37
votes
What's your game that you'd really like to see made?
For me, that'd be a multiplayer war sandbox. Take a big map, plop down resources, population centers and factories, spawn a lot of players who then have to organize to fight a war.
- To add espionage, you'd not hard-code a lot about the way teams organize(think removing the guild mechanics from WoW), so one could sign up with one side, then run to a different side (who do not know you already signed up with the first) and do ye olde double agent.
- Research could be modelled as being done by NPC civilians, which produces documents detailing the results. These documents must be on-site whenever the research is being used (i.e. present at factories), so it could be stolen.
- No need to go factorio on the production chains. Their purpose is mostly to enforce cooperation to reach better hardware for the guys in the field.
So, what's your crazy dream game that's never going to be built?
Never played For Honor. I presume it's got relatively detailed melee combat mechanics? (Next thing I played I could offer as comparison, I think, would be KC:D - which would be rad with lightsabers.)
Why not just cut out the middle man, and give the dragons lightsabers?
I have some graphics I made I'd like to show you..
honestly if they just ported movie battles II to consoles with enhanced graphics that would sell like hotcakes.
I mean, theoretically that's not too hard to achieve. 30 binary choices that all impact the end is a billion different endings, potentially. What's hard is making those endings feel unique. Dwarf fortress does this to an extent, but what it lacks is a kind of... cohesiveness. I'm imagining a AI like rimworld's story teller creating a fallout new vegas style outro after you've won dwarf fortress and had fun. Probably still a few years away though.
I think a game like this might just end up coming about in the near future.
This is me too. I'd love to see an RPG that gives everyone a unique experience and ending. It'd have endless replayability.
Something like that would need lots of random encounters that altered the story. Think the random encounters in fallout 3, but each one impacts your story and alters how the world shapes up. Eventually, everything ends up leading to your own unique story and unique ending.
I'd play a game like that forever.
So... D&D in video game form?
tildes' first novelty account? :D
Nope, dude just really likes DnD. And why not? It's basically the best game ever!
Honestly, yes.
Have you played rimworld? It's not an RPG, but the way the AI storyteller works might be influential in future RPGs too. The game basically tries to have a AI storyteller akin to a dungeon master, dishing it out hard when you're doing too well, easing off when you're at the brink of losing, trying to control the tempo of the game basically. Giving you phases of relative calm or ... not. It's a really cool concept (though it could use some more love imo) and I think it could work well to control random encounters in RPGs too.
That actually sounds pretty cool. I'll have to check it out sometime. Thanks!
I appreciate that some people want this, but just for the sake of counter point - I don't think I'd find something like that interesting in the slightest! I find that for me, the best game experiences are the ones that are more tightly controlled. With a single ending (or handful), the dev can ensure it's a really good ending (they may or may not succeed). With billions of endings? There's no way to hand-curate that.
Replayability is the goal here. And individual playstyles. If you could pull off a 80% quality ending like this, a player could get several of those out of the game. Very curated experiences? Yeah, first time is gonna feel like 💯, but the next playthrough will maybe get half that. It generally depends on how much time you want to spend with the game.
Prediction: you either don't play much or don't mind paying extra for a short and sweet experience. Right?
Story games like Life is Strange or The Walking Dead but that push the genre even further, with more mature themes and stories (think Dostoevsky adaptation).
I've thought about in depth dialogue systems before. What I'd like to try out ist imulating npcs knowledge as a graph of entities with relations. And morrowind-style you can ask about anything that ist in your knowledge graph. Your partner will then figure out if this is a change of topic (large distance) or just a slight change. Given a slight change they will generate an answer that traverses their knowledge to what you wanted to know.
Oh man, could you imagine a narrative game based on Demons?
Man, I absolutely agree. When I play a game with a really good story, I sometimes wish it were like that, even if the gameplay is good.
Minecraft done properly, which to me means taking the game as it existed during alpha and early beta and building on that, rather than creating a weird RPG / combat / MMO platform / voxel modeling thing that fails to live up to any single area.
how would you have liked to see the game evolve?
Basically just added more content but kept the core gameplay loop of dig blocks -> put blocks down. Basically just a big sandbox where you had to move the sand manually.
I feel like you can still do all the that at its core. To me, achievements, the end, etc, are optional tasks there for people who would like some more direction.
