Modern multiplayer games already do the "world model" calculations on the server, with clients simply rendering the little data they have.

Yep, definitely not new tech. (You probably know this next bit, but just for anyone scrolling by).

Having your server verify every action is the only way to prevent hackers from simply sending a packet that says "I got a headshot on that guy". Instead the server replays through all their actions. You never say you got a headshot, you just say when you clicked and the server figures out if you hit.

And because like you mention, clients only receive gameplay data and not the fully rendered screen, this allows clients to do things like interpolate the (delayed) server data with their own realtime input elements. Otherwise every single action you take would be delayed by whatever your latency is, and every other player's actions would be delayed by a similar amount. Instead, it appears seamless until your latency gets so bad that you just begin teleporting between positions because the server can't track you well enough (often called multiplayer rubber-banding).

Instead (well designed) game servers run their own authoritative version of the game world and track each player's latency, and use that to replay the previous few ticks of the game when an input comes in from any player. So if you have a 100ms ping, and press "W" to walk forward, when the server receives your forward input it treats it as if you have been moving for 100ms already.

Basically, saying every player shares the world model is incorrect unless you can set them all to the same latency.

https://developer.valvesoftware.com/wiki/Lag_compensation

https://developer.valvesoftware.com/wiki/Latency_Compensatin...

Edit: additional source, these gafferongames articles are a bit old (2004) but still extremely relevant.

https://gafferongames.com/post/networked_physics_2004/

https://gafferongames.com/post/what_every_programmer_needs_t...

This isn't entirely true for most games, clients run world simulation for prediction and sometimes can make authoritative changes. Almost no multiplayer games are complete dumb terminals ,you'd end up with the same latency issues that Stadia has.

Complex physics with high object counts can be difficult because synchronizing a lot of state is bandwidth intensive. Most games with lots of moving objects (RTS for example) use deterministic lockstep so only inputs need to be transferred, even if there is an authoritative server.

That only partly helps with cheating. Client side cheat software still has access to the internal game state and can modify rendering (so you see in the dark for example) or input (point the gun at any visible opponent). If you only remotely control a video stream, that's harder. I guess the next step would then be computer vision on the video stream.

>That only partly helps with cheating. Client side cheat software still has access to the internal game state and can modify rendering (so you see in the dark for example

Exactly, companies like "Artificial Software" (artificialaiming.net) have been selling aimbots for FPS for a decade or longer. I was curious back in 2009 and purchased access for a game in the Battlefield franchise.

It allowed me to have colored rectangles around all other players, my own team being green and the opposing team being red. You could see this rectangle through walls which would let you rack up kills and avoid getting ambushed. You could also auto-aim and tell it if you wanted the head or the torso, you could headshot people that were so far away you could not see a single pixel of their body. It also allowed you to start using your knife rapidly if an enemy got within a certain proximity of you.

It would even allow you to spawn in on enemy team memebers instead of your own team on maps that allowed you to spawn on team members AND it would change your name on non-hardcore servers to appear to appear as an ally to the enemy so you could spawn on one and casually follow them around, them thinking you were on their team, and just knife or headshot them when you wanted or follow them to a group of them and det pack/grenade/spray and pray and kill a bunch of them.

A quick look at their site and they still have many bots that still allow many of these features (and others like radar hacks, no recoil, no bullet spread etc) for various titles because all of this is still on the client side.

> I guess the next step would then be computer vision on the video stream.

Matching output seen on the screen has existed for decades under the umbrella of macros. I remember writing my own for an MMO around 2005 using AutoIt; it was significantly easier than any client modification.

Then what’s stopping us from making a MMORPG with the physics of a simulator?

I would hazard to guess that latency would be a killer for that. It's painful enough if you have 100ms ping on tactical FPS games like CS:GO for instance, but if the delay in a player's position updating caused a completely different physics cascade to happen (running into a wall rather than a stack of cardboard boxes for instance), that could get pretty jarring to see either the correction occurring constantly or the physics delayed after someone's actions.

Of course I'm just theorising - it could be an idea that no-one's implemented yet.

A lack of gameplay from the resulting physics to justify the server costs? Game devs don't simulate worlds for the fun of it. They have a purpose behind things. There's a cost/benefit involved.

> Game devs don't simulate worlds for the fun of it.

Unless you are Tarn Adams

Or about a thousand other small indie devs, but somehow I don't think Stadia has much interest in them.

P.S.: Don't take this statement the wrong way, I love small indie devs and donated for Dwarf Fortress.

The technical difficulty of making an MMORPG I'm guessing.

Hmm, I'm wondering if rendering on the server would make things a bit harder for bots? It seems hard to figure out exactly what to send without at least doing visibility calculations and clipping?

I have heard that the latest Counter Strike servers do player occlusion calculations on the server so that cheaters can't see through walls. If you can't see an enemy, than the server won't even send your client their position.

