Why Aren't Real-World Physics Equations Used In Video Games?

[@zoolookers]

But you don't need to run at a molecular level, many physics equations are based on "bodies" meaning you can use the same equations for a lot of things

[@noxabellus]

Yeah and we do use those equations in games...I dont really get where this video is coming from...Does _anybody_ believe that we simulate that crap?

[@ChrisD__]

noxabellus I think what he meant to talk about is how game developers sort of loosely base physics of certain things on real equations, but don't actually use real equations. The reason being is that game's often feel and play better when the physics are tweaked for the game mechanics.

[@Nikandros]

The video is coming from an idiot who thought they were informed enough to make a video on this very techincal topic.

[@Yizak]

+Chris D Yes it's not a general rule though. A lot of physics equations that model real situations are used just the same way in games as they would be in real world applications

[@Molochors]

Yeah, most equations you'd really care about is classical mechanics, and that can be broken down to forces and movements in 3 directions acting on a center of mass of an object. We already have graphical objects that move with thousands of graphical entities (if not more), creating an object divided in a dozen parts connected to each other and then calculating it's movement based on it's center of mass isn't that hard. It's like 3x3 equations (F=ma, v=v0+1/2at, x=x0+vt for each direction) requiring few properties as input (force broken down to 3 dimensions, mass, position in 3 dimensions, current speed in 3 dimensions - most of which you already have anyway in order for game to work)

[@ez4pz]

I thought you were going to talk about the exact numbers of gravity not being used or something. Not that you Can't run a universe in real time on a mainstream computer.

[@IkBenBenG]

The main reason for that it often just isn't fun to have the same limitations in a game as we have in the real world. In games we want to be able to jump higher, move faster and be stronger, which doesn't work well with the real physics models. If you'd be able to jump 3m in a platformer with realistic gravity then the physics would feel slow, floaty and hard to control. Those games then often add some in-air movement to improve controls and make gravity a lot stronger to make it less floaty and faster. In other games you may actually want to have those more floaty physics, since it allows you to do more elaborate stunts in the air. Another important reason is reliability. If you simulate someting then you give the player influence over the outcome. When you want to design a game where the player has to destroy a wall to move to the next level, then that sequence will often be scripted rather than simulated since with the simulated version the player may destroy the wall improperly leaving a still uncrossible pile of rubble.

[@OmniscientlyMe]

They did sort of explain that, too. Gravity doesn't work so well when nothing has any mass; remember everything you see is just a hollow shell of surface images.

[@IkBenBenG]

Game engines have a simple workaround for that. Even if it's an empty shell, its volume and the center of gravity of that volume can still be calculated. You can then assign a density to the volume and use the resulting mass in your physics calculations. Less advanced engines just let the developper specify the mass and center of gravity of an object. When those two parameters are specified, you can use the standard physics formules. Most games do it this way, though they often use different constants for stuff like gravity to make the physics fit the gameplay.

[@OmniscientlyMe]

And the counterpoint there is how much money can developers afford to spend assigning physical properties to the thousands of unique objects? Also, one has to consider if there's enough processing power available to do keep up with calculating for every unique object versus the simplicity of shared properties.

[@juckey6748]

"how much money can developers afford to spend assigning physical properties to the thousands of unique objects?" you're overestimating the work needed for this "one has to consider if there's enough processing power available to do keep up with calculating for every unique object versus the simplicity of shared properties" this is why developers have to optimize their games, and frankly, it's not that expensive either. See HL2 E3 demo footage for example; all the props have physical properties, and so does every physically simulated model in source games. and yes, it was running on a 2003-era PC.

[@hghg46204]

Because the universe copyrighted it

[@ReddoX30]

Albert Einstein and other guys tried to copyright some Physics properties, but the universe killed them.

[@wukennylee4926]

lmao exactly

[@helderdyela4822]

perfect!!!

[@xyz8206]

Think if we can copyright it. It can mean better understanding of creation. The ultimate formation. The everything. Although the won't be made up of same material the will be only computer generated, but we can watch what is happening.

[@theprophet_-__cxvii__-_8693]

Mousum Gogoi The universe was never created. Not by a god, not by a Big Bang. It’s an infinite existence. The one of few things our mind just cant comprehend. The word infinity. Gods exist, but not they way we think they do, or in some cases, think they don’t exist at all. The Big Bang isn’t true because its mathematically impossible to have occurred. And even if you say “well it happened, it’s a scientific miracle.” Than explain how after the Big Bang, other mathematical impossibilities occurred. For example DNA. Each strand is using a mathematical formula that only is possible from intelligent design. Unless you want to argue it was another mathematically impossibility for each strand of DNA. But at that point your preaching to something that doesn’t even exist. So yes, our universe is not based on evolution or the Big Bang. Because in order for that to be the case, multiple mathematically impossible events would have had to occur. Which is obviously mathematically impossible in and of itself. By the way, that’s fact. Science admits it openly, only they say its the only miracle ever to exist.

[@LoxagosSnake]

Nice video for mainstream gamers who want to gain a little insight into the games they play, but you oversimplified a couple of things and were wrong in others. As a hobbyist dev and physics student, who's taken a couple of computational physics classes, I can tell you that we *do* use exact physics formulae in some aspects of gaming, and statistical/approximating mathematical approaches in others. For example, Newtonian interactions not comparable to the speed of light (a.k.a. no need for relativity) are solved using finite element methods. Remember that v = dx/dt from calculus? Just take very small timesteps and the difference between coordinates and voila! you have just calculated the instantaneous velocity. In fact, this is the difference between analytical and numerical techniques: computers don't understand equations as we do, but they can produce ever so small deltas in fractions of a second that would take us weeks to calculate a simple in-game movement step by step, by hand. When it comes to more complex phenomena like optics or fluid mechanics, a statistical or macroscopic approach often works way too realistically. Lighting techniques use sampling, a solution which is scalable as you can easily tune the accuracy up by sacrificing as much performance as you can allow. And if you crack open a GLSL shader file, you'll see that waves dead-on include sines and cosines. All in all, we *are* able to take some equations, unwrap them into numerical methods, translate them in a way the computer understands them and use them in games. When we lack this ability, we rely on robust statistical models. And if push comes to shove with some theoretically feasible but practically unattainable physics, we just us the plain ol' 'smoke and mirrors' approach.

[@JumpingMonkey]

This is so much more informative than the video

[@childfs6865]

Thank you for the text! I wish I'll get to work with computational physics and game dev some day in the future.

[@whanowa]

The art of saying NOTHING in 7 minutes.

[@RennieAsh]

whanowa that ain't working! that's the way you do it - money for nothing and your chicks for free

[@RennieAsh]

pro hacker I should have learned, to play the guitar, I should have learned how to play them drums

[@mikeyfn-a6684]

Rennie Ash im 36 and can tell you(as I've played in a couple bands..even wrote a song that started a pit)..you still can 👍

[@Rufal]

you're just an idiot who can't understand what he's saying lol

[@ezpz4646]

whanowa Lol

[@WwZa7]

Acurate physics looks cool, and gives immersion. That said, no one needs physics on a cellular level. Give us ability to slice stuff with our badass sword up to 20 pieces, tear clothe objects apart, brake wall to pieces, and that's all we need. But than again, it only adds immersion. Even physics based games like Kerbal Space Program bend reality to make things more fun. As Falcon said - FUN is the most important factor. Anything else is just immersion.

[@Putzl52]

Furebel yeah I agree. Real work calculations are not done to such a degree. There’s a lot of assumptions to make systems ideal to calculate

[@skuzzyj]

Furebel The thing about the sword reminds me of the original Bloodrayne game. Nazis made of sausage. You could cut objects (meaning chunks of dead nazi) in half as many times as you could line up the blades. It was awesome. The sequel added significantly more detail to the slices but lost the magic of making actual mincemeat out of NPCs.

[@WwZa7]

@SkuzzyJ To be fair, I was thinking about Metal Gear Rising. The realistic cutting physics was great there without need of quantum calculations. Gaming and 3D CG in general was always about faking realism

[@skuzzyj]

Furebel I've never really played any of the Metal Gear games aside a couple of demos back in the day, so I didn't know Rising had that function. Neat.

