what's the next leap for video games?

None

Refining what we already have, shit like your webm serves no purpose in a videogame

I think that muh physics is the biggest thing that could improve gameplay and game design

The problem is that there's no real way for video games to improve at this point. The whole industry is just stagnating or pining for the past.

>>386943103
actually good gamedesign maybe?

Global illumination. Lighting and shadows still look like crap even in the best-looking games.

File: tear.webm (1MB, 400x417px) Image search: [Google]

1MB, 400x417px

>>386943235
it would allow for fluid physics puzzles and neat shit like that

also if it was like lava or something it could be an interesting dynamic obstacle, there's tons of shit you could do with actual fluid simulations

Moving away from making videogames solely for profit

We need better fucking AI in everything.

>>386943103
>What's the next leap for video games
good and fun games

>>386943352
They're insanely expensive though, you can either have a game or you can have a webm like the OP. Unless there's a major hardware leap simulating actual water to that degree will be prohibitive.

>>386943352
Physics puzzles are only good if the physics work the same way every time. having the water randomly move about would only frustrate the player.

These are:
https://www.youtube.com/watch?v=x19sIltR0qU
https://www.youtube.com/watch?v=bXouFfqSfxg

And how will this help making better videogames exactly?

>>386943338
shit like VXGI is probably the next step towards that, that shit takes like a 30-50% hit to your frame rate though

>>386943623
New WaveRace game

>>386943103
I don't see the purpose of crap like that in videogames, unless you really like crappy physics puzzles that are all the same at the end of the day.

Muh detailed physics are just as useless as muh hyper realistic pebble/grass graphics

>>386943561
Computer are predeterminist anon. Unless there a RNG or some uninitialised values it should always do the same thing unless there is player intervention.

I would love for more accurate cloth, fog, smoke and particle physics. That stuff always blows me away.

>>386943103
None because of consoles

File: 11122222222.webm (3MB, 570x536px) Image search: [Google]

3MB, 570x536px

>>386943549
yeah that post is about 100k fluid particles and it uses around 70% of my GTX 1070 to reach 60fps, it's not ready for games
>>386943561
you're definitely wrong, I can think of bunch of robust physics puzzles involving water, it's just a matter of how you design your puzzles, I think that having realistic fluids would serve more purpose than just some puzzles though

File: 1474271523558.gif (536KB, 381x512px) Image search: [Google]

536KB, 381x512px

>>386943819
>game runs at 60FPS
>clothing animations run at 30FPS

For fuck sake, Ubisoft.

>>386943814
he obviously means that fluids appear way too chaotic to our monkey brains, he's completely wrong though

>>386943814
Then there's no point in physics that sophisticated, is there? Waste of time

>>386944032
sure there's plenty of advantages in having fluids that maintain proper volume always

>>386943103
we need better harware,
look at quake 2 running with global illumination, it's grainy as fuck on a titan xp.

>>386944032
>unless there is player intervention
You know, the whole point of video games.

Bring back and improve GTA 4's Euphoria (ragdoll animation engine).
That shit made enemies feel REAL, and I'm absolutely baffled the amount of games that use it can be counted on two hands.

>>386944132
more like on one hand

that shit is apparently a neural network powered movement system, that seems pretty cool

>2004 game
>objects move about when explosions happen or are moved by the player
>2017 game
>everything is static

>>386944121
Name one in a video game context that couldn't be done/isn't already done in a more simple way.

>>386944108
https://www.youtube.com/watch?v=x19sIltR0qU

>>386944214
because you want to have your precalculated CGI lighting in modern games, so nothing can move around

>>386943945
>I can think of bunch of robust physics puzzles involving water
List a few, then. And remember that they need to require actual fluid simulation to count, anything involving weight/volume/buoyancy/etc. can all be fudged using much simpler methods.

>>386943945

That technology could be huge in H-games.

>>386943352
This needs to be wrapped around a girl.