You can, but the fluff gets in the way of the experience. I've tried in the past to mod away all the things that annoy in the new versions, but it's so much work that I decided it was easier to just play the versions I enjoyed.
Oh bless you. Funny story, I actually got into programming because I wanted to make a minecraft mod and the Forge API is sooo bad.
I'm still not clear which bits you think are fluff and which bits are extra content.
Can you give examples of the annoying bits of fluff, and examples of good extra content?
And why playing on creative or on survival peaceful isn't good enough?
Right now my grail is an in-depth Sci-Fi RPG. Something spacey and futuristic, with a deep and developed word. Something 3D, first or third-person. Something recent, with a high polygon count and high-resolution textures. Something with a robust character creation. Something really immersive, where I could become my character and get lost in the world. Mass Effect 2 is as close as I've found so far, but that's too on-rails and the graphics are beginning to look dated (although I understand there are HD mods). The Deus Ex games are pretty good as well, but could be better.
I hope Cyberpunk 2077 can scratch the itch, but that's not so spacey and is a few years away presumably.
Bethesda announced Starfield at E3. There wasn't much info other than 1st person sci-fi adventure set in an entirely new world. But, seeing as how Bethesda is responsible for Fallout and Elder Scrolls, it might be worth keeping an eye on.
I was thinking of Star Citizen until you mentioned Mass Effect 2. But then again I'm probably projecting too much RPG into Star Citizen. And it's got a... questionable... business model.
But yeah, I could get behind that game. You looking for MP or SP, guided tour or tell-your-own-adventure?
Have you heard of Bethesda's upcoming Starfield? Bethesda is starting to lose my blind faith, but of course I'll keep watching them.
Star Citizen looks interesting, from skimming the gameplay trailer. But it looks to be pretty focused on the starship aspect of things. I might have to try that out in the future.
Preferably single player. I've played a few MMO's, and they all tend to share the problem where everyone is the chosen one. While the world looks busy and bustling, it's mostly just people on the same quests or storyline that you are. Which, feels pretty lonely. Solid single player games with well written NPC's can feel more alive IMO. Something open with multiple storylines would be great. One of my big problems with the Mass Effect franchise was the ineffectuallality of all my decisions, when told that they mattered. That combined with the false open-world, and the mostly on-rails gamplay was just not the combo I need. Maybe that's asking too much.
I haven't til now. I'll have to keep an eye on it. The last game I gave Bethesda money for was Skyrim, and while I enjoyed it enough to spend 400 hours on it, it was still pretty lacking. Especially in the graphics and character creation aspects. I watch some Fallout 4 over my partners shoulder, but that didn't grab me enough to play it myself.
Hell, even an isometric rpg (Ala Pillars of Eternity 2) set in a Mark Miller Traveller Universe would be fantastic. Really digging the Pillars and could see Obsidian using the same style of system for something like that.
I'd enjoy something similar to the Elder Scrolls series but more... reactive? Open world games always had an annoying limitation to them, where you can only interact with the world at large in incredibly specific ways.
With the Civil War subplot in Skyrim, you're only able to select between supporting the Empire or the Stormcloaks. You can't make a grab at the throne for yourself, install someone else as the High King/Queen, descend the land into anarchy, split the country into multiple fiefdoms, etc.
God knows how difficult it'd be to actually make the game though.
You sure won't be able to control what's going on as a developer. This would likely require emergent reactions based on complex systems interacting. But as hard of a time dwarf fortress has communicating it's depth, these kind of things can be simulated there, so in a completely procedural world with complex mechanics that the AI participates in, this could work. I'd love to see it too.
Have you seen Westworld? It's a TV show on HBO
Check out Pillars of Eternity 2. It's an Obsidian game. Currently about only 20 hours in and it's been a blast.
Warcraft 4, as an MMO RTS.
Base building: in the WC4 MMO, it's not base building as much as it's city building. I want many tiers of buildings with many variants down many branches which give me access to many different tech options and many different units and unit variants. I want oodles of defensive options and I want tower defense-esque gameplay when another player attacks my city. In this game, my "base" is a labor of love maintained over not minutes but months or longer. In MMORPG terms, your city is your character, the thing you agonize over planning, building, and progressing.