[@WwZa7]

Watch some gameplays or trailer at least. You really can just go into this ninja mode, where time slows down, and you swing mouse or analog stick to cut stuff as you wish. Some gameplay mechanics reward you for clean cuts, like cutting out cyborg arm with valuable data in it, or slicing balistic shield between explosives. After beating level bosses you also have this moment, where you can go nuts and freerly slice them into really tiny pieces like a madman :D

[@FearlessLeader2001]

It depends on the game. No, I don’t want realistic physics in Sonic and Dead or Alive. But I want realer physics in GTA, Watch_Dogs, and even Fallout. I’d love to have the most immersion possible in games like that, and TBH the ways the former two don’t punish bad driving pisses be off relentlessly. It’s too easy.

[@Dhalin]

Yeah, but a lot of people (including me) enjoy being able to drive recklessly. It's fun. I mean, take a game like Saints Row 3/4 for example. Being able to smack into 50 cars before your car finally blows up allows you to do all the things you wish you could do IRL. "GET OUT OF MY WAY!" **WHAM** etc. Can't do that IRL because the slightest tap disables your car entirely. That's no fun. I even had the same kind of fun in LA Noire, where you could fly down the streets of LA doing 50 MPH+ and the occasional fender-bender wouldn't suddenly kill you or cause your car to stop dead in its tracks. I don't think people like you who spout this "MAH IMMERSHIUN!" thing understand what you're really asking for, and how bad it would screw the games up if they had it. But I suppose, they could put an OPTION in these games to turn on more realistic stuff just so you could see how bad it breaks fun in gameplay. Or are you one of those people who would be against such options because you feel everybody should be forced to play (and enjoy!) the game the same way you do? I've met plenty of people like that too. "You Just Ain't Enjoying The Game If You Don't Have It Set On Hard+!" **sigh**

[@Dhalin]

Well... to be honest, I'm very... how shall I say it. "Eh" about VR. Or maybe I should say "Meh". I have plenty of fun sitting at the keyboard, looking at a monitor. There's a point there, where you just have to admit that a game is a game and it is not meant to replace or even mimic IRL. In fact, the whole point of a game is to get **away** from IRL. You don't get away from IRL by trying to make yourself believe you're in another IRL, lol. That's just..... stupid? IMO. I mean if you wanna spend that much money so you can try to make your pretend world look like the real world more... uh... whatever. I just don't see the point in it.

[@FearlessLeader2001]

I've logged 140 hours in BeamNG drive

[@PhilFernandes84]

Sukki Blue the driving part is related to how alert the AI of the game is configured.. u do not want to be chased by cops everytime u hit the curve of another car or run over pedestrians.. ??

[@FearlessLeader2001]

Every time I hit a ped? yes. What the fuck is the "curve of another car?" If you're talking about getting police every time I total another car, then yes. It'll give me incentive to become a better driver. Wouldn't you rather have only pristene Zentornos and T20s racing around, rather than the completely stripped, scratched up messes that are driving around now?

[@DrWub]

The more unrealistic the physics the more fun the game is.

[@cent8526]

edwin serrano Agreed

[@manneh7382]

Agreed

[@communityEsc]

Couldn't agree more. Video games that take you away from reality, in my opinion, are better than those that try to replicate it.

[@alittlebitintellectual7361]

Just cause....

[@spacehorse9581]

Debatable

[@ComradeTiki]

Don't waste your time. Here's the video in 12 words: "Particles are CPU-expensive. I think arcade physics are funner than simulations."

[@nieildilsonsouza4747]

it's all about fun. and realism is not fun because we live with it already. problem is people saying it is about realism, they are only a minority trying to rule over a majority.

[@bisterkding9249]

There are different categories of games, and within those are fanbases with their own respective majority and minority opinions. For example, you clearly disagree with the majority of racing sim fans. You can't just apply a blanket statement like that to something as stratified as the gaming industry without sounding like an idiot.

[@REALdavidmiscarriage]

arcade is funner than simulations? that is just bs. only a person with zero physics understanding could say something like that. cause you can't relate to that

[@FernieCanto]

"I think arcade physics are funner than simulations." More specifically, he says *everyone* finds arcade physics more fun, especially when he mentioned Sonic.

[@REALdavidmiscarriage]

fernie Canto I don't. I lvoe rl physics, thats why i study physics. just most people can't tell the difference because they are uneducated. most people prefer realisticv physics. its just that mose peopl think battlefield is realistic. while its actually not.

[@bigboss9544]

Imagine playing any RPG and having that 3 meter long two handed sword attached to your back with real physics

[@KrateFocksKratesCrateofGaming]

we wouldn't have any more fat kids that's for sure

[@trollconfiavel]

Imagine playing Dark Souls and not having a body sliding on the floor like it is ice when you are just trying to walk through a corridor

[@KapitanPazur1]

Sounds boring as fuck

[@nullpoint3346]

The three inch thick one or the three quarters of an inch thick one?

[@Luxalpa]

@@rafaelvillalobos9145 Still it wouldn't work because the Center of mass would be outside of your character and you would fall over. So either your character must be extremely heavy, or you have to carry the sword in a different way; most likely both.

[@ManUMinute]

Gameranx quickly becoming my fav channel

[@afiefabrahams4801]

Great vid guys like always your my favourite channel to watch and the topic of this vid is just so amazing and interesting anyways keep up the good vids

[@vullord666]

It wasn’t already?

[@albertwillemse5259]

TanTan I prefer their videos by Jake Baldino but yeah a good channel

[@AngryBulldogGaming]

What? What other channel is better? I’m speechless

[@BLOODY__FATALITY]

Lol your pfp looks like Etika as a mii.

[@ArcadianGenesis]

Depends on the genre. Some types of games *should* be as realistic as possible, but others shouldn't.

[@manuelsaavedra8081]

ArcadianGenesis imo, I believe a super realistic fantasy game with magic tacked on the physics engine would be sick, Dr. Strange levels of sick if you catch my drift

[@trager8933]

ArcadianGenesis And people say bf is realistic ha !

[@Avetho]

Manuel Saavedra You sir are an absolute genius! I'm imagining some To Aru Majutsu no Index level kind of game. Magic or Psychic? Straddle the line like Tsuchimikado and gain practical skill, become the strongest Magician like Aleister Crowley, or become the strongest Esper like the fabled level 6 Mikoto Misaka from the FanFiction Ace of Academy City. Or, of course, decide to play as the one cursed with misfortune, who can negate most all of the aforementioned power with your right hand, and have it severed multiple times, enough to defeat Crowley or Aiwass or Othinus by learning their AI's attack patterns and purposely having your right hand severed to let IT come out to play ;)

[@boxzonprazon5696]

like racing games

[@Cazinskye]

ArcadianGenesis like 2k

[@UnknownPerson-cq3qv]

Because they are to busy making more micro transactions!

[@tyusjonesfan]

Unknown Person nice

[@gamingconnoisseur67]

Only $19.99 for realistic physics

[@optillian4182]

The intent is to provide players with a sense of pride and accomplishment for unlocking different particles.

[@gamingconnoisseur67]

Only 59.99 for a realism loot box

[@mateotierno3780]

Unknown Person actually Battlefront without microtransactions finished being a good game

[@Black-Dawg-Jesus]

When I first saw the title I thought: "Well, because it's technically impossible at the moment to run it." You managed to stretch that to over 7 minutes. Wow.

[@vaskedagame880]

i thought he was gonna at least use that as interduction on some other topic related to it, but no, he just wasted everyones time.

[@nada-zl9rs]

Hey, man. Jesus was black, you know that, right? Beware of youtube strikes for that character's copyrights.

[@ahlatagaci2018]

Not even 10 minutes, he does a little trolling.

[@dumbmelzan5186]

"Objects behave like objects." *Shows woman jumping* WHEEZE

[@cichan9094]

Could you guys not post for a week im trying to study for finals.

[@gameranxTV]

shit srry

[@guedosproduction3129]

Whacky Buckle no no no

[@kqj693]

Whacky Buckle probably should skip a week then play catch up or watch KZbin as a brake time.