>>386944214
b-b-but they look so much more realistic and cinematic! who cares about interaction, I want to play movies

>>386943103
> vore simulation

>>386944182
Definitely two hands:
>Backbreaker, Backbreaker Vengeance, GTA 4/5, RDR, Star Wars TFU & TFU2, Max Payne 3, that one mobile game made by NaturalMotion themselves

Also I think Euphoria has been merged with Morpheme, the other project of NaturalMotion, though I don't know what Morpheme does.
https://youtu.be/0Pm0Cvm0zdI

>>386943814
>what is sensitive dependence on initial conditions

>>386943945
>hotglue simulator 2018

>>386944314
>>386944421
I like the way you think.

>>386944340
it needs to be wrapped around a girl's butt

>>386944307
well the first and most basic idea I got was creating a physical path for the fluid to flow through to reach some destination

another one would be a puzzle involving weight distribution

then basically anything not scripted

>>386943103
Dying in the ass chasing whales with multiplayer only loot box crap and/or splitting limited good devs between being indie, VR and mobile.

>>386944521
Both of your ideas do not require realistic fluid simulation to function.

Would be nice if augmented reality would catch on. Having entire raid parties on my table in actual 3d, maybe being able to interact with my character via motion/gesticulation/drag and drop would be sick. Having entire maps span around my room in a random generated map akin to diablo endless potential to fuck around with for 1/10 the costs of a virtual reality goggles

File: 1497675119426.png (56KB, 209x248px) Image search: [Google]

56KB, 209x248px

>>386944276
>>386944374
>looks more realistic
who cares about that??? i WANT A VIDEOGAMEY GAME WITH FUN AND INTERACTION!
like timesplitters, JSR or XIII, even SMODs

>>386944593
yes they do unless you create some one way solution scripted puzzle

>>386943508
This absolutely

>multiple people in this thread calling better physics useless because it serves no purpose

What the fuck is wrong with people nowadays?
Why does everything need a purpose and have to directly improve gameplay in a big way?
Better physics should be something the gaming industry should aim for no matter what because it improves the game as a whole.
Just compare Bioshock to Infinite, 6 years between both and Infinite feels worse than Bioshock because physics are nearly nonexistent since that's apparently something modern games don't need because they got scripted sequences and particle effects to fill that hole.

>>386943235
>shit like your webm serves no purpose in a videogame
>never seen nurin
>never seen cellfactor revolution
>never seen bet on soldier

Too ahead for past and current time and your feeble mind.

>>386944307
Anything where you'd have to section of a body of water into smaller bodies, that shit would turn out garbage without defined and consistent volumes for the fluid

>>386944650
Play From Dust. A simplistic simulation is good enough for gamey applications.

>>386943103
nvidya flux is fun

>>386945049
>good enough
flat shaded models get the job done but you can do way way better

I'm honestly not going to play that game though because it looks pretty bad

>>386944650
The path one would only be true if you were required to use unusual/unconventional objects to create a path, which if you thought about it for two seconds, would be much more impractical and frustrating than it would be fun. Sure it might be entertaining to come up with some insane rube-goldbergian contraption to solve the puzzle, but it'd only be fun for like five seconds and only the most hardcore autists would find entertainment in it in the first place. The only way I see it working is if you make an extreme niche game that's almost entirely comprised of complicated physics-based puzzles.

As for the weight one, I fail to see how any implementation of that would require any kind of complex simulation at all.

File: 11221.webm (2MB, 695x492px) Image search: [Google]

2MB, 695x492px

>>386945313
the path thing was the most obvious example

the other one is about having fluid which has varied density, maybe it's sectioned to parts or something

The problem is that the current frontiers of vidya, AI and physics, are both incredibly hard on the hardware, meaning you can only have one or the other unless quantum physics decides to stop being a dick and allow Moore's law to continue. This is not helped that scripting engines, the mainstay of modern games, also take up considerable power because compiling code is also hard on the hardware.

