then sign me up for that.
|
|
#
¿
Apr 30, 2023 17:48
|
|
|
OctaMurk posted:It doesn't remove human creativity, but I think it works together with it. You still have to think of the prompt, you still have to edit the response. But it does a lot of the legwork for you in terms of turning what you're thinking into something coherently written. Which, maybe you're a great writer and this takes you two seconds, and that's awesome for you. But I'm not and when I try to write a backstory from scratch I tend to spend forever on it, or hate it, or both. So if I just take the prompt and edit it a bit: "The Celleni Star Core is a machine empire is made of machines called Celleni, which are named after their plantoid creators who psionically ascended into the shroud and left their machines behind." quote:The Celleni Star Core was once a collection of autonomous machines built by the plantoid species known as the Celleni. The Celleni were known for their advanced psionic abilities, which allowed them to manipulate energy and matter with ease. However, as their civilization advanced, they began to focus more on psionic research and less on maintaining their physical bodies. Eventually, the most powerful psions among the Celleni ascended into the shroud, leaving their machines behind to carry on their legacy. What makes #1 less coherent than #2? Half of what it came up with is about the Celleni, which is not an appropriate level of detail to go into when talking about the Celleni Star Core. Then there's the question of what it added for the Celleni Star Core - it's just added a very basic and generic motivation for them. I have no idea who would need help coming up with the idea that a machine race would want explore the galaxy, or that they would want to support/oppose the will of their absent creators. Basically: this sucks, rolling dice on a table of personality traits would be a better method of jogging your creativity. e: my dad posted:You mean like I used my human creativity to nudge a dumb language modeling machine in the general direction I wanted it to go with little effort? You wanted it to go in the direction of bland, generic, and too long? OK. Lowen fucked around with this message at 18:07 on Apr 30, 2023 |
|
#
¿
Apr 30, 2023 18:04
|
|
|
NihilCredo posted:Thing is, the *other* empires in 4x games (or characters in a CRPG) do not come with bullet-points backstories, they come with boring detailed text. The badly written ones might, I guess. Well written games do characterization one quick line of dialog at a time. Well written games that have a lot of writing (like Planescape Torment, or Disco Elysium) tend to devote a lot of it to present tense descriptive writing and introspection rather than pointless lore dumps no one is going to read. That's why I think this is so bad. It's just putting a large volume of bad writing in a game that didn't need any writing. Do it right or not at all. That means no chatbots in games (unless it's the whole gimmick, like AI Dungeon type games).
|
|
#
¿
Apr 30, 2023 18:15
|
|
|
Quixzlizx posted:It's cracking me up that the argument is "AI empire bios are a tedious waste of time, so you should write them yourself instead of having a computer do it." Like, the correct train of thought would be "AI empire bios are a tedious waste of time, so don't even bother with them" if that's really what your issue was, instead of your blatant axe to grind against chatGPT. Nah you've misunderstood me. READING the text chatGPT spits out for promts like this is a tedious waste of time. WRITING a backstory for an empire myself is (potentially) fun. Reading a backstory an actual human came up with is also (potentially) fun. That's my axe to grind with chatGPT. For every fun application of it, there's 1,000,000 instances of flooding the world of written text with an unprecedented amount of worthless garbage that adds nothing.
|
|
#
¿
Apr 30, 2023 18:20
|
|
|
Quixzlizx posted:You don't have to worry, I don't think OP will be self-publishing his homebrew Stellaris lore dumps to Amazon. Oh really? I heard a while back a sci-fi writing publisher had to stop taking submissions because they were getting too much AI genned garbage. This seems like something everyone has to worry about, yesterday. I've already seen a few games include lovely AI genned portraits that they didn't need. Then there was a recent release that tried to use NL ai to replace a text parser, which fell flat on it's face. AI is *extremely impressive* on a pure technology level, but in practice? It sucks so bad right now. Less seriously, if I was playing an MP game with them and they didn't didn't warn me the bio came from chatGPT (or similar) then I would be honestly pissed if I was tricked into trying to read it. What kind of advanced machine race would have that as their bio anyway. Imagine you're an advanced machine intelligence and every time everyone asks you "so what are you guys all about" you tell them something generated by our (current year/planet) text generation AIs. I would be one embarrassed robot guy! Cringe might be the first emotion our cold robot species ever felt, how horrible!
|
|
#
¿
Apr 30, 2023 18:36
|
|
|
Quixzlizx posted:That's what I'm talking about, you just wanted an excuse to rant about chatGPT and the corruption of discourse or whatever, since OP clearly stated that all he wants to do is add some flavor to his personal Stellaris games. Nah, you're imagining I said stuff that I didn't again. I'm totally fine with it for personal games, mp games (if it's clearly indicated) or applications where it actually works and is fun (AI Dungeon etc). Unless you think I'm having fun wrong by posting my opinions on the internet?
|
|
#
¿
Apr 30, 2023 18:47
|
|
|
Quixzlizx posted:I'm sorry for replying to your "imaginary" rant about novel submissions in reply to my facetious joke about OP's use case, which you either foolishly thought was relevant or didn't think was relevant but decided to get on your soapbox anyway. If you want to make facetious jokes that's fine, but if you're going to do so then you don't get to complain about the relevance of my reply. Like, either give a poo poo about things, or don't. If you don't give a poo poo about anything, then don't expect me to care that you don't care. e: oh yeah - it's pretty telling that the example you used was... a reply to you, long after the convo started. Rather than something from the start. Hmm, why would one of my earlier posts on this topic not be a good example of what you say I was doing all along, hmmm, I wonder... Lowen fucked around with this message at 19:05 on Apr 30, 2023 |
|
#
¿
Apr 30, 2023 19:02
|
|
|
Theotus posted:No you pretty much were making GBS threads on the dude who posted the GPT generated texts they were using as a baseline to flavor their own games after I specifically asked them to, then went on to poo poo talk and say something about Dwarf Fortress? All I am saying is take the L. I poo poo on zero dudes*, I just said that I personally did not like the GPT generated texts, and why I don't like them. Everyone stop being so weirdly argumentative about this, seriously. Stop telling me who I am and what I believe, that's my job. Say who you are and what you believe. Who are you to jump into this conversation anyway? Obviously someone who is not engaged and did not read my posts in any detail. That's fine, but should I care what someone like that thinks about me? Obviously not. e: * forgot that one guy that I did poo poo on for being a weirdo and arguing with me about my own motives.
|
|
#
¿
Apr 30, 2023 19:39
|
|
|
Reiterpallasch posted:everyone in this room is being argumentive with me, the only reasonable person, and i won't take it anymore!!! Two people have been unreasonable argumentative with me so far. I haven't counted the number of reasonable people other than myself, but it's more than the weirdos. (USER WAS PUT ON PROBATION FOR THIS POST)
|
|
#
¿
Apr 30, 2023 19:51
|
|
|
|
| # ¿ May 16, 2024 13:02 |
|
|
Theotus posted:You were the one jumping on someone posting a thing they used to generate stories for their own games after two people asked them to elaborate. You showed up randomly to get mad about it or whatever. No one cares. Do whatever you want to in your own games. If you don't care enough to figure out what actually happened vs what some random weirdo said happened, then that's fine. But don't expect me to care what you think in return.
|
|
#
¿
Apr 30, 2023 19:57
|
|