• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Graphical advancement vs Gameplay advancement

Dice

Pokémon Parentage Conspiracy Theorist
This came back to my mind because of X, but really it applies to SO MANY games. Basically, I see technology advancing so far that the visual depiction is at odds with what happens in gameplay, and the better graphics get, the greater this chasm becomes. It happens in a few different ways...

Wounds: Does it bother you that we're still often just attacking enemies that act completely unharmed until they simply keel over? We seem to have the concept of the headshot down, but beyond that you're lucky to get much else. Shooters do try to seem to at least have human enemies act wounded for 5 seconds before they go back to 100% mode, but for things like X there is pretty much nothing. You have this massive creature and you're shooting it and slashing at it and there is no signs of damage either visual or behavioral until it just dies.

Some are trying to move things forward significantly. Dead Space and Binary Domain especially, but other than them, I just don't see much progress in this. Games are actually working towards ways of gimping the player when they get hurt at a faster rate than figuring out how to affect the enemies in a similar way. Bottom line for me is that if I shoot a huge beast with a rocket launcher, a chunk of that beast should be gone. Anything less pulls me out of the experience because the visuals (image and movement) are too real anymore.

Destructible environments: This seems to be getting more attention, but it's still incredibly limited. After playing Red Faction: Guerrilla I was hoping to see the idea take off with middleware helping developers do this more often and easily, but it just didn't take off. Battlefield now seems to be moving forward, but I don't see much else happening.

Town scale: There seem to be valid attempts to move this forward, but it's fairly slow. Just Cause 2 and Far Cry are getting the "vast jungle area with spots of villages" thing down, and GTA and such are developing the city concept. On the other hand, you have stuff like Skyrim where you are supposed to feel like the hero of a nation but the supposedly grand cities of the land are filled with maybe 30 people. To me they come off like a joke because it's not like you're playing a Final Fantasy on SNES anymore. Mass Effect at least tried to convey larger metros while trapping you in a small corner of them.

AI: This gets more complicated since you're messing with player expectations and gameplay focus and such, but AI doesn't only have to mean combat tactics. Advanced AI could be used to create entirely new genres. I play something like L.A. Noire and while the conversation system is interesting, it's still just taking a linear path through the branching options. Maybe not with this particular game, but in others we could have really dynamic personalities and work out a new kind of game entirely. Antichamber makes me think we could give environments AI somehow...

The problem: As the title of the thread implies, I think the main problem with these is the huge investment of graphics development. We are constantly trying to push that bar forward, and it's incredibly expensive to do so. Anything you expand would have to be visually rendered at a high quality. But I wouldn't mind falling back to say, Left 4 Dead 1 visuals if that's the investment tradeoff that would be required to get these other gameplay dynamics in working order.

Ultimately, that kind of decision is what it's going to take. There are projects with "older" graphics on Wii and such, but the Wii has the old hardware limitations on gameplay as well. The only way to move all this forward is to have high hardware resources but then choose to use them for gameplay advancement rather than pushing the graphical detail bar higher. How long will we go before we finally decide that the image is pretty enough and the rest needs to catch up?
 
The best possible outcome of the next generation is:

- Technical graphics stay the same, or even recede if necessary.
- The extra processing power and budget that is freed up is used to power and fund resource-heavy gameplay advancements.

Of course, this will never happen because 1) trailers of shiny games sell and 2) innovating is much harder and riskier than polishing.
 

meta4

Junior Member
I agree with everything you said. I am actually really disappointed that AI has not evolved much in games. If there is one thing I would like next gen is a huge enhancement in AI.
I am hoping TLOU does something great in the AI and holds as good as the video's show.
 
Also, I want all action games with Meele to feature mutilations a-la Soldier of Fortune, it would be really nice, we were getting there in the 90s but now enemies just fall to the ground when dead :(
 
My most memorable moment of this gen was Red Factions building destruction. Shame the game was dull and the sequel was utter crap.

The tech needs another look in and brought into next gen.
 

cdevl

Junior Member
Yep. Agree with the majority.

Priority:

1. Gameplay
a. AI
b. Mechanics

2. Better story / settings
......

10. Graphics
 
I agree with most of your points except for the wounds one.

as much as I'd love to see more detailed wounds and wounds in general appear on enemies, I think it would result in games becoming too graphic and creating a slippery slope where developers try to outdo each other and we end up with a manhunt 2 situation where games end up causing outcries from the sensitive steve's and sally's of the world.

I'd love to see it though. bullets tearing through flesh and coming out the other side. bones protruding from limbs in beat em ups and so on.
 
You make an interesting point. The more graphics advance and gameplay stays stagnant, the more disconnect will happen. I personally would like to sees strides made in animation and physics across the board. Also, gameplay mechanics should always take priority.
 

BlackJace

Member
AI is my number one complaint I want rectified. For the stealth genre to really shine, AI needs to be excellent.

Also, your part about wounds was very interesting. Hope that takes off a la what I saw from X.
 

KirbyKid

Member
1) You should learn to Embrace the Abstraction. Everything you said basically is in response to making games more realistic to match a sub-set of more realistic looking games. If you play some more abstract games or learn to embrace the abstraction in games, you wouldn't be taken out of the experience so easily.

