A Life Simulator (Sims) with Less Focus on Graphics

How do you think a life simulator game such as Sims would do if the focus was less on graphics. I'm talking maybe Sims 2 quality graphics with some updated shaders.

Without the focus on graphics, they could create tons of actual gameplay and content, which I feel the Sims 4 is very lacking. The world could be fully fleshed out so you can actually do literally everything, such as opening a specific drawer in your room and storing a key in there, or having the option to go into a store and buy specific things, or even steal. The personalities, traits, and different actions available could actually mean something instead of just playing different animations. I feel the possibilities for this game type are endless and completely unexplored if the focus is more on gameplay and not just how good we can make the graphics.

 

I love The Sims but isn't there already a Sims game with Sims 2 quality graphics..called Sims 2?

Having been part of that community in the past, I can tell you that anything less than actually Sims quality, graphics and mechanics, will not be acceptable to them. lol I love the idea of a life simulator but you may want to start from scratch, not use The Sims as a jumping off point. It is far to easy to make a bad The Sims clone than making a new, unique, life simulator.

 

Sims don't have a high bar for graphics anyway, they already prioritized the AI other complex shader and effects. I think you overestimate the graphics and underestimate the asset production cost, even at lower graphical fidelity it would be hard to pump all the objects and have the ai react to it (ie having correct animation variation to different objects type and tie breaking for many kind of interaction, on top of emotional variation).

Unless you figure out a way to have fast "lifelike" (not necessarily realist) animation at low cost driven by a cheap enough to compute AI but strong enough to cover all the cases for the hundreds of smart objects and character in a coordinate plausible way (take a breath).

 

I wasn't really talking about making an exact copy of the Sims.. I just used it as a reference to know which type of game I'm talking about since there aren't that many life simulators.

I really don't agree. Their AI is mindnumbingly simple. Most of the actions Sims can do are not actually doing anything except playing a different animation with no impact on the world. It has a great illusion of complexity with very little actual substance in the game, very little is actually going on. The customization of content and content itself for Sims was at an all time low in Sims 4.. unless you expect everyone to buy $300+ worth of expansions just to have a decent amount of content.

Also, all the things you mentioned are related to having a focus on graphics and animation.. so I don't really see your point. The exact idea is to have a fast lifelike animation that is not as realistic so the focus can be on developing content and not just cool animations and graphics that, in the end, don't add much to the gameplay.

 

That was my point, it's simple because it's a harder problem that you might think about, there is enough talk about this kind of AI and the problem they pose, in term of gameplay too (like feedback) that's a known quantity. You will be surprise how even such a simple AI can quickly become obscure to communicate to the player (which is essential for the AI to be perceived as smart), or have exponential explosion of computation to reduce, or to find simple production ready way to implement social convention in a timeframe that isn't the heat death of the universe.

It's a graal that everybody has been chasing and the current crown holder is VERSU a game that is entirely text based.

 

My problem wasn't exactly that the AI itself is simple, but the resulting actions of said AI don't really impact the world. Example when someone is angry and they poop they do an angry poop animation. Which is a nice little feature.. but when all the AI's actions personality, mood,and such only amount to that and don't actually do anything but play different animations and have no impact on the world around them. The system they have in place would be fine but it's not fleshed out at all, and feels like they were heading that way and just didn't finish, which leads me to believe all the effort they spent on making nice graphics and animation took up too much of their time away from gameplay. Each of those animations for angry poop and everything else takes time, and I think the effort takes away from the AI having meaningful impact on the world.

Obviously art and coders are separate but they can only appropriate a certain number of hours to their workers and I'm inclined to believe they spent a great deal more effort on graphics than gameplay.

 

But that's speculation, not only they have spoken about their design choice, but the AI game dev community have been facing problem for a long time into turning AI into something playable, there is many issues I don't want to cite here you might hadn't considered (mostly about how the state of the world is communicated to the player and player's reaction to that), and I was just pointing that, even that simple AI was a hard thing to solve, let alone MORE.

Also why I redirected you toward versu, which the most flesh out example of social AI and gameplay flair.

 

I agree if you are focused on Sims 2.

But..Sims 3, which had much more elaborate AI allowed the Sims to impact their world all the time. Not only would the Sims become AI when you were not playing them, but they would actually interact with other AI elements. I would go back to a family I had played previously and found out that not only did my guy sim find himself a lover, but he also became a dad...all while I wasn't playing him. Families would move into cheaper houses, get new jobs, move in with other Sims, get married, fall in love, get divorced, and sometimes even die. I loved the unpredictability of the game....or rather the simulation since there was not a lot of game to it.