Combat: I want the actual combat to be an evolution of WC3's combat: small armies centered around a hero who also has a persistent progression. I build armies in my city, and maybe my city can "house" more than one 90-food army, but when I take an army and go raiding, I take a single 90-food army. When I enter matchmaking with an army that my city has produced, I can either attack an AI town for a small reward, or another players city for a big reward, where I pit my army against the other player's city defenses.
So pretty much Clash of Clans as a full-on desktop PC game, but without balance-altering microtransactions and much richer city building & combat.
This is why I like my RTSs slower, to the point of being 4x games. I'm too much of a perfectionist to just build my base and then walk away from that game. Similarly, I find it agonizing when besieging a castle in stronghold when I can observe enough of the battle to know what I want my units to do, but I can't control them fast enough. Even worse when you have a relatively detailed plan in mind and just can't execute it. Drives me insane and leads me to use simpler, more brute force attacks. But what does one need for raw force? A powerful economy and a solid defense. Which I can't just walk away from, so I'll end up just... amassing stupid amounts of troops and then curbstomp the enemy even more than my brute force plan would require. </rant>
I like the way you think though. I'm imagining building cities the likes of which you'd expect in WoW, in terms of character. :)
Precisely. In fact what I've suggested is meant to remedy both of those tendencies. Elaborate and thoughtful "base" building is encouraged, and curb stomping your enemy isn't an option: you get a 90-food army and your enemy has defense "budget" allocated such that your 90-food army vs. their city defenses is a balanced matchup. You're not going to raze their entire city, but you may get to one of their warehouses, banks, or armories and pilfer something valuable or interesting.
Really knocking someone's city out would require the MMO RTS equivalent of a "raid." And planning it would require communication which could be intercepted by the enemy faction, thereby giving them a chance to coordinate a defense.
In any case it will never happen: we're talking about a fusion of game genres which are both well past their peak. But that is what you asked!
Board games nowadays keep getting larger and more complex. Modular dungeon tiles, customization miniatures, multiple sets of different kinds of dice, and more and more complex rules. But what if there was a board game that had all of that plus more?! Well now there is! Introducing the ultimate in table top experience, Set Up: The Game. Comprising of over 1000 different tiles, 200 different types of currency, 50 different types of dice (including some non-transitive dice), and a 300 page rule book comprising a whole semester at your local university this board game is sure to fill the monotonous void in anyone's boring life. Watch as you and your friends campaign for votes to see who places the first tile. Marvel at the complexities of the in game stock market with over 70 different commodities that must be randomly generated over the course of a week in order to distribute wealth based on social status and race. Try and decipher the unique complex rules with properties and interactions that still need to be understood with set theoretic logic. And when you finally finish placing all the tiles, dice, figurines, money, tokens, and character creation sheets you get to start over and do it all again! Set Up: The Game! The most time consuming and difficult to understand board game you'll never actually play.
Edit: As much of a joke as this game would probably be I would actually pay to have something like it created.
I feel like there's actually a brilliant idea in there somewhere.
I'm imagining each person gets different rule books, parts of which could be in english, others in an uncommon language or even in glyphs, but each book has different criteria to set up the game, such that one person might have to build a tower using certain pieces, while another person needs to use those same pieces to put something else on the game together. Maybe the rule books reference each other or have missing words that another rule book lays out in a different page. I'm imagining SpaceTeam (https://www.playspaceteam.com/).
Just got to work out what the end goal would be. Something that doesn't make the same game over and over again, but would give variation in the final output game?
I probably am doing a bad job of getting my point across. The joke is that you have to spend hours setting up the game and that's the game. It stems from a friend of mine who likes to buy board games with huge numbers of pieces and tiles that seem to take longer and longer to set up each time we hang out.
No, no- I think you did a great job of explaining the game. I just kind of took your idea and ran with it some because I think it'd be a pretty neat game to actually play
I think having a board game that is generic enough to be any board game is hilarious. I would pay to see this made.
Edit: Play to pay, oops.
Once VR matures, I would like to see a RuneScape-esque medieval fantasy MMO using it which really understands the value of skilling and an economy which relies on personal connections and skill as a trader (in RS terms, pre-grand exchange).
I want to roleplay a trader and play out my Spice & Wolf fantasy god damn it.
It sounds like you want real life, but with VR-chat character creation. Have you thought about cosplay economics?
one day.