2) Plenty of games do have the detailed gameplay cause and effects that you mention. If you play Fighting games, there are many unique hit stun animations and hit effects that make these games rich. If you play a shooter like Kid Icarus Uprising, you also have a variety of stun, knock down, knock back, etc. effects.

3) If you're talking about games like X, it's clearly moving in a particular direction as an evolution of other types of games (mainly RPGs though). They're merely exploring the branch or style of representation/presentation that they like. The trick is to embrace the gameplay systems, embrace the visuals, and let them gel in your mind. After all, artistic works are artifices designed to convey ideas and experience mostly rather than be straight simulations or direct representations.


I see that people not actually talking about gameplay design and gameplay advances to be a far greater problem. In my experience, even when GAFers think they're talking about gameplay, they're really just talking about visual effects and non-gameplay concept that are related to interactive systems, but not necessarily gameplay systems.

The beauty of abstraction is to focus experiences that the conveyance of ideas on particular details.
 

LiquidMetal14

hide your water-based mammals
There are many aspects to this that make certain games types fun while others in the same genre not so fun. Core game play decision, map design, and reward structure are a few things that can make or break an online game. Other SP genres can suffer from this as well.

I'd say some of the funnest games might not have been so fun but playing them with friends really enhanced the experience.

Many of my most played games have been either online or fairly non linear.
 

i-Lo

Member
One thing about the wound I would like state is that as we move closer to reality, I want the gore to be less specific. Perhaps right now, were are still away from the apex but come the time of PS5 or 6, I really hope not to be looking at a highly realistic individual as swipe of the sword cuts him in half. from top to bottom

However, in regards to portraying progressive damage according to the health situation of the character, I concur.
 

Xanadu

Banned
as much as I'd love to see more detailed wounds and wounds in general appear on enemies, I think it would result in games becoming too graphic and creating a slippery slope where developers try to outdo each other and we end up with a manhunt 2 situation where games end up causing outcries from the sensitive steve's and sally's of the world.

fuck em
 

Kimawolf

Member
I agree with most of your points except for the wounds one.

as much as I'd love to see more detailed wounds and wounds in general appear on enemies, I think it would result in games becoming too graphic and creating a slippery slope where developers try to outdo each other and we end up with a manhunt 2 situation where games end up causing outcries from the sensitive steve's and sally's of the world.

I'd love to see it though. bullets tearing through flesh and coming out the other side. bones protruding from limbs in beat em ups and so on.

Well that'd earn any beat em up a M rating, and with things seeming to get stricter I only see this kind of thing regressing. And I'd love for the extra power be put into other areas as well, truly interactive environments/lush environments with lots of physics, all kind of random AI Routines going on, something that really feels like a world, not just a level in a game. because once the shiny graphics wear off you're left with just the gameplay.
 

Dice

Pokémon Parentage Conspiracy Theorist
I agree with most of your points except for the wounds one.

as much as I'd love to see more detailed wounds and wounds in general appear on enemies, I think it would result in games becoming too graphic and creating a slippery slope where developers try to outdo each other and we end up with a manhunt 2 situation where games end up causing outcries from the sensitive steve's and sally's of the world.

I'd love to see it though. bullets tearing through flesh and coming out the other side. bones protruding from limbs in beat em ups and so on.
Well for me the wounds my issue was mainly from a gameplay standpoint. It first hit me hard when I was playing Silent Hill Homecoming. There was this weird creature kind of hobbling about slowly on one shaky leg. I figure I'll knock it over and beat the shit out of it on the ground... no go. The game had enough sense to make my character cough and gag when it spit smoke at me (helloooo, just hold your breath, moron) but when I hit the thing with a heavy melee weapon it didn't budge an inch despite looking like it'd fall over on its own.

Like I said, it strikes me the most with these large creatures. I keep hitting it over and over and over, often in the foot, then it dies. It doesn't even start limping with the leg I was hitting. It's a joke. Dead Space creatures were very resilient and the violence graphic but at least there was a physical response to my attacks in proportion to my attacks. That's what I'm looking for, accurate graphical depiction or not.
 
Well for me the wounds my issue was mainly from a gameplay standpoint. It first hit me hard when I was playing Silent Hill Homecoming. There was this weird creature hind of hobbling about slowly on one shaky leg. I figure I'll knock it over and beat the shit out of it on the ground... no go. The game had enough sense to make my character cough and gag when it spit smoke at me (helloooo, just hold your breath, moron) but when I hit the thing with a heavy melee weapon it didn't budge an inch despite looking like it'd fall over on its own.

Like I said, it strikes me the most with these large creatures. I keep hitting it over and over and over, often in the foot, then it dies. It doesn't even start limping with the leg I was hitting. It's a joke. Dead Space creatures were very resilient and the violence graphic but at least there was a physical response to my attacks in proportion to my attacks. That's what I'm looking for, accurate graphical depiction or not.

As a side note, Binary Domain does this, but with robots.
 
As a side note, Binary Domain does this, but with robots.

That is what made Binary Domain so much fun.

I personally have been looking at indie side of games a lot more as far as gameplay innovation since a lof of AAA games are taking a safer route. However, I mainly just want to see old school PC gameplay with shiny graphics.
 