This is simply an incorrect statement. Even with a dedicated server, most games simply replay game logic with deterministic lock-step simulation based on ordered user input events. There's some exceptions, but (partial) state synchronization is tricky especially with complex physical interaction and complicates/limits game design with a huge amount of edge cases so it's not a wise choice unless you absolutely need to do this, like open-world or MMO games.

What games are you talking about? For example, Natural Selection 2, Counter Strike GO, and Overwatch all do a full simulation of the world on the server, and the server is 100% authoritative. Clients are often running the same simulation locally to predict state despite network latency. Players also can experience packet loss or reconnect at any point and still have full state synchronization.

Rendering is the computationally heavy part.

I would agree - if every purchase of a game on their platform included a coupon for a physical copy of the game on my console/PC of choice.

The fact that I have to pay full price for a copy of a game that may go away at any moment is ridiculous.

That’s a problem with their business model. I’m not defending Stadia, but the concept of cloud gaming.

This is the problem with modern gaming. Everyone wants to fix all the problems that aren't the core problem we have with gaming: microtransactions. No one cares about how nice your game looks or how low-latency your game is if you designed it around microtransactions with pay-to-win gameplay and a laughably greedy content update schedule.

Microtransactions are a pet peeve of the hardcore crowd, but ask any average gamer and it probably doesn't rate in the top 10 disagreements with modern gaming. It's an artifact of the hardcore gamer filter bubble online, not a widespread grievance.

The problems Stadia addresses - eliminating 100GB downloads and constant 10-20GB patch downloads that might take hours to pull down, removing the cost of entry in the form of a console, being able to play AAA games on your phone - these are things the average non-hardcore gamer cares about.

But it replaces multi gigabyte downloads with multi gigabyte per hour downloads.

It allows instant-play on 10mbit connections, where you would need to wait 10-20hrs otherwise to play. Even on gigabit connections, you're going to wait ~1hr to play AAA games as opposed to starting immediately. Instant gratification is powerful.

Sure but I’m highlighting the cost of that cause it’s not free. If you’re playing at 4K (which you couldn’t on 10mbit) you’ll be downloading 15Gb of data per hour. I also predict this is going to go up rather than down as we see more data packed in to help reduce felt latency.

Not until the world's internet gets a heck of a lot better.

There's no beating the laws of physics here. RTT in normal installed titles with modest system requirements is already an issue for gamers, let alone the latency this requires. Maybe in a few years when something like Starlink is in wide use, this can be revisited.

I really don't get who Google's target market is for this. A flat monthly fee I can see, that's what Playstation Now and, I think, XCloud is doing, but requiring full price purchases on this platform is lunacy.

Starlink also has to contend with the laws of physics (in particular, the maximum density of ground stations is quite low). It's a service for rural customers and a select few urban/suburban customers at best. It's not a mass-market technology that's going to put cable/fiber ISPs out of business.

Darn.. I hadn't heard of this density limitation yet. I was getting rather hyped up for nothing :(

Until you play COD with a latency of 80ms

You kids today. We would play Quake over a serial line back in the day. Uphill both ways. And we liked it.

When playing Quake over the net, the connection only has to synchronize the players positions etc. And Quake was pretty clever extrapolating the players motions, so even when packets were slow, they seemed to behave naturally. This frequently caused them to jump, when the positions were updated by real data.

With Stadia, the whole gaming graphics are streamed to your local client, so when the network is slow, the graphics will lag. And of course, it takes magnitudes more of data to transmit.

QuakeWorld did. They original was not so clever about networking. Much like streaming games.

I know you're joking, but serial line afaik has lower latency than the USB stuff being used today

It was actually Heretic, now that I think of it, which had a SLIP networking capability for two-person duels. Those were the days.

Maybe not, routers could have serial interfaces back in the day. They were way slow. Maybe they even ran frame relay.

Pretty sure routers still have serial interfaces today, for when everything goes to shit.

They don't. I'm not talking about console ports but interfaces for forwarding.

In the snow!

Yes, I don't get streaming for realtime games.

You already do, just with different consequences. Our brains are very good at mitigating input latency since it deals with it all the time, so we adapt quickly to 80ms. Compare that to the current system where what you see on screen is not what actually happens in the game. You get used to it, but every game solves it differently and it always feels bad. So I think that a lot of players would actually prefer this kind of latency over the kind they are used to now.

Also note that many who play for example COD does it on TV's and consoles which has hundreds of ms already without people noticing it, so playing with 80ms input lag would be fine for 99% of players.

> Also note that many who play for example COD does it on TV's and consoles which has hundreds of ms already without people noticing it

This is just straight up not true in the slightest. There's a reason HDMI ALLM exists, and it's not because "nobody notices hundreds of ms."