[@RackedandStacked]

I feel dis

[@waitingforasubway3790]

Omg same. Tomorrow I have a Math and Spanish test. I should be studying but, that's boring...

[@Exoclypse]

Very confusing video, he doesn't much understand what he's talking about. Even the title is wrong. To sum it up : - Video games do use real-world physics equations, in increasing amount as time goes on (even Pong used Physics). They are used in a wide variety of ways depending on the game, but mainly two are usually present : 1 - "Physics" : this term usually refers to simulating gravity, collisions, explosions, projectiles, vehicles, ragdolls, cloth deformation... The simulations may not be very accurate but the same equations are used as in science, if not the same numbers. 2 - "PBR" (Physically Based Rendering) : this is mostly about describing more accurately how light is affected by a surface (or "material"), thus affecting how the surface looks. Real world measurements can be used by scanning real objects. Now a better question is : Why not use more Physics? (which Falcon somewhat answers) - Hardware limitations. More precise simulations or rendering techniques are used in pre-rendered movies but are impossible in real-time, without going molecular. These include more accurate physics, fluid simulation by particles (although it has already been done, it would be too power-consuming in a full game and mostly just irrelevant) and most of all raytracing which models bouncing rays of light (instead of the much faster rasterization used in games which "draws" objects one in front of the other according to their distance from the camera). - Gameplay. Accurate physics may not behave the way game designers want to or may not be predictable enough or even could be game-breaking. - Irrelevance. Not all games need all this or can fit it in terms of performance or budget.

[@CMoore119]

Exoclypse Jesus Christ dude Noice

[@Exoclypse]

Thanks ^^

[@Exoclypse]

It's nice that they're trying to understand how games work but they should really ask anyone from the industry before making inaccurate videos about it.

[@qwertzuio4295]

Exoclypse this comment is better than the video

[@Lolwutdesu9000]

Well said. Much more informative than the video.

[@evora4944]

Y'all watch Richard and Mortimer

[@gameranxTV]

who

[@evora4944]

Also love your content keep it up.

[@jemarcoballard6911]

gameranx rick and morty

[@twodollarking8009]

Mystical Ronin who dat

[@onefistdaddy]

The Purple Quarian ah.. fellow intelectual

[@jasonsage1417]

I'm a developer, I've written a Physics Engine, I used Newton's formula's.. they work like a charm... kinda cool, but I wouldn't do it again - lot of work and I didn't get into fluid physics or anything that nuts, but they were real equations. :) It's actually really awesome when you get your first "3D-Geometric" basic shapes and stuff visible on screen to watch say something lie a pool table work after you coded all the math, and quanterion math four 4-axis rotation to avoid gimbal effects of 3D Coordinate systems not coping well with full range of spheres without something called Gimbal lock.. Anyways... IT's complex stuff. I've also written games that looked like they had real physics and it was all formulas and you tweaked the "gravity" and things by eye... that was just as entertaining. :) Edit: Definitely Easier to cheat.

[@NaeemAcademy]

bro we need to learn physics to make games in unity or any programming language ?

[@SirPetten_Physicist]

I really want to play this game you wrote. Do you have a link to us?

[@101-r4k1u]

@@NaeemAcademy if you want to create an entire new assets, well yes you need math and physic, unless you use peoples assets or free than you only need to learn progamming

[@jatinoham]

@@101-r4k1u What you are totally wrong it doesn't required to have any maths skills to make game assets just some knowledge of blender, houdini or any other software 🙂

[@J.Wolf90]

Hey Mr developer. I'm bored of video game environments being nothing but an art canvas to look at. These are supposed to be GAMES not art museums

[@lomborg4876]

This is the equation to gravity F=m•g•h m=mass g=gravity constant h=height of object Then you could mix in wind resistance, but I don’t remember the equation for that

[@pc_screen5478]

thought he was gonna explain why, even with these simplifications on the physics, most objects on games don't behave like their real life counterparts (for example falling objects accelerating at rates that greatly exceed the 9.8m/s² of the gravity)

[@MadScientist512]

AH SEU VOU That would have made more sense, as games do use real-world equations for their physics, they just use different values for variables like the 9.8m/s2 Earth G you mentioned, according to another video Mario's supposed to be 9G!

[@peterfisher5931]

im believe in game object actually fall more slowly less than -9.81m/s^-2

[@aidanconcannon5710]

AH SEU VOU aright u cant exceed acceleration. U can exceed velocity. I think you mean they have greater acceleration than real life meaning they can achieve un natural ammounts of velocity

[@pc_screen5478]

Aidan Concannon that's exactly it. I thought he was going to explain why the devs used such high (or low) values when making game physics instead of just using real world values

[@HawooAwoo]

"I don't think people can pull off mid air headshot if you fall at realistic speeds also that would include significant self damage and possible death if you fall high enough." That's not even physics; that has to do more with the limitations of the human body

[@EmperorFett]

Watching this before my Physics final at 4:30 lol

[@BlueDragonMusicPromotions]

Emperor Fett heh heh good luck bruh

[@Daniel-rf2hm]

Emperor Fett smart

[@EmperorFett]

BlueDragon Music Promotions thanks fam, it’s over electricity, magnetism, relativity and light essentially lol

[@gameranxTV]

good luck!

[@DarkDodgers]

Hope you passed man. I got Engineering tomorrow...

[@Alucia0]

I honestly think that both realistic physics and none realistic have their place in certain games. For example ragdoll physics, which is more fun a corpse that hardly moves that you fall over if you go near it, or one that looses all weight and can be kicked around like a ball? Depending on the game I would say that both work.

[@DissociatedWomenIncorporated]

Stopped watching as soon as it became apparent the question being asked here was the most stupid possible variant, i.e., "well why _can't_ I simulate the entire universe in realtime???"

[@feynstein1004]

Fancy seeing you here :)

[@thatboringone7851]

He uses a ridiculous example, but that's more or less the case even for something that might sound reasonable (a single box in an empty room, even). Take collision systems: you need a system that looks right and works realistically (or at least works for the design choices and the players) without getting objects stuck in each other or behaving incorrectly, but the more you try to simulate actual physics in a collision system, the more it's going to require from a computer to keep a stable frame rate, because actual physics is more complex than what our current technology can reproduce in a game at a reasonable framerate. The worse the frame rate gets, the more it stresses the limits of the computer it's running on, the more likely something will fuck up and get stuck where it isn't supposed to, fling itself out of the world, bounce off and hit the player, etc. Plus, real physics are hard to reproduce when not every programmer knows much past what they need to approximate it, and would be a nightmare to debug. The goal in programming these systems, or more accurately improving them, tends more towards "make it look and feel better for the player, find ways to allow them to do some awesome things we want them to be able to do", rather than simulating real life systems more accurately than we would exactly need to for the player to enjoy the game. Even a game focused on physics as part of it's core design and game play doesn't need much from actual physics to act reasonably enough like they should, or would in reality. [Saying this as a bachelor student in programming for games, currently working on reproducing physics through collision detection and collision response, for context]

[@DissociatedWomenIncorporated]

ThatBoringOne yes, I get that, but your whole box in a room example is pretty much another variant of the one I said. Again, simulating a full set of physics in real time even for the contents of a tic tac is obviously beyond a computer's abilities, and it's that incredible obviousness that made the video disappoint me. As a (primarily game) programmer myself, I'd hoped I might learn something interesting, like if there were weird quirks that occurred if you set a few of the simpler variables to be realistic, or because real jumping physics would suck for Mario (although that would also be a bit obvious), or something like that.

[@thatboringone7851]

pixel girl Well, no, I wasn't talking about simulating every tiny detail, I was referring to just general physics (like how in reality if we push against a wall, we're applying force to it, which will be applied back to us since both walls and people are pretty solid, and the wall isn't the one likely to move, whereas that's generally not what we'll reproduce in games). My point was that the closer you try to get, even with basic stuff, the more difficult it gets to handle, the less reason there will be to use it- the guy in this video wasn't wrong to present it that way. I get if you were hoping to learn about something more, that would have been great, but that doesn't make the point made in this video unreasonable.