Stumpokapow

listen to the mad man
Wounds: Does it bother you that we're still often just attacking enemies that act completely unharmed until they simply keel over? We seem to have the concept of the headshot down, but beyond that you're lucky to get much else. Shooters do try to seem to at least have human enemies act wounded for 5 seconds before they go back to 100% mode, but for things like X there is pretty much nothing. You have this massive creature and you're shooting it and slashing at it and there is no signs of damage either visual or behavioral until it just dies.

This won't be fixed, because consider the implication in terms of what would happen when enemies shoot you. If every time you got shot you died, that would be very frustrating. If every time you shot an enemy they winced and yelled and rolled around and basically any body-shot was enough to completely incapacitate them (like a real human), it'd really prohibit any kind of fast or arcadey or gutsy movement/level/player design. This isn't a tech thing, it's a game design thing. Bushido Blade is what it is, but not every game is Bushido Blade, nor should it be. Metro 2033 had Ranger Mode, which moved closer towards that, but it carries with it an extremely high degree of difficulty and it wouldn't be suitable for most games in most cases. But most shooters have a "pain" mechanic, FYI, going right back to Doom.
 
There have been games that tried to do some of the things you said but you already hit on why they didn't sell well - the graphical trade off. Red Faction: Guerrilla had a great destruction system but it didn't look good. They could of probably made the cities more dense but then there would of been an even greater hit on the graphics.

And your suggestion to just focus on gameplay for awhile so it can catch up without a visual toll probably wouldn't work because there'd be other games where graphics is all they focus on so what looks good now won't look good later. But then again perhaps the disparity between the great look games and ok looking games will be small enough for games to focus resources elsewhere and still be successful.
 

graywords

Member
Wounds: Does it bother you that we're still often just attacking enemies that act completely unharmed until they simply keel over? We seem to have the concept of the headshot down, but beyond that you're lucky to get much else.

A fun experiment into the wounding concept was used way back in the PSX days - Bushido Blade. I'd like to see something similar to that in a decent-budget modern game. (thus excluding things like Deadliest Warrior)

Also: beaten horribly to mention of Bushido Blade. Oops.
 

Kai Dracon

Writing a dinosaur space opera symphony
I would say a balance is needed between much of the OP's points and what KirbyKid posted in regards to abstraction. Games are about abstraction, and it's easy to get lost in a hall of mirrors when you begin thinking too literally. Eventually, you end up striving for nothing more than a 1:1 simulation of reality, and aside from being technologically and artistically draining, that introduces barriers to actual game design.

However, I think AI is the most universally salient point brought up by the OP. AI is something that does directly enable game design, mechanics, and play strategy whereas much of the other suggestions are largely cosmetic.
 

Dice

Pokémon Parentage Conspiracy Theorist
1) You should learn to Embrace the Abstraction. Everything you said basically is in response to making games more realistic to match a sub-set of more realistic looking games. If you play some more abstract games or learn to embrace the abstraction in games, you wouldn't be taken out of the experience so easily.

2) Plenty of games do have the detailed gameplay cause and effects that you mention. If you play Fighting games, there are many unique hit stun animations and hit effects that make these games rich. If you play a shooter like Kid Icarus Uprising, you also have a variety of stun, knock down, knock back, etc. effects.

3) If you're talking about games like X, it's clearly moving in a particular direction as an evolution of other types of games (mainly RPGs though). They're merely exploring the branch or style of representation/presentation that they like. The trick is to embrace the gameplay systems, embrace the visuals, and let them gel in your mind. After all, artistic works are artifices designed to convey ideas and experience mostly rather than be straight simulations or direct representations.


I see that people not actually talking about gameplay design and gameplay advances to be a far greater problem. In my experience, even when GAFers think they're talking about gameplay, they're really just talking about visual effects and non-gameplay concept that are related to interactive systems, but not necessarily gameplay systems.

The beauty of abstraction is to focus experiences that the conveyance of ideas on particular details.
Sorry, but this is 1) assuming of my mindset and 2) bullshit in context to the conversation.

Using the word "abstraction" and saying to simply let go of the issue is not a solution to the problem. What I am talking about are very real gameplay dynamics. Cause and effect, range of possibilities, interactivity, exploration. I don't give a shit if you want to say something is an abstract representation of something if the manifestation of that ends up meaning the exact same form of interaction I've been having with games for the last 15 years.

I'm all for independent, innovative games. More and more I've been looking into new kinds of games and enjoying what I find there, and they are most often very abstract in what they are conveying because they are low budget. That isn't what I'm talking about in this thread nor is it the solution to the problems I am presenting in this thread. What we are talking about here is progress of gameplay on the path of the perceived goals of the established top genres.

If you want to talk about meta-progress of gameplay of videogames on the whole, innovating new types of it and how it might be represented in new kinds of games, you can go and make a thread about that. That isn't what we're talking about here and I don't want the thread derailed in such an obnoxious way. We are all very damn well acquainted with the idea of abstraction. We've all been utilizing it since we first made a little guy made out of 16 pixels "jump" over a gap. This thread is about the boundaries of where we can continue making things less abstract and more real in interactive rather than purely visual ways.