So you either have amazing physics, amazing AI, or easy moddability and assets.

Also I'm gonna go shill BeamNG here, cause the devs are literally using algorithms more advanced than those in the latest resaerch papers just to get their physics to run at a decent speed, and I think that's cool

>>386943508
This.
AIs still behave like total retards in pretty much all games

>>386943508
This, specifically, player's interaction with the AI.

Also we need more open ended game design, especially in RPGs, im tired of 'open world' meme games that are anything but open when you're still railroaded through 2-3 generic story paths, rather then the classic design of having a singular goal and a billion dynamic ways to achieve said goal.

The AAA model falling out of favor with the masses.
Development time and budget shrinking dramatically.
Rise of AA studios.
Games costing less
4K 60+fps for $200 or less

Gyro controls becoming a standard option in games
Rear buttons and sticks becoming standard for controllers

>>386945514
It's clear that you haven't given this very much serious thought and are just being swayed by impressive-looking tech demos. I know it hasn't occurred to you, but those tech demos's appeal is literally the same as those "le cinematic grafux whore's wet dream" games you hate so much. They're designed to appeal on an emotional level rather than a logical one.

Your ire is misplaced. There is a correlation between better graphics and simpler gameplay, but not a causation. Games are simpler because executives(and developers themselves, many times) WANT them to be simpler, not because of limited resources. Complicated puzzles have existed since the days of point-and-click adventure titles where nothing was actually being physically simulated at all. Consider, for a moment, that before the advent of modern supercomputers, all design problems, from plumbing, to cars, to factory equipment, to missiles, were solved without ANY kind of simulation, only very basic equations and models. I myself find it a little insulting that so many self-proclaimed "nerds" seem to think that brute force methods like particle simulation are in any way advanced or aspirational.

>>386943352
as neat as those physics are, I don't think they will ever look realistic since in real life doesn't use frames like a computer does, so even at a smooth 60+fps it wouldn't look natural, but hey there nothing wrong with trying

File: peon.jpg (21KB, 154x158px) Image search: [Google]

21KB, 154x158px

>>386943103
There are so many games out there that utilize water physics not just for show but as a part of a gameplay mechanics.
Truly amazing

Stop spending the majority of time and money and graphics that look shit anyway

>>386946572
it's not only realism but hanging on a drapery which could tear because someone is shooting on it or something would be pretty cool
>>386946421
I don't know what you are on about

File: 100-5801-BLK-F01.png (3MB, 2000x2000px) Image search: [Google]

3MB, 2000x2000px

>>386946421
Where did all of these accusations come from?

>>386943814
when you get into complex liquid physics, a player model or object in the 'wrong' place can fuck everything up

>>386943103
Devs actually caring about the games they make.

>>386947043
>hanging on a drapery which could tear because someone is shooting on it or something would be pretty cool
This still doesn't require the kind of simulation displayed in the various webms in the thread and could very easily be scripted.

>I don't know what you are on about
If you're going to complain about something, at least have the courtesy to have a basic understanding about what it is you're complaining about first.

>>386943578
I just tried that quake path tracing demo on my gtx 1070 and holy shit that shit is grainy, even when I increase the sample amount to the point where I get 1fps

we need like 1000x more powerful GPUs for that shit

Hair physics.

https://www.youtube.com/watch?v=HvHq4JIcneY

>>386947645
why would anyone want scripted gameplay over flexible gameplay

the drapery would react to any force accordingly if it was simulated vs if it was scripted you have to script a thousand scenarios to make it appear reactive enough to fool anyone

>tfw no 2017 version of powder toy

>>386947645
what am I complaining about? seriously what the fuck are you on about? all I am saying is that shit like proper fluid simulations would allow for interesting gameplay, puzzles being an example to what fluid simulations could be used for

File: Laura_46.webm (2MB, 1280x720px) Image search: [Google]

2MB, 1280x720px

>>386947712
>tfw you will never jerk off using lara's silk-like tress-fx enabled hair

>>386943945
Oh boy, Flubber technology, so amazing.