My idea is similar to yours, but rather than an identical game for every player, you have a flight sim for people playing the aerial units, an FPS for the foot soldiers, an RTS for the commanders, etc. with potentially dozens of interlocked games, and all of those sub-games interact with one another. A flight sim player can strafe a foot soldier, an commander can give orders to the others (which show up as objectives in their games), a foot soldier sees the AI units the RTS player is controlling and so on
I fear that if you hard code too many of those interactions, it'll feel too "apart". I'd rather have everyone play a fps game first and foremost, and just add some mechanics to support different playstyles. Aircraft are vehicles that handle like in a flight simulator, commanders can acquire a map table, and an AI NPC updates unit positions on that table. The commander just radios commands. Since you're specializing anyways, most of the complexity of any other playstyle won't overburden you.
What about two separate games released to be intertwined where one is a tactical command game and the other is a FPS which the troop orders and movements are directed by the players playing the tactical game?
Probably wouldn't work because the number of players would be messed up so the wait times for games would suck
Also, commander assassinations, infiltration, espionage, all get so much more... natural if the enemy's commander is just another soldier, just at a desk.
The Planetside games are a bit like that, they need to make a third one
MMO RPG VR game set loosely in the Harry Potter Universe, but without any reference to the actual Harry Potter characters.
The player could brandish a wand and hold down a button and then make a gesture to cast spells. Internally there would be a complex mathematical backend that took as input the pose trajectory of the controller and gave as output a vector in an abstract space that determined the properties of the spell that was cast. There would be no in-game guide to learn how to do magic, and indeed even the game designers would be unable to fully predict the complexity of the 'magic-function', so players across the world would have to collectively backwards engineer how to do magic, or learn it from their friends.
There would be an inventory, loot and crafting system similar to many RPGs, but here "crafting" would be potions. There would be a complex underlying mathematical system for mapping ingredients to effects, again creating a large discoverable and undocumented space for people to explore. The spells you can perform with your wand could also be used to cast charms on your items, just as the items you collect could influence or augment your baseline magical powers. Items in the game would be explicitly rivalrous - no multi-drops for the whole party or anything like that. There would be an elder wand in the game, with mechanics for motivating the holder of the elder wand to expose themselves to risk.
Once you learned how to cast deadly spells, you could obviously kill other players in all areas besides, say, Diagon Alley. There would be natural stimuli within the game to push people towards either becoming aspiring dark wizards, whose goal is increasing their power and self aggrandizement, or aurors, who join up in teams to try and win the bounty on particularly nasty dark wizards. In general, dark wizards could make dark bargains that increase their power but make them permanently unwelcome in common spaces, forcing the division. The goal would be to make people be the bosses - "raids" would be attempts to kill and steal the loot from another person playing the game. Think EVE. One explicitly included gameplay element would be the creation of horcruxes, which would be difficult but possible to pull off. You would have to leave your horcrux in the gamespace out for anybody to find it, making convenient "sub-quests" for aurors who are trying to take down a super powerful dark wizard.
Direct movement would be restricted to a local area, but you would be able to quickly apparate around the environment. This would be one of the few spells handed over to the player early on and made easy to cast. You would also be able to acquire a broomstick, which would be controlled using a handheld controller held in the natural broomstick-y way. Floo networks would be used to take you around the main pre-generated parts of the game, whereas the wilderness full of beasts and dark magic would be procedurally generated, and generally have to be explored manually. Occasionally, random items that one found would secretly be portkeys that took you to long-forgotten caves, tombs, etc.
Yeees about the magic system. I don't know how one could solve that, but that sounds so cool. I've got a few ideas floating around my head, because this task screams "neural networks" at me. You'd basically try to match a set of known gesture-effect combinations. The way these combine to form new effects would depend strongly on your mathematical representation of the gesture and of the effects. Same goes for the potions, but here representations of the input is much more straightforward. I'm thinking chemistry lab, so the reaction is governed by present reagents and conditions, and the output governs effects.