This won't be fixed, because consider the implication in terms of what would happen when enemies shoot you. If every time you got shot you died, that would be very frustrating. If every time you shot an enemy they winced and yelled and rolled around and basically any body-shot was enough to completely incapacitate them (like a real human), it'd really prohibit any kind of fast or arcadey or gutsy movement/level/player design. This isn't a tech thing, it's a game design thing. Bushido Blade is what it is, but not every game is Bushido Blade, nor should it be. Metro 2033 had Ranger Mode, which moved closer towards that, but it carries with it an extremely high degree of difficulty and it wouldn't be suitable for most games in most cases. But most shooters have a "pain" mechanic, FYI, going right back to Doom.
That's why it mostly bothers me when fighting massive creatures.
 
1) You should learn to Embrace the Abstraction. Everything you said basically is in response to making games more realistic to match a sub-set of more realistic looking games. If you play some more abstract games or learn to embrace the abstraction in games, you wouldn't be taken out of the experience so easily.

2) Plenty of games do have the detailed gameplay cause and effects that you mention. If you play Fighting games, there are many unique hit stun animations and hit effects that make these games rich. If you play a shooter like Kid Icarus Uprising, you also have a variety of stun, knock down, knock back, etc. effects.

3) If you're talking about games like X, it's clearly moving in a particular direction as an evolution of other types of games (mainly RPGs though). They're merely exploring the branch or style of representation/presentation that they like. The trick is to embrace the gameplay systems, embrace the visuals, and let them gel in your mind. After all, artistic works are artifices designed to convey ideas and experience mostly rather than be straight simulations or direct representations.


I see that people not actually talking about gameplay design and gameplay advances to be a far greater problem. In my experience, even when GAFers think they're talking about gameplay, they're really just talking about visual effects and non-gameplay concept that are related to interactive systems, but not necessarily gameplay systems.

The beauty of abstraction is to focus experiences that the conveyance of ideas on particular details.

I thought I was the only one who thought this all along, thank you so much for sharing this. I always get a little bummed out when people, even on NeoGAF, discuss new games and then talk a lot about the advancements in graphics and realism, not gameplay. There is nothing inherently wrong with advancements in graphics and realism, of course, but the focus on that particular aspect of videogames is way to strong.

I am not alone ;_;
 

sono

Member
Great post. Its kind of interesting when you have lower quality graphics your imagination fills in the rest,

Is a part of the answer that more sophisticated game engines are required that cover the points you mention in effect automatically (e.g wounds, types of AI) so the developers can focus on gameplay.
 
A lot of people say 'better AI' without really understanding what that means for AI.
People don't actually want "AI" in games, because actual AI (Artificial Intelligence) would be no fun to play against.

Don't just say "better AI"; really think through what it is that you actually want, and how it could be achieved.

NPCs that learn from your actions and don't fall for the same trick twice, so your available in game actions decrease with each encounter?
NPCs that can use their superior reflexes, aim, and timing to destroy you with no means of beating them?
NPC bosses that have no pattern recognition gameplay or 'trick' to defeat them, so that you can't beat them?

What does 'better AI' bring to the table for gaming, outside of filling PvP slots?

EDIT:
I say this because a lot of research has been done into AI, and a lot of designers have done a lot of experimentation with AI, and used cutting edge AI techniques from academics to develop AI routines.

These are almost universally no fucking fun to play against.
 
This came back to my mind because of X, but really it applies to SO MANY games. Basically, I see technology advancing so far that the visual depiction is at odds with what happens in gameplay, and the better graphics get, the greater this chasm becomes. It happens in a few different ways...

Wounds: Does it bother you that we're still often just attacking enemies that act completely unharmed until they simply keel over? We seem to have the concept of the headshot down, but beyond that you're lucky to get much else. Shooters do try to seem to at least have human enemies act wounded for 5 seconds before they go back to 100% mode, but for things like X there is pretty much nothing. You have this massive creature and you're shooting it and slashing at it and there is no signs of damage either visual or behavioral until it just dies.

Body parts are cuttable, as in Monster Hunter kind of way.

Who the hell needs wounds to be shown in an RPG anyway?
 

Jive Turkey

Unconfirmed Member
I agree fully with the items in the OP. The only problem is the graphics are an easy selling point for games simply because you can't see complex AI interactions in screenshots. Nobody flooded threads with Killzone 2 gifs showing off the AI.

A lot of people say 'better AI' without really understanding what that means for AI.
People don't actually want "AI" in games, because actual AI (Artificial Intelligence) would be no fun to play against.

Don't just say "better AI"; really think through what it is that you actually want, and how it could be achieved.

NPCs that learn from your actions and don't fall for the same trick twice, so your available in game actions decrease with each encounter?
NPCs that can use their superior reflexes, aim, and timing to destroy you with no means of beating them?
NPC bosses that have no pattern recognition gameplay or 'trick' to defeat them, so that you can't beat them?

What does 'better AI' bring to the table for gaming, outside of filling PvP slots?

The AI I'd like to see improved on is non-combatant AI primarily in open world games. It sucks to be driving around town and have the city's citizens diving for cover...Directly in the path of your car. Feels like you're in a city full of suicidal mental patients.
 
There should be more complex damage and healing I agree.
AI will always be messy; and for the record ND, Last of Us AI are basically stealth AI, its been around for fucking years. Get off your high horse.

AI remains the annoying point for me, because they all wank off with it.
I do accept its difficult though; which is why level design should come first.
 