>>386943945
It looks like the fingers don't get changed at all

File: 3331111.webm (863KB, 410x489px) Image search: [Google]

863KB, 410x489px

>>386948338
bukkake bunny

>>386947790
>you have to script a thousand scenarios
Name three meaningfully distinct scenarios.

I myself can only think of two: either the cloth is damaged to the point where it's no longer climbable, i.e. you fall, or it's damaged to the point where you have to do some basic climbing/"platforming" to get past.

Whether it's shot, burned, sliced, rotted, etc. doesn't matter, it's functionally the same from a gameplay/mechanical perspective. The only thing that would change would be the visual assets and the gameplay trigger, i.e. in the case of shooting, it'd only get torn/damaged if, say, you were doing a stealth section and tried to climb across just as a guard below was looking up.

If the asset were made fully physically interactive, like in the webms, then it'd just be frustrating for the player. In some cases, the guard might be a bad shot and damage the drapery by accident. In others, he might intentionally damage the drapery to make the player fall. And regardless of his intent, he will randomly hit the player, the drapery, or perhaps even the wall behind it depending on aim, the type of gun he's using, etc.. Every time you play through the scenario would be different. If this sounds "good" to you, consider that this randomness would make it impossible to formulate a good strategy. Whether or not you got through such a scenario would largely be dependent on luck, or you'd otherwise be forced to choose a plan of action so conservative that it'd render the game boring.

The real world is full of situations like this, where only one or two out of the infinite possibilities actually coalesces into a meaningfully distinct outcome. It's why we invent stories and games in the first place. It's counterproductive to try and make something which by definition is not supposed to be realistic, more realistic.

>>386943352
>it would allow for fluid physics puzzles and neat shit like that

Which almost never matters or results in anything substantial.

>>386943561
The chaotic nature of particle physics based stuff means that it's almost always something cools that looks great in tech demos but there's rarely reason to actually think it would enhance the gameplay of anything significantly.

>>386948639
you're not efficient at typing it's fucking annoying to read ten lines of text you could have written in two, then all these epic big words just make you come off as the average autismo

maybe you'd want to slice a part open from some fabric to see through, maybe you'd want to burn it, maybe tear it, you want some scripted breaks the same way everytime shit

>>386948771
wave race 2017 with insane water physics would be cool as fuck

>>386948771
>Which almost never matters or results in anything substantial.
a non-blocky minecraft with fluid fluids would be so much better though

File: adachi.jpg (25KB, 227x249px) Image search: [Google]

25KB, 227x249px

>>386948969
Not really, it would be just wave race with a much bigger price tag due to the hardware involved.
>wave race in the 90's
>i'm racing on waves!
>wave race in 2020
>I'm racing on waves but I paid $2000 for my rig so that the water looks cool
Fucking. Useless.

>>386949069
pessimist

it would be cool as fuck as a game, have waves that actually splash and shit and water reacting realistically and shit

>>386945248
>you can do way way better
In what sense way better? More accurate physics means giving the player more freedom to fudge things.

>>386943561

well now i know you've never watched the crystal maze. water puzzles are awesome

>>386949179
More accurate physics means giving the player more freedom to fudge things.

>>386943103
Consoles with Ryzen or better CPU's.

>>386949131
It would be the exact same thing as the original game, but with bells and whistles attached. Do you get excited when people jingle their keys in front of your face as well?

Go work as a 3D artist if you care about that shit, like in the automotive sector. They're miles ahead.

>>386949252
well if the water physics were realistic it would play differently

what games do you play? all games are copies of existing games

>>386949252
>the video in a video game is irrelevant

>>386949327
I play a bit of everything, recently got into spacechem.

Machine Learning based AI.

>>386943508
This a billion time, I'm still waiting for something more impressive than f.e.a.r back in 2005 and it's not coming.