[@DissociatedWomenIncorporated]

ThatBoringOne, I guess it's an informative video for people who really don't understand much about either or both of computers and physics? And I get the point you were making but yeah, that seems kinda obvious to me too, that even a subset of real physics algorithms being applied even just on a per-object basis would be very hard to pull off in real time... and doesn't even _need_ to be done that way for games, for example I've never seen a game bother to calculate a bullet's path and motion... something that I don't think would actually _be_ particularly taxing on a CPU, but for something that happens faster than the human eye can see anyway it's pointless to not just use an easier method like "first target in line with the gun takes damage".

[@siryoloswagling6728]

*This is an arrogantly misinforming video.* 3:49. It's clear you don't even understand what a shader is. Shader's are used predominately for graphics. ONLY compute shaders are used for physics. All other shaders (vertex, fragment, etc.) are useless for anything outside of the rasterization rendering pipeline.

[@siryoloswagling6728]

The whole video has no substance

[@christianantillon3305]

That techsource setup though

[@highdownhybrid]

Terrible video. Video games do use real world physics equations, at least newtonian physics. „Unrealistic physics“ are not unrealistic in the sense that they use a wrong or unrealistic euler integration, just that forces, mass and friction are tuned differently.

[@MadScientist512]

highdownhybrid My thoughts exactly, same equations with different variables, the title couldn't be more misleading.

[@jcm2606]

And they're operating on a much larger scale. That's ultimately what this video is about, scale. We use equations derived from real-world physics equations, just we do everything at a *much* larger scale than the universe does.

[@TheJustin500]

When you said unrealistic physics, I immediately knew that you're an idiot.

[@paulpereira7735]

Knight Davion oh yeah, because ad hominem makes you some higher level intelectual. Calm down snowflakes

[@TheJustin500]

+paul pereira you don't know what you're talking about. You're accusing me of ad hominem but you yourself call me a snowflake. lol. Dumbo

[@PimpMatt0]

Maybe because physics is hard.

[@wallachia4797]

Explicit Tech i think you meant USELESS

[@cluckeryduckery261]

Physics is useless? Seems legit.

[@paroxysm_brian7817]

E = mc^2 E = energy m = mass c^2 = light

[@cluckeryduckery261]

It's Me AwesomeGaming actually c represents a constant, which is the speed of light in a vacuum, roughly 299 thousand kilometers per second

[@mxcherryblue5943]

It's Me AwesomeGaming now use the Schrodinger's formula to know where is an electron

[@GeorgeCowsert]

I do think that we need an actual system for forces and gravity. Things like Newton's laws and gravity are significantly easier to simulate than the molecular makeup of an object.

[@craggolly]

That's already happening. While liquids are often just planes with normal maps, and smoke just a couple of particles with a smoke texture in games, rigid bodies for example use newtonian physics. This video is misleading and poorly researched

[@Cockalicious]

@@craggolly water definitely cannot be simulated perfectly, there's just too much happening for a single drop on a solid surface, let alone a large body of water.

[@craggolly]

@@Cockalicious not perfectly but we have some decent water solvers already. FLIP simulations are fairly accurate when it comes to large scale movement of water

[@stealthinator00]

" do you think that sonic the hedgehog would be fun if his physics would be accurate" Oh no you do not temp the wrath of Austin from game theory.

[@andrewevenson7798]

I think having hyper realistic physics would open the door to new types of games that revolutionize gaming. That isn’t to say they would be better, it’s to say giving artists such a powerful tool would allow them to manipulate it. For example, say we could replicate physics perfectly, would that not give game designers more freedom to intentionally screw with physics in more depth than they can now? You could have a game where the properties of matter change around you and you must adapt your gameplay to the ever changing physics. I think that sounds cool personally, but it wouldn’t replace traditional games. There will always be someone making almost every type of game.

[@nedduggan4099]

Andrew Evenson Andrew Evenson I think if we perfectly replicated physics in a program we might use it for a bit more than computer games. Cool concept though I do agree with your thoughts.

[@viharcontractor1679]

To perfectly implement physics wouldn't we first need to perfectly understand them?

[@andrewevenson7798]

Vi Con I think we have a great enough understanding of physics so we could implement what we know, but if you are getting all the way to physics on a quantum level, then of course we have a lot to learn. If we could just implement the physics we do understand though, it would be pretty awesome.

[@cpsyche246]

Roses are red Violets are blue Most poems rhyme This one doesn't

[@xX_DRKGORGNO_Xx]

Lol.

[@wallachia4797]

Lmao

[@cozmyc_reaper]

Roses are red Violets are blue I like rhymes And so do you

[@ShowMeSomething1]

This is not a poem it's a random blurt of words from a over emotional person...

[@HappyCynic]

Violets are not blue. They're violet. That's like saying oranges are purple.

[@Gorden121]

You could potentially use actual formulas. Actually, using the formulas derived from analytical solutions is the simplest and fastest way of calculation, since you only need values in and get values out. There's not much else that has to be done or accounted for. But the problem you encounter really fast is, that the vast majority of analytical solutions can only be derived from very simple problems. Most more complex problems cannot be solved analytically anymore, so you do not end up with a nice formula or even any formula. Physical simulations taking actual physics into account take weeks to simulate miniscule spaces. Accurate light scattering simulations of 1 cm^3 can take days to weeks, depending on how good of a result you want. And in video games you don't want to simulate tiny spaces, you want to simulate the whole environment in real time, so simplifications have to be made.

[@OmniscientlyMe]

Yeah, the real limitation is the sheer volume of calculations. # of objects (polygons) * # of effects (lighting/shadows, gravity, momentum, etc.)* # of interactions of either

[@dub16rider]

A way of using the actual conservation laws is quite straight-forward. You pre-define only a subset of any possible movement restricted by the set of animations you want to visualized. Simulating the actual, realistic movement on those on a rather coarse to finer scale leads you to approximations or even accurate results. Defining a subset of possible use cases (movements/ animations) comes in handy now: You calculate the many solutions to what you want to visualize and make a table look-up of what comes close to what is currently happening within the gaming simulation. Further mapping this to some kind of regression model would keep this within reasonable bounds. The amount of possible scenarios is then only restricted by the computing power of the machine you are running your game on.

[@AbandonedVoid]

You could, however, simulate what accurate light scattering would look like and then represent that accuracy in a fixed manner in the map, allowing it to be played in real time. And for most physics simulations, you really only need to calculate a few basic Newtonian laws of physics for it to present accurately. I still find it weird that games aren't assigning mass and force to objects, resulting in a plethora of violations of inertia and conservation of energy, etc.

[@AbandonedVoid]

Add a little bit of friction (including air friction) and you're done. Though i feel like a lot of games _do_ do this.

[@sebastianfeuerstein9306]

Gorden121 Nice, I finally understand what the video meant. Thanks Bro 😎

[@ChadeGB]

As i've been saying for many years, when it comes to games, more realistic doesn't equal more fun. The whole point of gaming is to be able to do things you couldn't do in real life and most of the time that has a lot to do with breaking the laws of physics or bending them to make them more enjoyable.

[@TheRagdollrejects]

The other thing about totally accurate physics would be that games would be pretty much unplayably difficult. -Your character has just fallen 2ft but their ankle went over and now it's sprained from a minor drop and a bad landing...

[@It9LpBFS37]

I completely disagree. You guys chose extreme examples: you dont have to simulate the entire world down to every single particle, as you said, to incorporate certain physical equations for the game. There are a lot of games that feature realistic physical simulation to certain extent. Although not much of them are using strict formulas in code, they use parameters that simulate the behavior in more or less the same way. And going even further, source engine did in fact start using real physics equations (in certain areas like sound dispersion-google it) a few years back.

[@MadScientist512]

The title is rather misleading; game code always uses equations to make things move even when it's simply x=x+1, and those equations are nearly always the same as those found in physics textbooks, the only difference to the real world are the variables used, like Mario's gravity's supposedly 9G, but it still uses the same equation for gravitational acceleration, as does Sonic for that matter, equations for friction and momentum are also used in those games. I'll have to find a source for that Source sound, sounds like Aureal 3D, that was a competitior to Soundblaster and it used a kind of ray-tracing of sound for a realistic sound simulation but lost out to Soundblaster's inferior fake reverb and the technology got 'quietly' buried legally, gotta love those innovation-destroying patent laws (and puns)!