Derrick01

Banned
AI should be the number 1 thing people work on advancing next gen. It would improve the challenge in so many games and make the player get more creative in handling situations within the environment.
 
It seems like better graphics is an easier thing to understand and create than solving the problems you and others have talked about.

Which I'm guessing is why we don't focus on fixing them unfortunately.
 

Bombadil

Banned
This won't be fixed, because consider the implication in terms of what would happen when enemies shoot you. If every time you got shot you died, that would be very frustrating. If every time you shot an enemy they winced and yelled and rolled around and basically any body-shot was enough to completely incapacitate them (like a real human), it'd really prohibit any kind of fast or arcadey or gutsy movement/level/player design. This isn't a tech thing, it's a game design thing. Bushido Blade is what it is, but not every game is Bushido Blade, nor should it be. Metro 2033 had Ranger Mode, which moved closer towards that, but it carries with it an extremely high degree of difficulty and it wouldn't be suitable for most games in most cases. But most shooters have a "pain" mechanic, FYI, going right back to Doom.

Excellent post. I've thought about this as well, and my conclusion to this dilemma was to basically move away from shooting and violence as the main focus of the game.

It would beneficial to games, and society in general, if developers changed their focus from shooting to puzzles and logical quandaries and exploration. In that way, they can advance the gaming medium as well as making combat wounds more realistic and less common in the game. Maybe you'd only be forced to injure or kill a handful of people, and each of those instances would be so realistic and heavy in presentation that the player would be discouraged from engaging in violent acts.

Perhaps that would even reduce the level of desensitization that violent media are said to induce in impressionable people.
 

Dice

Pokémon Parentage Conspiracy Theorist
whereas much of the other suggestions are largely cosmetic.
I don't think it's merely cosmetic when a Mammoth in Skyrim doesn't act like I've been slashing at it's leg for the last 3 minutes before suddenly dying. Especially if I am trying to be more nimble than it as a strategy.

I don't think it's merely cosmetic when my buddy downstairs is in crisis so I use an explosive charge on the floor to get down to him before he is killed.

I don't think it's merely cosmetic when the combined forces of the enemy "army" would get their asses whooped by my character, but I still have to fear them out of concept and options are unavailable to me because of that.

Yes, with all of these, the concept of abstraction plays into accepting the limitations that may be present, but they are not merely cosmetic issues.
 
A lot of people say 'better AI' without really understanding what that means for AI.
People don't actually want "AI" in games, because actual AI (Artificial Intelligence) would be no fun to play against.

Don't just say "better AI"; really think through what it is that you actually want, and how it could be achieved.

NPCs that learn from your actions and don't fall for the same trick twice, so your available in game actions decrease with each encounter?
NPCs that can use their superior reflexes, aim, and timing to destroy you with no means of beating them?
NPC bosses that have no pattern recognition gameplay or 'trick' to defeat them, so that you can't beat them?

What does 'better AI' bring to the table for gaming, outside of filling PvP slots?

EDIT:
I say this because a lot of research has been done into AI, and a lot of designers have done a lot of experimentation with AI, and used cutting edge AI techniques from academics to develop AI routines.

These are almost universally no fucking fun to play against.
Yup. It's the term I see tossed around everyday, and there are people, even within development, who confuse the "better/smarter AI" vs "interesting AI". An AI that uses cover constantly is smart in an FPS, but is it fun to play against? You'll just end up swearing at the AI up and down for being a bitch and not giving you a fair chance to shoot at.

The OP's topic evoked a slightly different, and somewhat interesting idea: Have you ever noticed game companies's staff roll/hiring postings? Wanna venture a guess what's the breakdown of programmers/designers/artists? I think you can quickly see why gameplay seems to always take a backseat.
 
It was bad enough when graphics were the only thing that really detracted from the art of game design. Nowadays there's a a bumper crop of of tech gimmicks (motion controls, touch controls, 3D, second screens, the second coming of VR) and newfangled reek-of-desperation business models (subscriptions, DLC, microtransactions) that seem to be getting more attention than fixing classic technical problems like the kind mentioned in the OP or pushing game design in a way that takes advantage of the increased hardware muscle. Decentralization will doom us all!
 
The AI I'd like to see improved on is non-combatant AI primarily in open world games. It sucks to be driving around town and have the city's citizens diving for cover...Directly in the path of your car. Feels like you're in a city full of suicidal mental patients.

It's easier to do 'always avoid' player AI than it is to have AI that fucks up and may / may not avoid you gracefully.

Play PS1 Driver, or DC Crazy Taxi.

You'll never hit a pedestrian, no matter how hard you try.

Yes, I can see the benefits of AI in non-combat oriented games.
For example a 'sleuth' type game that understands natural language and can respond based on pre-defined psychological templates to any question asked.

However, this would take years to develop, as you would be writing in depth psychological profiles for multiple characters, but more importantly, it would not sell to 'the core' who would immediately dismiss it as 'casual'.
 
I don't expect the look of games to change much. I think the graphical appearance of games for next gen are going to look very same. Lighting and shadows will appear more realistic and rendered in real time and we will see some more natural environent modeling, but I think the biggest advancements are going to be in physics and AI.

My Expectations for next gen are games that looks at least as good as Assassins Creed 3 with a physics model that allows for full environmental destruction (Geomod 1 basically, or for those to young to know the original Red Faction, think Fracture + RF:Guerilla combined) with some decently populated (and realistically scripted) NPCs.