>>386944593

that's a very high bar to set to argue for realistic physics. you could script any puzzle to fool the player into thinking that the physics involved are realistic. realistic physics would make for a much wider range of experience in attempting/failing to solve the puzzles though.

some sort of machine learning neural network ai shit would be cool

>>386948995
This is one of the cases where I could actually see it working if done right.

>>386948909
>epic big words
>meaningfully
>randomness
>counterproductive
>conservative
>epic big words

>maybe you'd want to slice a part open from some fabric to see through, maybe you'd want to burn it, maybe tear it
All of which can(and should) be scripted. Your "criticisms," if they can be called that, have more to do with design than technology. Everything you've described in this entire thread could easily be achieved using current engines and current hardware. As I've said, games are DELIBERATELY designed in a simplistic fashion these days, it has nothing to do with consoles or technology or SJWs or any of the tired old bullshit this cesspool is constantly spewing.

File: 34324618[1].jpg (2MB, 2560x1920px) Image search: [Google]

2MB, 2560x1920px

Speaking of water, what are some games where you flood/drain areas a lot to manipulate the environment? There's just something about an area becoming completely different simply by filling it with water that hits the spot for me.

>>386946421

>it's clear that you haven't given this very much serious thought because you have different preferences and priorities to my own

>I myself find it a little insulting that so many self-proclaimed "nerds" seem to think that brute force methods like particle simulation are in any way advanced or aspirational.

no you don't you pompous ass. having the ability to incorporate physics into gameplay is clearly aspirational for the industry as a whole. individual developers/designers are under no obligation to give a shit about it, but it still increases the available range of options to developers to a point where their imagination is entirely unconstrained by the need to create simulatory fictions.

>>386943103
Imagine playing a game where combat occurred in the open water with those physics. It would be fun as hell.

>>386949639
I don't know what you are talking about but if you are saying that an interactive fluid simulation can be scripted, you are talking about some shit tier garage looking minecraft water because fluid simulations have unlimited states

why does vidya oceans look like shit if they can be scripted to look real and behave realistically?

>>386943103
60 FPS
>consoles

>>386949439
openAI has a functional 1v1 bot for DOTA right now.

Deepmind's started training their starcraft 2 bot:
https://deepmind.com/blog/deepmind-and-blizzard-open-starcraft-ii-ai-research-environment/

should start becoming relatively common in 5 years.

>>386949739
Ocarina of Time and Twilight Princess.

File: YES.gif (4MB, 663x398px) Image search: [Google]

4MB, 663x398px

>>386944340
holy fuck, this

>>386949819

you can script them by limiting the range of ways the player interacts with the water.

File: 1497547214540.jpg (133KB, 1024x719px) Image search: [Google]

133KB, 1024x719px

>>386948481

NO!

>>386949887
well there you have it, by scripting you compromise

>>386943103
Datatization of our bodies so we can physically be in the game

>>386949639
you forgot coalesces, just a bunch of unnecessary words shoehorned to make a ten line post to make it seem like it has more substance

>>386945649
Yeah surprised bng isn't memed on here more often.

Shame multiplayer is impossible

>>386949795
Like you are a pirate and after the ship crashes there is floating planks and debris everywhere and you have to maneuver as enemies come at you. Also there is a shark in the water.

>>386950203
>coalesces
>big and complicated
How fucking underage are you? Alright smart guy, how about you summarize in your own words for me then? Show me how a real man speaks.

>>386950520
needlessly complicated as you wrote a bunch of ten line posts with these snowflake terms when you could've just written everything in two lines, I'm repeating myself here

>>386943352
isn't that the psyhsx demo from like 2008?