This is really cool, I like it. Thanks for sharing! :)
Hey I meant to respond to this a couple weeks ago and never hit the post button, but just wanted to let you know I appreciated the feedback! The idea of the chemistry lab is exactly the sort of idea I was going for. I can give you one example of how I thought a "mathematical" potion system might work:
Think of each magical ingredient (hair of unicorn, eye of newt, nightshade! etc.) as a vector in a very high dimensional vector space. If you take integral combinations of each of the ingredients, they together form what is in mathematics referred to as a lattice, which is a subset of the vector space. So what you could do is choose points in the vector space that represent hotspots, and assign to each one a magical effect that the potion might have. When you put together your ingredients into your potion pot, underneath the hood it's adding a bunch of vectors together into one, effectively selecting one of the points in the lattice. You could then measure the distance between that point and all of the different magical effect hotspots, and the closer you were to a hotspot, the more of an effect that potion would have to the associated magical quantity. For enough dimensions it would be hard even for the game designers to know which combinations of ingredients produced the strongest results, because finding the closest lattice point to another arbitrary point is in general an NP-hard problem. So you would have a genuing discovery process where the community was learning things the designers could never have anticipated.
Halo 3 Anniversary. Is that crazy enough?
I'm so completely unfamiliar with Halo, I can't even tell what's crazy about that. :D
Halo CE - released in 2001, got an anniversary in 2011.
Halo 2 - came out in 2004, got a remaster for 2014.
Halo 3 - released 2007, it’s 2018 now. :(
AR lasertag. I've messed around a bit with vuforia before, doing simple AR stuff, but I don't have the coding chops to hash out a complex environment. At this point, I just want someone to make it happen.
That's rad! I love stuff like that. I saw some sort of open source AR project recently that had gesture recognition that didn't require gloves. I looked through their GitHub a bit and found their tutorial for making a set of goggles. I've been seriously considering making a set, partially to also make a how to video, but also just to play around with it a bit. I kinda wish I knew more coding, but it's such a deep rabbit hole that I don't feel like I'd be devote enough time to get up the learning curve.
It's still tempting though because the idea of being able to combine game elements into open spaces is so compelling to me.
A Breath of the Wild style MMO.
A mage-centric 3D RPG with real-time gameplay and a "combo" discovery system similar to Legend of Legaia. The idea being that certain (intelligently) randomly generated input combinations allow casting of different spells, you're not told what those combinations are until you discover them, and the maximum input sequence length increases as you progress further in the game. Individual spells that have been unlocked could probably be improved using skill points or the like as well.
I've always found mage classes in games severely underwhelming because they either a) become useless almost immediately as MP quickly dries up, b) can't effectively operate without a party, or c) just plain suck compared to literally any other class available. They're always designed as supporting characters and I find that disappointing.
Or as a mage you get OP at some point. Both Skyrim and Gothic 3 had the problem that they relied on Mana for balancing, but then allowed you to either reduce cost to zero or get crazy regeneration at some point. I completely stopped my mage playthrough of Gothic then and there. Too easy if you're just a WMD on legs.
I wonder what would happen if you were to have a energy system underlying these mechanics. Basically, conservation of energy for all your magic energy, and as a mage you need to acquire such energy somehow. Your skills (both player and character) determine how efficiently you use that. So that would mean you have to manage that energy, and the better you get at manipulating the world into doing your work for you, the better. Think the eragon books, where a primitive but effective way to magically hunt deer is to use a pebble as telekinetic bullet, not some firebolt nonsense. A better way in Eragon is to just sever the spine telecinetically.
Different classes/types of magic users would acquire their Mana differently. Druids would leech off of nature, warlocks would exploit living or demonic creatures, maybe some others use astral energies.
Of course I prefer not to have hard class boundaries, so you could mix and match.
Holy shit, I completely independently had a very similar desire. You should check my post lower for a take on this. My idea is that the spells themselves are vectors in an 'effects-space', and then you would use a complex mathematical mapping from input trajectories to spell-vectors to make the 'rules of magic'.
I like OP's idea, but I'd want a game that emphasize on terrain tactics. Something like CIv5 combined with Tower of Defence.
To conquer a stronghold/city, you have to choose your route carefully through forests, rivers and mountains. You can to establish a support line. Epic battles can be carried out on strategic choke points.
Yeah, considering that you'd probably have to handle logistics as well, my concept would probably imply raids on supply lines as well. And since you probably want to send AI for those boring tasks, insurgents have a viable option of acquiring materiel. So that'd be interesting to see play out. (Add in the fact that no one wants to play as yet another goon for the world-spanning empire, because insurgency is more fun, and you probably won't even have to worry too much about resetting the world after the war is won, because that empire will fall. OhMyGodIWannaPlayThisNow)
For terrain, I'm currently looking into how detailed free GIS sources or OpenStreetMap are. If you could recreate cities and rural areas with reasonable accuracy... yeah. That's terrain tactics for you. :D
Yes, that's probably what I want.