As you can see, my graphical expectations aren't to high. With that said, I also expect mindless shooters like CoD to look impressive with real time lighting and shadow effects. The time of baked lighting is over in my opinion. Halo 5 should look like Halo 4, minus the baked lighting.
 
The AI I'd like to see improved on is non-combatant AI primarily in open world games. It sucks to be driving around town and have the city's citizens diving for cover...Directly in the path of your car. Feels like you're in a city full of suicidal mental patients.
This one, surprisingly, is a technical issue than a design issue. Most open world games (or games with massive NPC/Enemy counts) on screen use very simple (or even non-existance AI) and other tricks (like flocking behaviours) to simulate actions. The behaviour you're describing is a basic "if condition reach -> dive somewhere". Depending on the implementation of the AI, the cost of calculating and instructing each actor in a scene to calculate a dive to location may be too costly and slow the game down.
 

Jive Turkey

Unconfirmed Member
It's easier to do 'always avoid' player AI than it is to have AI that fucks up and may / may not avoid you gracefully.

Play PS1 Driver, or DC Crazy Taxi.

You'll never hit a pedestrian, no matter how hard you try.

Yes, I can see the benefits of AI in non-combat oriented games.
For example a 'sleuth' type game that understands natural language and can respond based on pre-defined psychological templates to any question asked.

However, this would take years to develop, as you would be writing in depth psychological profiles for multiple characters, but more importantly, it would not sell to 'the core' who would immediately dismiss it as 'casual'.

I'm not asking for anything too complex but I'm also not looking for anything so cut and dry as "always avoid player". Basically when my player character gets hit by a car in Saint's Row the cop who witnessed it shouldn't start shooting at me. Cops don't normally approach a pedestrian vs. vehicle collision and shoot the victim do they?

What I'm saying is there's a bit of complexity that can be added to fix the behaviors of citizen NPCs without giving each some sort of full psychological profile.

This one, surprisingly, is a technical issue than a design issue. Most open world games (or games with massive NPC/Enemy counts) on screen use very simple (or even non-existance AI) and other tricks (like flocking behaviours) to simulate actions. The behaviour you're describing is a basic "if condition reach -> dive somewhere". Depending on the implementation of the AI, the cost of calculating and instructing each actor in a scene to calculate a dive to location may be too costly and slow the game down.

Yeah I figured it was a little more technical than anything but still the scenario above should never happen. That's just silly.
 

Helscream

Banned
A lot of people say 'better AI' without really understanding what that means for AI.
People don't actually want "AI" in games, because actual AI (Artificial Intelligence) would be no fun to play against.

Don't just say "better AI"; really think through what it is that you actually want, and how it could be achieved.

NPCs that learn from your actions and don't fall for the same trick twice, so your available in game actions decrease with each encounter?
NPCs that can use their superior reflexes, aim, and timing to destroy you with no means of beating them?
NPC bosses that have no pattern recognition gameplay or 'trick' to defeat them, so that you can't beat them?

What does 'better AI' bring to the table for gaming, outside of filling PvP slots?

EDIT:
I say this because a lot of research has been done into AI, and a lot of designers have done a lot of experimentation with AI, and used cutting edge AI techniques from academics to develop AI routines.

These are almost universally no fucking fun to play against.

You make a valid point. If we really wanted competent AI, we would probably be getting our arse's kicked.

I think as a whole we would still like to have enemy encounters/boss battles possess more innovative mechanics. While a game like Twisted Metal have its drawbacks it really itched the "right spot" in terms of game play.

I do agree with the OP on just about everything. So many games are just one-dimensional and flat when it comes to gameplay. Normally when you lower/raise the difficulty it just turns into enemies having more life, and the PC having less life. Which in some cases that can be adequate enough granted a game has great game play mechanics. If it is a game like Uncharted (amidst its popularity and reviews) the game just gets downright frustrating on crushing. Some encounters are completely devoid of fun and that is a serious problem.

There is another honorable mention like a classic RPG style game like Dragon Age: Origins is a real treat on PC. There are several dimensions you have to take into account when faced with a enemy encounter. Other games like DMC3/4 had such a crazy combat system that even on the hardest difficulty if you mastered the mechanics you could literally be untouchable and it was a blast in the process.

To close my opinion I just really wished games focused on gameplay more so that games like Unreal or Quake the glory and the fun was strictly in the execution/kill alone. I think when a game can be 10 years old is when the gameplay is timeless. (Heroes of Might and Magic III).

Great post OP. Couldn't word it better myself.
 

pixlexic

Banned
As far as graphics needing to very accurate and precise I say nah .. That's just being anal the older I get the more I appreciate different art styles and rather have something that's actually fun and not something to brag about on forums.
 

Boss Man

Member
I agree with the notion that there are a lot of aspects to games that ought to be reevaluated. There's a tendency to build on what's worked, instead of thinking about what, why, or how something might work.

However, I disagree with your main premise that there is such a thing as graphics vs. gameplay. On the contrary, I think that the technology used to drive graphics ends up being one of the main drivers of gameplay innovation. For instance, the idea of injuries in a game is completely a design decision and the fact that it's not done has nothing to do with graphics holding it back. Destructible environments involve some complicated work and they are something that you are going to get more of as graphics improve, not the other way around.