[@xponen]

The original Mario had complex character falling rate, one rate for when user press jump for the character, another rate when character reached max height, and other rate when character is going down... people will confront & present example that real Physic is not required by quoting iconic & notable game like this.

[@seancarroll9849]

This largely depends on the type of game, too. For example, games like Gran Turismo or Forza require actual physics to do what they need to do. Racing style games are as close to real world as you will get. Then we have space based games like Elite Dangerous. The whole Newton's Laws thing reigns pretty supreme in some variations of these games. Some shooters try to treat shooting accurately, but there is always a little fudge room for these games. Honestly, though, not all games need to react to real world physics accurately to be fun. The ones that can pull it off, though? They're unusual gems.

[@BotanistJeff]

KSP anyone?

[@marcobarrientos4953]

GTA 4 got pretty close to realistic physics compared to GTA 5 - anyone else agree?

[@pradeepb5963]

Marco Barrientos I agree... But there's a reason why GTA 5 is more successful than GTA 4, because of its ragdoll physics.

[@marcobarrientos4953]

Pradeep B - Successfulness = "shark cards and customization" go look at GTA 4 gameplay before answering please - it has actual kinetic energy when you shoot someone with stumbling balance loss by the victim, shoot any ai with any gun and they will die the exact same way in GTA 5. Cars cannot be controlled in midair in any realaistic scenario - rag doll physics are nonexistent in GTA 5

[@pradeepb5963]

But you still got to accept that GTA 5 is more fun... I mean, Come on...

[@marcobarrientos4953]

Pradeep B - I never said it wasn't fun, its just not realistic or slightly believable at all compared to its predecessor

[@RatedLFIVEFORTHY]

pradeep i dont get your point he is talking about the realism and the physics and in gta they truly are a joke compared to gta 4

[@banksyy6182]

1:48 TechSource xD

[@xlaatikko]

And i thought i was the only one to realize

[@diego2112gaming]

That went a totally different direction than I was expecting by the name of it--you're coming at it from the approach that to have TOTALLY 100% realistic physics we'd have to simulate quantum states exactly as in the real world, rather than using the way we know objects to interact with each other as the basis of how it do. In Skyrim, one of the first mods I install (after Gopher's iHud) is Realistic Ragdolls and Force--I don't like shooting someone with an arrow and they go spinning off the side of a cliff (not just for loot reasons, either). I don't expect molecular simulations in the game, but I do expect plausible physics approximations. Really all you said was "I like the over the top animations, and simulating things on a molecular structure is impossible anyway in realtime, so deal with it." When people ask for "real" physics in a game, they're just asking for a more plausible representation of of the real world. As to why we'd bother, the whole point of games is to escape--yes, escape reality, but into an alternate reality where you can CONTROL the aforementioned approximations as if you're a being from a Type 6 Kardashev Civilization. I mean, at least that's as far as *THIS* gamer goes.

[@Cockalicious]

Dude the whole point of games is actually fun, it doesn't need to be the matrix to do that

[@DustinRodriguez1_0]

Yeah, games that sought to emulate actual reality would not be terribly successful even if technically masterful. It's like making movies. Making movies is HARD. You can't just point a camera at two people punching each other and have the audience think it's a great fight scene. The vast majority of modern people don't even have the slightest idea what actual violence looks or sounds like. They would see reality as either stupendously underwhelming or gut-churningly brutal. A real punch landing sounds like a piece of meat with some hard bones in it hitting another piece of meat with some hard bones in it. What you hear on TV or in a movie or in a videogame is usually the sound of celery stalks being snapped. For all the idiotic hyperbole about videogame "violence" desensitizing people, it seems to have been ignored that the violence depicted, aside from being between fictional characters who have no feelings, doesn't resemble reality in any way, shape, or form. We call it the same word and that is apparently enough to trick essentially everyone.

[@lmao.3661]

I think minecraft has the best physics, very good

[@MixZup]

yup ...

[@lmao.3661]

Dixie Normas How did you know???

[@michaelmac5261]

We use video games to get away form reality but some games need good physics

[@Daisho32]

Realistic physics would mean no double jumps in many actions games. :(

[@ChucksSEADnDEAD]

Jetpack

[@vsm1456]

double jump is not physically impossible if you have a force to apply to your character. like jetpack or even something magical/sci-fi-ish. even more, there's no such thing as full realism either in level of realism or in gaming areas covered by that realism. so you always have partial realism that itself only applied to some parts of your game. you can have quite realistic bullet physics but with arcade movement (no inertia), or realistic movement but with hit-scan bullets, or anything else, etc.

[@Waffle4569]

People seem to associate "realistic physics" with stuff you could only do in your personal life. Realistic physics just sets a guideline for consistency and explanation. If you want double jump, realistic physics will just push you down the path of having something explain why you can double jump, like rocket boots or something.

[@Simboiss]

Although, you can have "magic" double jumps, but everything else more "accurate" physics.

[@Zappina]

Simboiss Not really. Shrapnels and even bullets doesnt work like in real life. Not to mention injuries. Arma was somewhat realistic reality but still very far from it.

[@NintendoPolitics]

Gameranx needs to talk about the science of jiggle physics

[@boltstrikes429]

Nin10 They most certainly do

[@MichaelStories]

But we do.. we use Navier Stokes Algorithm to make the water. That's how water moves. Similiar to explosion we use Navier Stokes with combustion model because smoke behaves like water but in the air and explosion is basically a density ofthe air that expands really fast. Several other example; We use Reaction Diffusion for chemical reaction or coral growth, PBD(Positional Based Dynamics) for grains of sand, or Fem (Finite Element Method) for pressure of flesh or something soft. What are you talking about? haha

[@LazyDev27]

When someone who doesn't 3D model talks about 3D modeling lel

[@crossofintimidation]

If the laws of physics no longer apply, God help you.

[@CyberLillix]

portal :D

[@ShikaRoddy]

in short, real stuff too complex, easier to do darn good replicas atm.

[@danbodine7754]

Don't even need to watch the video to get that.

[@oyuyuy]

This is bullshit. Laws of physics are extremely simple calculations and "Molecular level modelling" isn't necessary for realistic physics. You simply need to give objects weight and density. The weight doesn't even need to be distributed properly in most cases. All that gamers want is to avoid random bullshit such as objects flying into space for no reason when they collide. It doesn't need to be perfect or anywhere close to it. We only want "decent".

[@Dhalin]

I don't think anybody is arguing against this. However, there are some players out there who want ultra-realistic stuff. Yeah, okay, stuff flying around randomly is stupid. But yet some people complain when, for example, a body doesn't fall off a balcony JUST right. "Oh, his legs would have fallen that way instead of this way!" "The body didn't bounce off the ground just right!" or "That car should have caused the other car to roll 3 times instead of 2 times on that crash!" that kinda junk.

[@oyuyuy]

Well, to those people I'd say that "ultra-realistic" is an ultra-unrealistic expectation. "Car+Car=/=Space ship" however, is a fairly realistic one. Yet, it's the reality of most games for some reason.

[@Dhalin]

Because in those games, it isn't *meant* to be realistic. Like for example, the Saints Row games. They are silly because they are *meant* to be silly. Obviously, a dubstep gun is just ridiculous, but that's why it is in the game in the first place. I find that most games are reasonable for what they are -- games that have absolutely terrible physics are either older games, or games that aren't meant to be serious in the first place.

[@oyuyuy]

That's not true. Some games are definitely meant to realistic. Take PUBG for example, a modern, realistic game where vehicle turn into spaceships and bombs on a regular basis.

[@Dhalin]

"Realism" isn't a 0 or a 1, though. It's a sliding scale with an infinite amount of choices in-between totally silly and IRL. PUBG might be further towards realistic, but yet allows some hand-waves, ones it deems necessary for a fun game. You might disagree with that, but you're not the developer of PUBG. In the same way, Saints Row 4 is a "mostly silly" game, but yet even that game forces you to reload bullet-based guns and still has at least some physics like gravity.