Sims 2, on the other hand, was more of God-Mode game. Your sims did little without your direction.

I have not played Sims 4 as I decided it was way too much time and expense to start over again and even at the end of my Sims 3 time, I was mostly making new worlds rather than playing. I can do that in Unity. lol

If I were ever to make or consider playing another life simulation game, I would much more prefer the AI of Sims 3.

 

I was talking about The Sims 4 specifically, the other Sims games were great.

 

Ah, the only one I haven't played.

 

Although unrelated, I always wondered though how much their method of making tons of expansions has cost them customers. I was extremely into Sims 2, but had bought like 6 expansions by the time Sims 3 came out I was interested, but not interested at all in buying a bunch of more expansions. Even if I was rich it seems kind of ridiculous to pay $300+ for all expansions, considering how little you get in each one. They added a new bowling expansion or something where you can't even keep track of your score or anything and it's just showing them doing bowling animations while the ball knocks over some pins. This is the type of thing that I'm wondering why they would not flesh that out, it's a freaking bowling expansion pack and the only thing that is nice is how it looks. You can't even really bowl as it's actually played. This is the type of decision they do very often that leads me to believe they're focused way too much on graphics. That is how a majority of things in the Sims world are, things look nice, but there's nothing to actually do.

That was also just one aspect I feel is lacking, the AI. They can say one thing publicly just to appease people and actually be doing something totally different in their actual design process. EA is one of the only companies that I actually see as greedy, choosing whatever will make the most profit over whatever will make the most enjoyable game. EA has gotten a lot of hate over the past few years over seemingly greedy decisions. Since we are not in there with them designing the games, everything is speculation, but a series of decisions in their game design is turning a lot of people away either way.

Their reactions to criticism is not that great either. After Sims 4 came out it got tons of hate for lacking some basic features, and EA's response was "If Sims 4 isn't successful, we will get no Sims 5" and so their decision was to make a ton of new expansions, each with as little content as possible.. Yeah I'm not seeing Sims 5 anytime soon.

 

AI was fine in other Sims games. I cannot speak about Sims 4 though. They simplified the game so it could be played on lower spec computers and they were very open about that. Sims 3 required a gamer set up for computers to be played with all the expansion packs. I know..because I was constantly updating my computer to play them.

So...if Sims 4 is not your cup of tea, then well, okay. But still, lots of people play it. I decided not to do so but that was my choice.

If you can make a game with better AI than the Sims games, then good for you. But...before you tear apart the game, try to do it. I think the games are all, even Sims 4 which is rather dumbed down, are amazing games with unique mechanics that appeals to a wide cross section of gamers, from casual to hard core (believe me there are hard core Sims players) and to men and women of all ages.

Again, try to do it better. I would love to see more life simulation games. But move out of trying to make a better The Sims game. Make your own game.

 

The Sims 4 does a lot right, but I think AI is really toward the bottom of that list

Their AI has been known to just randomly stop in the middle of rooms doing nothing when they are starving or have to go to work, or having to go to the bathroom really bad they'll just be standing there. I just find it kind of humorous when people talk about how advanced the AI is in this game when it really only appears advanced on the surface when you have lots of Sims walking around doing stuff it's 'close enough' to real life that you won't really analyze it any further. But when it's just one or two sims around and you start analyzing it further you notice glitches like that where people will just stop doing anything you start paying attention to how unintelligent the Sims really act.

Most of the moods are not that complex either, it seems as simple as "Mood value > X" or "Mood value < Y", look for object to do actions associated with that mood in Z range, 'pathfind to that object's room', 'interact with found object', and then place priorities upon actions to do next on a sort of list of queued actions. But they couldn't even get that part completely correct because of the issue with Sims just stopping and doing nothing for hours. Same thing happens with queued actions where actions will be queued but pathfinding fails and they stop all actions completely. Sometimes the player has to tell a Sim to do an action over and over before they finally do it, the Sim will cancel the actions and just stand still in a room or do nothing instead.

 

That is too bad. Glad I stopped with Sims 3. I loved that game, lots of AI surprises that were delightful.

 

Sims needs a lot of very good animation. Tons.

I don't think the AI would be that "hard" - but it will require constant iteration and revision. Even if the problem itself is fairly clear, making things 'feel right' is time consuming, high detail work.

At a high level - things like mood = find item - these are not where the real work is. Making behaviour seem full, fluid and complete will be much more work than you expect.

_______________
I do agree though that this is fertile ground, not much has been worked on in the area and there is a lot of great experience left here without needing huge budgets.