Also about AIs, I think we can some kind of mechanism to delegate portion of your units/legion to AI, aka, a prefecture. You can focus your operations on a small group, one major city for micromanagement fun.
However, if you isolate AI long enough (or a political deal system) there will be mutiny, think the divide of Roman empire. Then you have to conquer the "rebellions". In this case, the game could go on forever, because of the transportation/communication/logistics limitations and terrains as natural barrier.
I think using enforcement learning (GAN) could spawn very interesting AIs that can inhabit certain types of procedurally generated terrains to form a "country" in the game, then we can have dynamic campaigns, like Clash of Clans but with more variety.
I'll get back to you shortly, but can you briefly clarify how you wanted to use GANs for reinforcement learning? I'm studying this stuff and I've never heard of that particular combination. Next best thing I could think of is a actor-critic model for Q-learning.
Edit: So yeah, you could use an actor-critic model for reinforcement learning (and from my limited understanding, it makes the most sense, considering you have a large action space). But really, GANs and RL are two different matters. GANs help you create things that are alike a dataset. They imitate. Maybe you could imitate a player, but it won't try to play well. RL actually tries to maximize some utility, so you need a goal for the AI to follow.
Mutinies might well follow from having an AI that is trained to maximize it's own direct influence (however that will be measured). So if you teach an AI to a) play as part of a team and obey orders (because the higher-up can punish it) and b) organize a team and give orders, and that AI knows that it'd rather lead than follow, it might take you up on any opportunity you give it.
LOL dont take this too seriously I am total n00b to this. Just buzzwords I found on openai and stuff.
I assume RL and maximize the terrain aspects for AI to learn "tactics", and use GAN to train "heros" maybe? It would be fun to have this family tree style traits from randomly generated "hero" where you can recruit and battle, marry and have descendants, Crusader Kings style.
Ahhh ok. Your description fell right in between "knows what he's talking about, has unique ideas", "knows what he's talking about, but expresses it non-standardly" and "buzzword bingo" and I wasn't quite sure.
But yeah, you could use RL to teach an AI to do pretty much anything you can easily quantify. Some things are harder, particularly if the way to that quantifiable goal is non-direct. But terrain usage? Sure.
You can use GANs to generate anything you have known samples of. Give me a few 1000 cat pictures and I can generate more of them. In principle, you could condition on a state and teach a GAN to play like a human, which would be interesting to create AI that is not too powerful. Think of a chess engine that won't roflstomp an amateur player, but won't have it's own machine-like playstyle either. It'll pretend being an amateur. So either you could generate that, or generate traits from a set of samples, if your traits are non-trivial to generate a-priori. (If you look at the big-5 personality traits, I think they're basically fitted to distributions in real life and made to be statistically independent, so generating plausible big-5 traits samples for a person is trivial. Other traits are harder to generate, for example when features like strength, dexterity, intelligence, physical appearance etc interact in complex ways. I can't just sample strength and then sample physical appearance independently, there's an obvious, but non-trivial correlation there. So I could use a GAN there.)
But really, for family tree style traits, I would go for a genetic algorithm. Maybe if you're feeling really fancy, you could learn your crossover operator (via NN) so it leads to more interesting (read unpredictable) results. That could be fun.
Yes, I imagined in the hypothetical strategy game, we can have terrain based habitants and their kingdoms, for example forrest people have a strong bow and arrow trait, their heroes tends to use ranged attack tactics against an invasion. For a civilization in a great plain, they are more likely to use pikes, maybe.
But yeah every hero can use either pikes or arrows, but their trait difference can amplify their damage.
So If you have a weaker army of pikes, you need to "capture" a hero or city from another region, or grind your hero to a high level by experience, or wait a miracle genetic mutant to happen to one of your new born heroes.
Because each traits counter each other, you have choose very carefully how you advance your army. Use the terrain smartly in defense or offense.
In this design we can lower the barrier of learning for newbies to the game, for example if you start city a town no near to the sea or river, you don't have to learn the naval skills at all. However in later games after you conquer a costal city you can delegate your original city to AI, and start learn how to manipulate boats!
This gives continues fun for early games, mid game battles and late game expansions.