[@theultimatereductionist7592]

THANK YOU FOR MAKING A VIDEO ON THIS SPECIFIC SUBJECT & ASKING THIS IMPORTANT QUESTION THAT I HAVE WONDERED ABOUT FOR THE PAST 31 YEARS!!

[@Kynareth6]

Molecular level physics simulation (not necessarily realistic) is out of question in the near future. Moore's law is slowing down, heat problems are very troublesome. But there ought to be more improvements in real-time physics, because games like Red Faction or Half-Life 2 were really fun. In 2020 there could be even better games like that (Half-Life Alyx is the closest one). Why can't games make a use of all cores that are available (Threadripper goes up to 128 threads)? Because game programmers are lazy and use outdated techniques!

[@skollrum]

Tony Hawk's pro skater probably had the most accurate physics.

[@Zeruss01]

Especially the grinding.

[@joshuanorman2]

Best game

[@stpirate89]

It had a realistic physics mode that you could turn on in THPS2, it was awesome!

[@kyleraccoon6195]

If our universe is a simulation, our universe probably doesn't even use real physics, so games don't have to use real physics either.

[@brockly7916]

agree. or perhaps is an Illusion from our *GOD.*

[@Gfish17]

Christians balieve heaven is more real then this world. but do you want to live there. I say NO!

[@primeirrational]

F U C K Y O U! Or the SPAGHETTI MONSTER

[@BicheTordue]

if god is a dev then maybe he is in a simulation made by another dev god, then we can also be gods.

[@KingofHearts67]

What are real physics then?

[@auggo7324]

This is a very interesting topic, great vid 👍🏾

[@Lattamonsteri]

Half Life 2 had pretty good physics for its time :D it also didn't have "shells" per se but the materials you used to create boxes defined how the box would behave. Wood would float and iron sink and the thickness of a wall could prevent a bullet from passing through. The E3 showcase can be found on youtube :D it's very cool to watch.

[@tux_the_astronaut]

Half life 2 from what i remember uses the havok physical engine.

[@felixb99]

'Not using physics equations' is very different to what this video is talking about. A lot of games will use, at the very least, fundamental kinematics equations to simulate objects moving around the game, interacting with each other and being affected by gravity. A lot of classical physics is really simple for any personal computer to simulate in realtime, and simulation at the cellular level - let alone the molecular level - is not needed to achieve extremely realistic macroscopic physics just as we experience every day, and no sane video game would invest in simulating such physics even if we had the computers capable of doing so, as the net gain would be so minimal for such an extreme amount of work, especially for FPS genre games where people aren't going to notice the fine details of individual atoms, just as no person is going to notice processes acting at a cellular level or below outside of a laboratory. Even if a game wanted to simulate subatomic processes, these could be simulated 10 or 100 at a time within the scope of whatever microscope the game allows the player to use, and then portray the net macroscopic change across a large object without having to simulate each molecule within that object. As an engineer and video game enthusiast, I am disappointed at the misleading nature of this video, and how it misinforms those who don't know better.

[@Avvura]

ragdoll is my favorite

[@LucianC137]

Just Cause 3: RIP Physics

[@BertoLaDK]

No, still tired of trees being rock solid even doh you fly a 747 into them, they still won't move the slightest...

[@cheetoschrist5685]

This is why I think Dwarf Fortress is one of the most complex games ever. It keeps track of the tissue, bones, blood, etc of every enemy and how it's affected. It's like the exact opposite of normal games, as it aims for an accurate simulation rather than a pretty shell

[@user-tu9ox4hj9g]

What were asking is: Is low level physics > high level physics? And the answer is hell no. The closer the physics to our realm the better. Imagine a catapult lunging a great stone into a castles wall. Each stone of the castle wall blows out of position, dust fly's into the air, and the ruins do not instantly disappear. Then imagine a catapult lunging a great stone into a castle wall and all that happens is the wall HP variable decreases and a section of the wall just disappears when the HP is depleted. Which one would you rather watch? "Realistic" physics is more creative and what is more creative is more interesting to experience.

[@FarrakhanWolcott]

2 minutes in I felt like I was click baited

[@KasperDahlTangen]

This guy doesn’t know what he’s talking about, just sayin’

[@User-xw5mk]

I agree

[@gplgs4640]

I may get alienated for saying this, but I really like realistic things in video games. Graphics, physics, control, etc. I just want my games to be as immersive as ever.

[@fartnutte1724]

go outside then lmao

[@seriyooow310]

I've thought about this too. If there were such things as a molecular simulation, where you input the properties of certain molecules and see how they react depending on the conditions you provide. Then scale it up to a real objects interacting with each other. Imagine the possibilities of engineering.

[@carmatic]

for me, having more realistic physics means increasing the depth of the gameplay while keeping everything intuitive... it gives more avenue for showing off your skill while keeping everything feeling fair

[@affluenzashot]

I can't help but notice something being neglected when valuing other physics systems over realistic physics. Sure, the joy from experimenting with fun-based physics is apparent at face value, but the joy from realistic physics comes not from its crazy and unique behavior, but from the crazy and unique objects that can interact with it. This is the perk that comes with having a virtual space, and it is beneficial not only because it allows players to discover crazy and unexpected results, but also helps them better understand the very physics around them. These are generally basic concepts that most game designers like me understand who design simulators, and it surprises me that none of this was mentioned in the video. TL;DR version: Realistic physics can be fun and more useful with proper game design. :p

[@jaymethysell5111]

There is a difference between fun and interesting. Real World simulations are very interesting. And they can also be very useful. But that's not really fun for most people. You're not going to keep going back to the simulator after you've learned what you needed from it because it's not inherently fun. Realism detracts from games. I don't want a realistic fishing game, I want to actually catch fish instead of sitting and waiting. I don't want a realistic fighter, I want to knock people to the moon similar to Smash Brothers. My favortive football game is NFL Blitz, which is not realistic at all. I don't even need or want realistic graphics or photorealism, some of my favorite games are 8-bit or 16-bits and others have amazing cartoony art styles. When I want realism, I'll stick to doing stuff in the real world.

[@affluenzashot]

That is a good point, but again, it appears that something is being neglected or undermined when comparing real-based games with fun-based games in this light. Even in simulators, the ideal aim for game designers is to create an environment that is compelling enough to motivate the player to explore the most out of their creation. One of the key values of having a virtual space is to enable people to interact with things that are impossible or ridiculously costly to replicate in real life, or in short, to go beyond than just "doing stuff in the real world." Realistic physics can have more playability than you may think, since there is such a pronounced amount of knowledge of its complex behavior, which although it would be very difficult to accurately replicate it in a simulator, it would have far more depth than a custom physics system that is limited to the ideas and efforts of the few programmers creating it. Real physics may be fun in surprising ways too, such as experimenting with dynamic buoyancy, black holes or the crazy properties of quantum physics. A good simulator is one that draws in more than just the ones interested in its subject. Simulators also have more value to society since they are designed to educate and interest people in various fields of knowledge, like how physics simulators can be used to introduce people to engineering or astronomy. On the other hand, the core mechanics of a lot of classic game titles were formed primarily from coding bugs that ended up being unexpectedly fun to play with, and from there they generally metamorphosed as a sort of means to acclaim the artistic efforts of the creators and the computational power that was so mystifying and lesser known to the mainstream audience at the time. My main point though is that these are qualities are not exclusive to fun or fantasy-based games. Games with real physics or biology or history can be fun and artistic as well, I just feel that society's game design community needs time to evolve in order to hone it effectively. I'm not ranking on classic games like NFL Blitz, Super Mario, Halo or F-Zero. It was games like these that inspired me to become a game designer and programmer. Gaming culture and technology are maturing though, and I believe part of that involves using their unique qualities in a way that has an impact on the vast possibilities of the environment around us. That being said, this is why whenever there is a debate of using real-world mechanics in games, this concept I explained is a necessary point to bring up. Sorry for the long spiel.