80ms is very noticeable, it's like moving around in sludge. hundreds of ms is straight up unplayable. This is no where close to what consoles & TVs are currently doing, which is closer to 50ms.

I didn't say that specific people didn't notice it, just that a most people are surprised if you tell them that they have hundreds of ms latency when they play on their console because they thought it was practically instant because their brains tricked them.

The fact that console gaming is so popular despite it having such horrendous latency is proof that latency is not an issue for streaming services to go mainstream. I'd look more at unreliable latency and data caps, people definitely notice 100ms latency spikes.

> just that a most people are surprised if you tell them that they have hundreds of ms latency when they play on their console

They're surprised because that's absolutely not even remotely close to true. Consoles & TVs do not have hundreds of ms of latency. This horrendous latency does not exist. That's why it's not an issue.

They do have hundreds ms of latency, the fact that you don't they they do means that you didn't notice hundreds ms of latency for your whole life.

Look here, the games averaged 100ms for a monitor with no display lag, add another 50 - 100ms for a typical plasma tv: https://displaylag.com/console-latency-exploring-video-game-...

The smallest hundreds is 200ms. 100ms is half that.

And I do notice 100ms which is why I mostly play on PC with sub-50ms end to end latency. But this is besides the point.

> add another 50 - 100ms for a typical plasma tv

The typical TV has never been plasma? LCD TVs have had game modes for a decade and those are 10-20ms latency. Consoles automatically switch into game mode, too, and have for a long time.

This is incorrect. Furthermore, just about every consumer TV has a “Game Mode” setting to further minimize latency by turning off post processing — a feature that wouldn’t exist if people didn’t care about latency.

I had an intuition that quantum entanglement might be used some day to give us worldwide 0 latency communication, until I understood that this would break the universe’s causality :)

This is laughably first world problems. I've been playing cod and other multiplayer games with >80 ping my entire life. It's fine.

I was expecting people to be worried about stadia being 200 or more ms latency but 80? Come on.

These 80ms were just fine because game engines already employ a lot of nifty tricks to minimize the negative impact of latency. Tricks which mostly only work as well as they do because the engine has an inside view on the gamestate. 80ms of video lag are far worse for the gaming experience than 80ms of "conventional" client-server lag.

Try to track a moving target when your screen updates almost 0.1 seconds after you adjust your joystick.

I expect Stadia input latency to be much greater than that, though, because most display hardware adds at least 30ms on top of whatever latency Stadia introduces.

Always online is really just for smooth microtransaction delivery.

Microsoft is working with Sony and Nintendo on this in some form or the other. Cloud gaming has a future, but it's not going to be Stadia by the looks of it.

the thing with multiplayer games is that they are often latency dependant like first person shooters or fighting games, or mobas. Portable and cheaper single player games may be a better application in practice, because you're more rarely fighting over advantage over fractions of seconds with other agents, and NPC behavior is likely easier to predict for the machine learning next frame prediction stuff I imagine they are using

Hardcore gamers may not like it, but it's hard to argue against the observation that the overall trend everywhere is towards hosted services. Yes, there are some limitations imposed by physics but applications requiring specific high-performance local [client] hardware are going to be increasingly niche. (As, arguably, a lot of the most hardware-intensive gaming already is in the scheme of the population as a whole.)

> Hardcore gamers may not like it

I'm not a hardcore gamer -- I think hardcore gamers would call me a "casual gamer". But Stadia is not interesting to me at all. It offers tons of important limitations, and brings no actual benefit that I can see.

You can't see any benefit to never having to download a game before playing it or running 4K games without needing the latest gaming hardware? Given the obvious theoretical benefits regardless of how Google actually delivers on them, it's hard not to take your claim of "no actual benefit" as disingenuous.

I should have said "no benefit to me"

The benefits you cite are not things that matter to me at all. They may matter to others, of course, but I can't speak for others.

I'm a very casual gamer and have no particular opinion about Stadia in its current form. And am perfectly willing to believe that it's pre-1.0 in its current form. But it's the direction things are headed.

Context matters. For gaming the main trend is move to mobile.

Agreed. Hence my comment that traditional gaming is already something of a niche.

Plus gaming has a very sharp peak utilization at a certain time of day and then much lower. When not being used for gaming they can use the GPUs for training, so the marginal cost of providing the gaming stuff could be very low compared to if they had to idle all that hardware most of the day.

I agree, but I think Microsoft would end up leading the way. They have all their Azure data-centers already, and I believe they're working on Xbox/Cloud streaming. Looking forward to what game developers can do with powerful dedicated servers.

Google (GCS) has huge gpu server farms everywhere too. that's not the factor here

The anti-cheat potential would actually be a killer feature for me. Unfortunately the games which would benefit the most from it also tend to be latency-sensitive.