If things worked the way that you're implying, then Nintendo's games should have featured more of the things mentioned in your OP. When in reality, games working on older technology tend to play more like older games (which isn't objectively bad, but it's worth note).
 

SkyOdin

Member
Town scale: There seem to be valid attempts to move this forward, but it's fairly slow. Just Cause 2 and Far Cry are getting the "vast jungle area with spots of villages" thing down, and GTA and such are developing the city concept. On the other hand, you have stuff like Skyrim where you are supposed to feel like the hero of a nation but the supposedly grand cities of the land are filled with maybe 30 people. To me they come off like a joke because it's not like you're playing a Final Fantasy on SNES anymore. Mass Effect at least tried to convey larger metros while trapping you in a small corner of them.

Why are huge towns a good thing? All they do is slow the gameplay down. Back in the days of the SNES, towns consisted of four or five buildings and about one to two dozen NPCs. Those towns were quick to explore and it was simple to go through an entire town and do all of your shopping. Not to mention, there were a lot of different towns, making for large worlds.

Fast forward to the PS2 era, where towns swelled in size to the likes of Rabanastre, which was huge in scale, but horribly impractical. Getting stuff done in town in a timely fashion requires a warp system. Even with the teleporting the town took an unreasonable amount of time to get through, since individual areas were still rather large and you had to go through four loading screens, bare minimum, to go from an entrance to a shop and back to another exit.

Now in this generation we have examples of towns like the monstrous main city of White Knight Chronicles. That town dwarfs Rabanastre in size, with more than a hundred NPCs and multiple large districts. It is also a complete chore to travel through. Despite having the same services as an older RPG town, it takes at least five times as long to perform the same errands.

To be honest, I think games like Skyrim and Xenoblade already push or surpass the limits of practical sized towns in an RPG. I have never seen a game benefit from making towns any larger. As a whole, large towns are a completely superficial element akin to pretty graphics. While a large town might wow at first impression, that grandeur quickly outlives its welcome after you have to run through the giant mess thirty times.
 

seldead

Member
GTA IV and Max Payne 3 had quite realistic and satisfying reaction to injuries. Shoot a dudes leg and he toppled over and crawled away on his other one kinda stuff. Obviously when you have character's more reactive to damage it becomes not only a technological issue but a censorship issue. Soldier of fortune style gore wouldn't have a chance of being rated commercially next gen or arguably even this gen just because of how much more graphic it would appear than it did in the 90's and early 00's.
 

Dice

Pokémon Parentage Conspiracy Theorist
Who the hell needs wounds to be shown in an RPG anyway?
Not just shown, but part of the tactics. Think of it as the evolution of status effects.

Also, mind you, the realism I'm talking about would not necessarily be introduced by implementing the aspects I'm talking about. For instance, Skyrim has a lot of "finishing moves" that take combat to that level. Plenty of other games do something similar. They give you a taste of a higher realism in select moments but then want you to accept much greater limitations all the rest of the time. It doesn't make any sense. Either be abstract or push things in the direction you implied by the things you presented to me yourself.

A lot of people say 'better AI' without really understanding what that means for AI.
People don't actually want "AI" in games, because actual AI (Artificial Intelligence) would be no fun to play against.

Don't just say "better AI"; really think through what it is that you actually want, and how it could be achieved.

NPCs that learn from your actions and don't fall for the same trick twice, so your available in game actions decrease with each encounter?
NPCs that can use their superior reflexes, aim, and timing to destroy you with no means of beating them?
NPC bosses that have no pattern recognition gameplay or 'trick' to defeat them, so that you can't beat them?

What does 'better AI' bring to the table for gaming, outside of filling PvP slots?

EDIT:
I say this because a lot of research has been done into AI, and a lot of designers have done a lot of experimentation with AI, and used cutting edge AI techniques from academics to develop AI routines.

These are almost universally no fucking fun to play against.
That's why I directed the thought beyond combat. However, within combat you are still thinking according to conventions. Taking cover, flanking, using grenades, better aim, etc. are not problematic and frustrating on their own. Let's look..

NPCs that learn from your actions and don't fall for the same trick twice, so your available in game actions decrease with each encounter?
Why should future enemies learn from a guy who is dead? This would only be a problem if you kept encountering T2000 through the game or something.

NPCs that can use their superior reflexes, aim, and timing to destroy you with no means of beating them?
More intelligent aim doesn't mean superhuman. Snipers should need some moments of stability to center in and hit you, not just pop you in the head the split second it's visible. Likewise, common ground troops should have similar hindrances in terms of "centering in" before hitting you if they just popped up themselves, accuracy loss while moving, benefit from focusing through scope but greater chance of losing track of you if you move behind cover, etc.

NPC bosses that have no pattern recognition gameplay or 'trick' to defeat them, so that you can't beat them?
This is just wrong. I know I have played games where it was a more dynamic experience. DMC got somewhat into this but it is possible to go further. With proper movement options, combat can be more about smart positioning and good reaction on your part than prebaked to expose a planned weak spot.