[@helgenlane]

But game design should also be simple. Because if it's too complex, most people won't have time and energy to explore it. And realistic equals complex. People love realistic car-sims/space-sims because it's just as fun in real life. People love realistic shooters because they are "hardcore". But the more realistic it is - the more niche it becomes. Because complexity makes gameplay stressful. And why most people play games? Because they want to relax or just forget about real world. Realistic survival games... They are fun because you get immersed. And even though they are complicated, you already know most of the stuff because it's common in your life. But flight simulators aren't as popular and common, even though people love flying: they have to learn how to operate a plane, it takes too much time and effort. That's why people prefer arcade flying games - they are easy to learn and most people don't notice "unrealism", because they don't know what it's actually like in reality. And this applies to anything "realistic" in games. We also have imagination and can believe in something if it is true to the world it exists in. And often unrealistic things are more fun. That's why I believe "realism" in games should be used to give more possibilities/opportunities/choices to the player, and not as a main feature. Yes, there are ways you can make realism fun, but you can just as well make it unrealistic and most people won't notice it, and others won't mind because it can still be believable.

[@helgenlane]

Green Pulsar don't you want more people to play your game? And didn't you read my entire comment? I clearly said "more complex = more niche". It was actually my point, that simpler games have wider audience. And since you want to have people playing it, it should not have ultra-realistic gameplay, because it's complex and complexity sways people away.

[@helgenlane]

Green Pulsar but it is niche -_- It's niche is "isometric fantasy RPG". Did they sell at least million? Maybe it seems like a lot to you, but it's still a niche. If you target a niche big enough like "isometric fantasy RPG" of course it will sell. But "most" people won't enjoy it.

[@Ondrix]

I would say the argument of having, or not having, totally realistic physics simulation depends entirely on the game in question.

[@monkeyshell3858]

Haha they used Techsource's setup

[@theiathegreatest]

MonkeyShell haha they did

[@OleMariusSvendsen]

as a bachelor engineer and now master student i can confirm what this topic of video is trying to point out. For my current classes is about thermodynamic and its properties regarding a surtient pressure foruce it change (the properties). We engineers use sinplyfied models in 2d calculations for most parts to handle the information. Non of our equation on a high level are accurent, its lab tested to make assumption on how to describe physics. Take a bullet as an example. When you shoot it, it will increse its temperature as a function of forced applied to the gunshoot itself and its exit out the gun with friction. It will lead to property change on a thermodynamical perspective. When you apply our lab tested formulars it varies on how the friction will act on the body (bullet) regarding change and transfere of heat to the sourondings. This means; - gravity will work on the bullet in several ways; - gravity constant change on where u are on earth and what your ground mass is on it. Since air friction (or other fluids, and yes a air is afluid just liek water and oil) is a function on temperature, air windspeed, air density (change with temperature), viscosity (change and temperature) etc, change itself when the bullet is cooled downed its path will differnce. - The bullet will becouse of friction change form due to material elasticity from bullet exit, airfriction and temperature property change, differce in geometry. Every formulas wich engineer use are geometry based means new geometry change would require different formulas to predict assumptions on how the path of the bullet would be. This is a simplified analysis from me regarding a bullet alone. For windpath and temperature analysis for a single room alone, it can take hours to complete with just 2 finate parameteres alone. to implimated every thing into one big simulation its not possible with current tec. Also the knowhow from progamers are not even close to describe such in one big software.

[@marcuslombardo8615]

This is something I’ve been wondering for a very long time and you answered it. Thank you.

[@ZTheLastViking]

I really like the thumbnail good job.

[@adrianyocupicio341]

0:53 was.. was that Andrew with a fidget spinner on his nose?

[@gameranxTV]

Yes

[@adrianyocupicio341]

But.... but why though

[@adrianyocupicio341]

gameranx if you're reading this, thank you for the amazing content Tom

[@tetragem1887]

Sure was!

[@asubaruthathatesexistance6871]

Vault boy, because it's real life physics

[@thecaptain4630]

Imagine if lucioball had physics

[@TheFlamingPiano]

Yeah imagine NFS with BeamNG physics, hardest game ever, at least for those with pursuits

[@ChrisChronos]

The only thing I want totally accurate physics in is race simulators. The point of them is to accurately represent the cars you are driving. Every race simulation enthusiast thrives to have the most realistic simulation possible, building full cockpits just to play the game, because the chances are most of the cars we are racing in the simulations we will never drive in real life, so to get a model of it done accurately helps us understand what it would really be like to race it in a way normal racing games cant achieve, and through this we find it very enjoyable :D. Every other game on the other hand is fine how it is for me

[@isshaye]

It would be cool if there was a game where you had the super power to change the amount of realism of the physics in the game.

[@personperson9794]

If they could make that extremely open ended like turning some blocks into stairs it would be very impressive

[@calebshady1670]

Let's get EA on this idea and they can bring us micro transactions

[@jayfrmbrklyn4543]

Caleb Shady shhhhhh don't give them any ideas

[@diegog1853]

This is somewhat interesting but in fact, in the modern day with powerfull computers we can't even model a single cell without taking major compromises. Because of quantum physics and that we don't have a mathematical way of predicting the movement of three different objects that produce a field (like an electrical field) That means that just by having an atom with two electrons, you have to make compromises to the math. Molecules are in a whole different league, and with all of our compromises we today are just able to model protein, that is a hugely complex molecule but is really far from the complexity of a cell. So forget about modelling a space in a molecular level, its not just a matter of computing power, we don't even have the mathematical tools to do it. We don't know what is going on down there.

[@thijsjansen2052]

why would you want to simulate quantum physics for a model, there is no point. Those have very little effect on how an object behaves in space.

[@diegog1853]

Thijs Jansen if you simulate molecules and atoms you have to include quantum physics for it to be realistic. I only said that because they were going that far. And besides, saying that quantum physics have no relevance in a normal scale is not entirely true in every single situation. Most of the modern electronics work with quantum principles (a laser for example). When microscopic objects move through space you need quantum mechanics

[@thijsjansen2052]

yeah but if you want to use one data point per molecule in a scene, it is not relevant to compute quantum physics cause it does not have a large impact on newtonian physics and the only thing poeple want is blow up in games not not simulate how a laser for example works or a quantum computer. That aside this video is pure crap cause there is a difference between unrealistic physics and crapy physics, thereas unrealistic physics have there parameters wrong so the game is more fun, but crapy physics have just bad implemented physics systems. you can have over the top physics even with an really good physics systems just apply more force to rigid body's or turn the gravity parameter down.

[@diegog1853]

Thijs Jansen i disagree. I mean yes, if the physics engine uses some kind of large particles to better simulate some aspects of the world then you are right. There is no need for quantum physics, but if we are specificaly talking about simulating molecules and atoms then using newtonian physics is not enough. For example, the distribution of the atoms in a molecule is purely described by quantum physics, its the state of less energy, and the shape has a great impact in how the molecule interacts with others. If we go to the easiest of examples. The shape of the water molecule determines how it interacts with other water molecules to crystalize at 0*. And the structure of the crystal has the major impact of ice floating in water. Of course someone can program this properties manually into the computer, but if thats tve case then there is no need for programing every single molecule. The problem is way more complicated when big complicated molecules like protain interacts with each other, in which aproximating the problem like they are two dots will not take us anywhere. Of course, the video doesn't know what they are talking about. Is completely different to emulate random particles that to emulate specificaly atoms and molecules, like the video says. I was just trying to say that if they want to say that, the problem is way more complicated than having the best pc

[@thijsjansen2052]

well than it depends on how you interpret the "down to moleculair level" and it depends on what is relevant for your game, but i really don't have the authority to talk about this as just an hobby graphics programmer.

[@abeefyblackguy5524]

I would want totally accurate physics for racing games. Sure, I can race in real life, but can I race multimillion dollar vehicles at the click of a button? No.

[@almeidaariel9]

A full complex physics simulation of a car or something would take hours or even days to render just a single frame, because it's so much to calculate (the air flow, the explosions inside the engine, the piston movement, gears, tyre contact to the ground, etc). Basically if you want accurate real life simulation you can (that's what engineers use to develop cars nowadays), but you couldn't never do that at 60 or even 30fps with current hardware. Modern engineering programs can reproduce with up to 95% to 97% accuracy.