>>386943103

More shiny stuff with the same crappy FPS

>>386950859
2012 or something, it's the flex demo

that cloth simulation stuff is easy to do on the GPU now though

>>386950060

yes, and i'm not the guy who's bashing the idea of physics in games, but i do agree with one of the things he's getting at - the entire history of videogames in general is about compromise. games aren't necessarily more fun if they more realistically represent real world actions. they're just like board games or whatever in the sense that they have a constrained set of rules you have to learn.

where i diverge from him is that i think the real physical world is exactly the same in that regard and so it can be turned into a game as well.

>>386951116
if scripted, physics stuff is pretty much always scripted for performance, same goes for how audio echoes in spaces, scripted for performance

lots of shit is compromise due to need, sure a lot of stuff is scripted because there's no benefit to doing it in a more computationally expensive way

>>386943103
The next big leap for games is going to come with WASM, and the web version of Vulkan.
Developers will be able to deliver high performance modern games to any device with a compliant modern browser bypassing any stores or restrictions placed on devices by their makers.
The only way companies would be able to block games from their platforms at that point would be to not implement modern browser features which would quickly leave them behind.

File: 1499015523343.jpg (28KB, 663x485px) Image search: [Google]

28KB, 663x485px

Perfect water simulation would be as much of a leap for video games as the Source Engine physics. A fun but gimmicky addition that gets old very quickly.
We do not have enough fine motor control to actually benefit from such a feature.
Video games are abstraction. That's why when you walk forward when you press W, and don't need to actually perform the weight shifting necessary for walking.
There's a balance to be found, between abstraction and fidelity, for it to be fun and make sense to our instinct.

using more than 2 cores for starters
making AI smarter
Eyetracking. VR. horrogames.
better text-to-speech to dynamically generate dialogue instead of shipping with 15GB audiofiles

>>386952117
most 2016 and newer games use 8 cores fine

it would be cool to have a robust siri type ai in a game, so the npcs can talk about things more naturally

>>386952117
>better text-to-speech
AEIOU

>>386943352
didn't Mirror's Edge already implement this?

>>386951774
Every game should use qwop walking mechanics.

>>386952117
>making AI smarter
That's more a matter of having the right algorithms/code than computing power.

>>386951774
I don't know man, source engine physics took a while to get boring for me. I mean I was barely a teen when they came out though.

People have gotten some games to run at 16k and still be playable. Those games were Half Life 2, Minecraft and Civ. But still, we are getting pretty high up there.

>>386943103
it's mostly in the area of AI (both for individuals/combat-scenarios and for the in-game world's politics/economics/etc.).

Terrain destruction is another thing I'd like to see more of, especially in multiplayer FPS games, perhaps it requires a change in tools to make it easier for smaller developers to make terrain destructible/assign the necessary animations models, and other things to in-game objects, as well as netcode that can handle terrain destruction and larger numbers of players.

>>386952117
AI dialogue generation would make open world shit infinitely more interesting.

>>386952286
JOHN MADDEN

>>386943103
>what's the next leap for video games?
actually being good, fun and enjoyable, while not pandering to fanbase of lowest common denominator in their respective genre

Lighting and sound are arguably more important than physics and graphics to be honest senpai.

off a bridge

>>386943240
This
It will never be ai because that would make games too hard
Physics improve gameplay

>>386943103
Having good gameplay

>>386943103
Fully rendered, dynamic rope

>>386953459
How is lighting not graphics?

>>386953870
I don't know, you tell me

>>386953740
>It will never be ai because that would make games too hard
You are dumb. That's not what improving AI does.

need infintley better AI
need AI/bots in multiplayer
need an AI vs humans mode to teach the AI
need more of physics

>>386949739
Mega Man Legends 2.
More specifically, the Nino Island Ruins in it.

VR if they ever get their shit together

>>386952467
Borderlands 2 has some cloth you can shoot if you have physx.

>>386943103

Games that aren't afraid to openly reference things in real life

Like give me a game in an Islamic war state

>>386955784
Isn't there an entire genre of bro shooters dedicated to stuff like this? Ones that sell more than any other games, and have like, 3 new entries a year?