If you end with a large empire, your most remote AI would more likely to start a mutiny and declare themselves an independent country. That would be chaotic fun! You can also trick your opponent countries to split, cost zero soldiers to neutralize a nemesis!
I'd like a proper successor to Victoria 2, that manages to improve QoL without streamlining all the depth out. Unfortunately paradox seems to be going the latter direction of late, but who knows what the future holds.
Hardly a novel concept but give me an expansive open-world rpg (basically Skyrim) set in the Star Wars universe and I doubt I'd touch another game for a long, long time.
A game with a world like Fallout 4/Skyrim but the mechanics and story/quests of a classic CRPG. I really loved New Vegas and it was sort of like that (the world was a bit weaker than 3 though I think) but there hasn't been a game like that since and I don't really see there ever being one, the financial incentives don't really align. But still, that would be my absolute dream game.
What I've wanted for a long time is a game that has multiple levels to it; a strategic level where you shuffle around units and a down on the ground level where you're building things or shooting things up. And it would be cross-platform; you can play the strategic level on mobile like an idle game, click here and there to spend cash to create incentives for the people who are playing the FPS or minecraft-clone to do things.
This was already kinda done with EVE Online and their FPS (can't remember the name of it) but it didn't feel very integrated and the politics and strategy weren't really visible in the on-the-ground game, it'd also be nice to not just shoot things up; it'd be nice to build a city or something.
While this is kind of where my idea came from (playing CnC renegade and thinking how cool it would be to drop into renegade in a normal CnC match), I've since decided that I'd rather have it all in one cohesive game. I've outlined that a bit here: https://tildes.net/~games/3ey/whats_your_game_that_youd_really_like_to_see_made#comment-10t9
I think it will just in general improve the sense of being present in the same game. Also, creates no artificial boundaries which makes the game easier to reason about.
A Breath of the Wild-style Pokemon game. I'm not much into Pokemon, but an open-world exploration including to-scale Pokemon would be sooooooo dope.
Definitely this. From what I understand something is on the works like this (not Let's Go Pikachu or w/e)
Not quite what you want, but there's Pixelmon which is a Pokemon mod for Minecraft. It has all of the pokemon roaming around to battle/capture, it has traditional turn-based battle mechanics, and there are little towns with people, gyms, shops, and pokecenters. It's not bad at all.
Ready Player One - complete with all the IP from the book.
It'll never happen, but I can dream :)
Just so you all know, someone copied this thread and put it on AskReddit.
A successor to Lord of the Rings: Third Age. Sure, I could pick up an older Final Fantasy game, but ....
I have it in my library, but haven't gotten into it yet. In my defense, I couldn't sink hundreds of hours into hoi4 and Stellaris and then ignore a paradox humble bundle.
I wonder if such a thing would emerge from my concept, but I'm probably expecting too much from people. :D
I just want games like Dark Souls to have unlockable debug modes that, after I've played through the game once, let me create a character to my specifications for PVP or Co-op.
Civilization but an mmo where everyone is in the same game.
Late to the thread, but I'd like a single player stealth game about the SOE during WW2. The training they went through in the UK is a perfect tutorial level, and then you could be sent to various areas of the war to carry out sabotage, reconnaissance, organization of local resistance, etc. There could also a pretty wide range of character options, since spies had to be able to blend in well and they recruited women, disabled folks, older people, people of color, etc.
That was a deep dive into wikipedia. Thanks. Definitely an interesting topic. Should we tell the Battlefield devs? They can wiggle their way out of their women-in-ww2 shitstorm that way. :D
That was actually what got me thinking about the potential for an SOE game! One agent in particular popped into my head when reading about the uproar about one of the trailers-- Virginia Hall, a female spy with a prosthetic leg. (Not an arm, like in the trailer, but still~)
I'd enjoy an MMO RPG similar to destiny, but with a more intense focus on the hardcore side of the community. This means more grinding and raids and just overall what I believe Destiny should be.
All I want is a Mega Man Legends reboot. Capcom please!
Something strongly inspired by the lore, atmosphere, basic mechanics and multiplayer of the Dark Souls series, but without the mediocre balancing or awful netcode.
A 2D Space based rouge lite.
Risk of Rain with Command mod style upgrades.
An Elite style procedural generated universe.
Combat similar to Subspace.
A finished Star Citizen is all I need