[@arturkovacs3689]

Really good video, great job! Although I'd like to point out that a lot of games DO contain SOME real-world physics equations. You are absolutely right, that there is not enough computing capacity in the world to simulate everything on an atomic level in real-time, but parts of the physics in some games are actually just like in the real world. For example if you think about an object falling and hitting an other object the way they move and act on each other is based on real physical equations. (In many games at least) And I know that you kinda mentioned this when you talked about the water, but I just didn't feel that it was that clear.

[@aidandruck2423]

I interned for a foundry and we ran a pour simulation for pouring iron into a mold. It was a very in-depth physics model that did indeed take into account great detail in molecular interactions. It even modeled the formation of dendrites within the cooling liquid as well as the phase transformations in each area of the mold. We don't cast patterns bigger than 32" square. It was fairly normal for the sim to run for a few days straight. Up to 40 some hours for a whole sim, longer yet if we ran optimization. And this video is talking about rendering a whole video game map with that level of detail, with the added dynamics of player input. Maybe its possible with a supercomputer, but certainly not in real time.

[@angelbalashev]

1:49 Tech Source!?

[@raimo1985]

guys a little help, wich game is the one showed at 00:10?

[@froumts]

Hayate Everspace

[@ilumina8226]

if ur talking about the 2 girls jumping. It's Paragon

[@Arrotect]

what about 4:02?

[@cjeff99]

Husey 4.02 isn’t a game, it’s a blender (animation software) water simulation. Search water simulation and it’s like the 4th video that comes up

[@Arrotect]

cool, thanks man

[@techuser1619]

Can someone PLEASE tell me where 3:20 is from. The clip looks beautiful! Is that a game or a game engine benchmark?? Thank you in advance!

[@potatopower7893]

TechUser looks like The Vanishing of Ethan Carter

[@gerboog]

Roblox

[@Dzeroed]

All being said- those are some absolutley _beautiful_ game shots you've used in this video, I feel like I am breaking the law of physics by watching this on a shitty phone and not seeing them in HD glory as intended! Shame on me. I'm watching this over again when my kid gives me the TV back from his Minecraft session

[@phuturephunk]

I've been playing video games for decades and I don't think I've ever heard anyone unironically complain that the physics in a video game don't match reality unless the game was exceptionally bad. I would say a good deal of game players out there have at least a basic grasp that modelling particle interactions in real time are far...far..beyond what even the most powerful home computer can muster. I dunno man, with this vid I guess I was expecting some talk with developers on how they approach the simulation of physics in context of different game types. I remember a long while ago a channel that may be long gone doing a video about how the jumping physics in old games like Super Mario Brothers made the game. Essentially how you were able to physically navigate mario through the game world and how precisely it matched the launch, float and drop off of his jumps made it very satisfying. This shows too, in other games like the Castlevania series and Metroid. Stuff like that..

[@downer9098]

are you drunk?

[@anuragguria008]

Tech source

[@aaboboch9016]

What game at 3:20?

[@clintshidwood4511]

Aabo Boch it's "disappearing of ethan carter"

[@dolphinboi-playmonsterranc9668]

Aabo Boch Pokemon Snap 2

[@Luxalpa]

The main reason why physics equations isn't used in VFX is because of the lack of control that you get with them. Being able to simplify or complexify the equations into something that's more imperative allows us to art direct the results and give them the right feeling. The other thing is that material physics are actually surprisingly complicated. You can sorta see that when it comes to Physically Based Rendering, which needs to simplify some of the quantum effects of light in order to properly represent surface properties.

[@MagSun]

This video is one of the commentaries about "there cannot be realism in computer games until we are able to simulate everything". Heard the argument once from an Activision R&D guy, that realism in games requires a rendered face to have 20M polygons. Realism is much more than a poly count or how *deep* the simulation goes. You can always find flaws in the simulation (center of mass is a bit off, friction is to high, etc.). Most important thing is that it feels *right* (still envying the soldiers in Battlefield handling a 20kg machine gun like any other pistol

[@AkenoAkine]

How does an in game Ai see the game or the world within a game? Like do they see what we see? Do they actually see the player the objects and weapons ? Or do they just see lines of code?

[@MadScientist512]

Yuki Yukino Here's a good video explaining what they see and do: kzbin.info/www/bejne/pHbRf6efqtKaf6s

[@toastybatch565]

Yuki Yukino they don't see anything. AI is just a set of code that creates the illusion of intelligence, but really, it all depends on how clever the programmer is. For example, in a stealth game, a lot of AI draw lines in a 120 degree sector in front of them that goes out a specific distance. If the player object ends up in that cone, they get alerted. That's how they "see". It's just code.

[@mahikannakiham2477]

They see and feel nothing. A computer program is just a bunch of transistor gates with varying input voltages and a canal that forms when the gate's voltage reaches a certain threshold.

[@zheoferyth]

Mahikan Nakiham One could say the same for synapses in the brain.... Although no video game AI's are actual AI's so you're right. Some of the more developed AI's are getting scary with their depressive "thoughts" though.

[@mahikannakiham2477]

Billy-Joé That's true, but ultimately, no one really knows what is concious and what isn't. Is a rock concious? Is an atom concious? Is a computer concious? Who knows? But one thing I am pretty convinced is that the characters in video games are not concious. They don't even live in a virtual world, I mean, everything is just pixels that light up on a 2-dimensional grid...

[@supernaturalswampaids8083]

Dislike squad?

[@jaymethysell5111]

Arcade physics are so much more fun than real world physics. Look at GTA for example: each new iteration of the series introduces more realistic physics and the games have progressively become less fun. GTA5 is a still great game, but it's because of the great story and additional features that weren't in the previous games. However, the new, more realistic vehicle physics were somewhat of a regression. Gone are the days where you can flip your car into doing outrageous stunts and still be able to drive away from pursuing police. That was super fun, and it's missing from GTA5. There's a similar situation in Halo. It was way more fun when it was an arcade style FPS party game. Starting with Halo Reach, and then continuing in Halo 4 and 5, it has tried to become more like COD with more "realism" - Blah. For me that's a huge turn off. When I want realism, I'm just going to stick to trying out stuff in the real world. And consequently, I only want to use simulations to improve my abilities at doing real world tasks.

[@shamicentertainment1262]

I prefer realistic driving. I like having the feeling of solid cars, that don't just slide all over the place.

[@zeyadjy7229]

I am a beginner unity developer and I love your channel, In all game engines, the physics engine is intended to simulate real-world physics as you said games supposed to be fun experience not physically acquired experience so depending on the game type and mechanics the physics engine is twisted to give the desired results but most game use gravity and they handle collisions with physics too, I don't agree with what you said about 3d game objects and characters are basically just empty shells and they cant react to physics like real objects made of molecules, I am sure that you know every game object that affected by physics have components that give that object all the mathematical qualities of a real-life object so when it affected by the physics system it will react like real object will do, for instance anytime you let the rigid body fall to the floor it will bounce and stop in a slightly different angle - just like in real life.

[@zacharyamos6139]

I think realistic physics are always fun. The entire game could just revolve around the physics because having realistic physics is so fun.

[@AHurdOfBronies_]

I, for one, believe that people would actually enjoy game physics simulations if they were Totally Accurate. Just look at how many people enjoy Totally Accurate Battlegrounds Simulator.

[@paulrisk606]

Good video, good points. I very much enjoy Kerbal Space Program, a game which simulates physics in a way that I am not currently able to do in real life (cuz I don't run a space program). Because I don't get to see that kind of physics in real life, I would like to see the game be as accurate as possible. But in a game like the Sims or Half Life, there are mechanics that I am much more likely to experience in real life (walking, eating, putting objects on top of each other, shooting guns occasionally...). For me, I enjoy those kinds of games more when the emphasis is not placed on recreating elements of life that I already experience, but more on the things that I cannot do myself (jumping in and out of portals, building a house, exploding things). When games try to simulate too much of what is commonplace in real life, it makes me want to go home and rethink my life.

[@viharcontractor1679]

How would we ever have the fortune of witnessing glitch compilations if physics were perfectly implemented in video games.

[@BotanistJeff]

me: Why is my banana purple? gameranx: because it doesn't need to be an exact shade of yellow. this is basically how you answered the question of why don't video games have accurate physics.

[@chris94pacific43]

When you run out of video ideas so you start questioning video game physics