Furthermore, many issues stem from shooting gallery level design and the notion you always need to be fighting 20 guys. I remember a few times playing MGS games where I had extremely tense moments with just 2 or 3 guys. It was because they were reactive to my and the things happening to each other, sight and sound, and yet the level provided me with ways to try and work around it. Still, their responses kept me on my toes as I had to track them as much as myself and create my own opportunities.

This can happen just as much in a more traditional shooting game. What if you were in a scenario where you had to battle 2 guys who were really smart and just as tough as you yet also have movement and awareness contextual limitations of aim as you? We aren't looking to pop 2 bullets into them nor do we need to be paranoid about cheap deaths. It's about smart play, and the level design can be made to enhance that. Give it more aspects like smart wounds and I think you could have some winning gameplay. Similarly, they might use grenades, but only carry a couple themselves and not always know exactly where you are.

Zombie games have also pushed forward a concept I have seen rather cheaply implemented in FPS games, the concept of the blindside. In shooters I've been surprised by someone and hit with a melee attack that dazed the everliving shit out of me while 10 other guys shot me. It was cheap and not fun. However zombie games gave me a moment of pass/fail reaction to defend myself from the surprise. If failing, sometimes I would have a secondary option to dive over a counter or out of a window in my daze rather than merely stand there getting shot like a helpless doofus.

These can be combined as well, so if you're in that room with two smart guys, you might get the jump on one but your little melee encounter doesn't result in one of you dying, either because you or he bailed out, or because his friend started getting too close. You could even implement gun knock-aways and available item use so it plays out like a fight in a Bourne movie. I don't believe in problems that don't have solutions when you're talking about a world where you create every rule. Gameplay design can go anywhere. We can still push things forward to be more dynamic, engaging, interesting, and still fun rather than overbearing.

I also don't think this stuff is necessarily a hard sell, and advancements in it have been major selling points for past games like Resident Evil 4 and such. If you show gameplay doing things you could never do before, giving the player creative options in every scenario they come across, it will be appealing. Even less-realistic implementations like the fast hijacks in Saints Row the Third are a great triumph of using what you can do as a developer to make what a player must do in your game more dynamic and engaging.

Now in this generation we have examples of towns like the monstrous main city of White Knight Chronicles. That town dwarfs Rabanastre in size, with more than a hundred NPCs and multiple large districts. It is also a complete chore to travel through. Despite having the same services as an older RPG town, it takes at least five times as long to perform the same errands.

To be honest, I think games like Skyrim and Xenoblade already push or surpass the limits of practical sized towns in an RPG. I have never seen a game benefit from making towns any larger. As a whole, large towns are a completely superficial element akin to pretty graphics. While a large town might wow at first impression, that grandeur quickly outlives its welcome after you have to run through the giant mess thirty times.
That one for me is more loosely related to the topic of gameplay limitation being at odds with the realism of the presentation. Are they not portrayed as "great cities" in the game? You're fighting a civil war, overtaking thrones and such, but they are dinky little villages. Both the narrative and the presentation would lead me to expect at least something like the cities in Assassins Creed 2. I know that is especially difficult since all the buildings in Skyrim are homes that can be entered, but in AC2 the scale and population presented is equivalent to the type of figures and deeds portrayed. It's not even close with Skyrim. You're basically on the level of a 3D modeled version of a PS1 RPG, not even giving the sense of scope of Infinity Engine games.
 

ASIS

Member
I'm sorry I haven't gone through every post yet, I'll read them in another time, but in response to the OPs last question: Isn't that what the 8th generation is about though? I thought the new engines focused more on the physics and stuff rather than making the games look prettier no?

BTW excellent thread.
 
Wounds: Does it bother you that we're still often just attacking enemies that act completely unharmed until they simply keel over?
Would be too graphic/ would get banned from stores in some countries
Destructible environments: This seems to be getting more attention, but it's still incredibly limited.
Possible, but engine technology and performance lagged behind for a while. Should get better next gen.
Town scale: There seem to be valid attempts to move this forward, but it's fairly slow.
Taxing on CPU on Memory, which current gen lacks. Can be alleviated with better engines and more power next gen.
AI: we could have really dynamic personalities and work out a new kind of game entirely
AI isn't limited by hardware, but quality AI is hard to program and developers prefer to spend their development time elsewhere instead. Not sure what can be done.
The problem: As the title of the thread implies, I think the main problem with these is the huge investment of graphics development. We are constantly trying to push that bar forward, and it's incredibly expensive to do so. Anything you expand would have to be visually rendered at a high quality.
Graphics sell
 

QaaQer

Member
I'm sorry I haven't gone through every post yet, I'll read them in another time, but in response to the OPs last question: Isn't that what the 8th generation is about though? I thought the new engines focused more on the physics and stuff rather than making the games look prettier no?

BTW excellent thread.

Thats what the beefier 8 core cpus are for. But they will look prettier too because the gpus are also much better, allegedly.
 

deviljho

Member
I don't think it's merely cosmetic when a Mammoth in Skyrim doesn't act like I've been slashing at it's leg for the last 3 minutes before suddenly dying. Especially if I am trying to be more nimble than it as a strategy.

You are 100% correct. It isn't that hard to do either. And you don't have to have the same rules apply for enemies and the player. Ever played Monster Hunter? Monsters have different hitboxes for body parts that you can break or cut off.

What you are asking for though is going against the trend of AAA game design, unfortunately.
 
Top Bottom