The cloud is just over the horizon [2018-2019]

I also have newfound respect what Red Faction did with big structure physics on lastgen consoles. What most stands out in the released game is the weird lack of projectile-impact effect on geometry; it's almost as though projectiles vanish. No puff of concrete dust or spark on metal.

It was a mighty impressive game indeed. Running on 360's Xenon no less.
 
It was a mighty impressive game indeed. Running on 360's Xenon no less.
I read a reddit post from a volition dev on it. I wonder if I can find it again. He actually went to say that destruction was not impressive tech as he knew how it works. The impressive tech they developed was for AI to navigate destruction tech. As in if you blow down walls previously there they could now see past it etc. That was way harder and more complex.

And they also had a very hard time making mass destruction fun.
 
Last edited:
I agree with that sentiment. Playing devil's advocate though, is there really a technical breakthrough on the destruction, or is Crackdown just benefiting from faster Internet that old games didn't have? What's the BW use in play when things get busy?
Seems fairly complex as per many precautions they took here with the title thus some basic missing features at launch.
Also development hell as a result of the MP destruction likely points at many difficulties in making it happen.

https://www.crackdown.com/launch-day-update/
 
Is there any word on the reason for the downgrade? Is it:
1) that Internet speeds would struggle to keep up?
2) that more cloud resources would be required?
3) that more local resources would be required?
 
No to four, simply because the piece sizes could be smaller and more varied like the original demos without changing the nature of the game. That is, better destruction quality wouldn't affect the gameplay or fun. I expect the issue to be related to what was actually possibly with a guaranteed QOS over the internet. All the early demos were over local LAN.
 
My point was there doesn't necessarily need to be only technical reasons for the downgrade. It could be other factors. Gameplay could be one or even financial reasons could be another. I understand this is a tech forum, but there could be a host of reasons why they made the changes.

Tommy McClain
 
No to four, simply because the piece sizes could be smaller and more varied like the original demos without changing the nature of the game. That is, better destruction quality wouldn't affect the gameplay or fun. I expect the issue to be related to what was actually possibly with a guaranteed QOS over the internet. All the early demos were over local LAN.

I haven't played it myself, but can't you run, jump, and fight on the debris? At least the few videos I've looked at seem to show people moving on the debris rather than clipping through the debris.

That means you'd need collision detection for every piece of debris. That would get more and more computationally intensive the smaller the debris.

Perhaps someone who has played the game could clarify this.

Regards,
SB
 
I haven't played it myself, but can't you run, jump, and fight on the debris? At least the few videos I've looked at seem to show people moving on the debris rather than clipping through the debris.

That means you'd need collision detection for every piece of debris. That would get more and more computationally intensive the smaller the debris.

Perhaps someone who has played the game could clarify this.

Regards,
SB

Based on what @AzBat says, your supposition is probably the most compelling reason for the downgrade.

Hopefully this tech can be integrated into some other games. Ideally, higher profile games like Gears of War or Halo, so they can comfortably dedicate some more cloud resources to them and dial up the procedural destruction all the more.

The original had veeeeeery realistic looking physics, whereas that which has released suffers from some degree of the polystyrene effect that plagues the physics of every game. That's not to diminish the devs' achievement, they've clearly worked wonders with what they've got, it'd just be cool to see it with some more resources thrown at it.
 
Great selection. These videos make it clear the scale of C3 are indeed beyond anything that came before it on console. It's not too unlike RFG, but with 10x the scale (although somewhat similar debris size)
As for BF games, yes, the destruction is less detailed. The models are segmented in fewer bigger pieces, bit the physics are still procedural the same way. And C3/RF still also has "pre-baked" parts at which geometry will break. Geometry is not segmented in real time, the pre made pieces are just separated.
Anyway, C3 sure is synching A LOT of moving objects at the stress points there. Cudos to the team to pulling that off. Say what you want about the game, the promised tech did come through (downgraded, but still beyond anything anywhere else)

Honestly, i see nothing impressive in these videos... it's maybe my memory, but i don't see anything especially more impressive than a game like Just Cause for instance. The physics might be a bit better in Crackdown 3, but at the same time the world is much more simplistic with more basic geometry, less details, etc.

To me the Cloud was simply PR bullshits...
 
Last edited:
Honestly, i see nothing impressive in these videos... it's maybe my memory, but i don't see anything especially more impressive than a game like Just Cause for instance. The physics might be a bit better in Crackdown 3, but at the same time the world is much more simplistic with more basic geometry, less details, etc.

To me the Cloud was simply PR bullshits...

I'm starting to wonder if the cloud is there less for the shared physics calculations than it is for the global shared world modifications. IE - the debris appear to stay for entire matches and as has been noted the debris from destruction affects player movement, sight, and presumably weapon shots.

That requires all clients to have the same world data. Considering destruction isn't canned with static pre-made destruction that means changes to the world state must constantly be updated to each client in a match anytime something is destroyed. I'm guessing local machines are then taxed with the computations required for collision detection with the new state of world geometry (buildings destroyed, and lingering debris on the battlefield).

In other words, this goes significantly far beyond anything else attempted in a game up to this point.

Whether it makes for compelling gameplay is up in the air. But the tech behind it is likely quite impressive.

Regards,
SB
 
I think they should have rolled with vehicles along with base defense/attack in MP to show off destruction. Spend credits to build up your base or weapons. etc.

Destruction is extremely superficial if the objective is simply TDM. Might as well get nVidia to provide their cloud GPU servers for this gimmick. o_O
 
Last edited:
I'm starting to wonder if the cloud is there less for the shared physics calculations than it is for the global shared world modifications. IE - the debris appear to stay for entire matches.
The DF video still has it disappearing, and the collision look very rudimentary - bounding boxes.

The main achievement seems to be lockstep physics data. The DF data ws side-by-side the same, although the consoles were in the same room I believe, so internet latency would be identical. It'd be good to see performance on machines in different places. Of course, that's a lot trickier without partying! DF seemed to think joining matchmaking at exactly the same time worked 4 out of 5 times, but that might be because their local was so similar they were always paired.

We actually need lots more info to know if the tech is anything special or not. We need amount of data sent and what the latencies are like, and whether their pair up low-latency players only, and is the destruction procedural or are all these pieces predefined, and all sorts.

The other big point about the cloud was local PC augmentation. All we have here is game servers using some new network skills, which isn't at all dependent on the cloud. So even if the debris synchronisation is something new, it still doesn't represent vindication for the Cloud in gaming.
 
Last edited:
Honestly, i see nothing impressive in these videos... it's maybe my memory, but i don't see anything especially more impressive than a game like Just Cause for instance. The physics might be a bit better in Crackdown 3, but at the same time the world is much more simplistic with more basic geometry, less details, etc.

To me the Cloud was simply PR bullshits...
I believe the Cloud was something MS was planning to use on XBOX for general uses. But then after the PS4's performance difference they adapted their communication and focused it on graphic/performance related functions to counter Sony's message which has turned it into PR bullshit. If they were communication Cloud based general functions (player based AI data, persistent game world changes by player input etc), with some performance related features being just a possible tick on a box people would have viewed it more positively.
 
I don't think MS had any particular aspirations for The Cloud regards XB1. Cloud computing is about making loads of money from services and big corporations, which is why MS built Azure. As an afterthought, they probably looked at what Azure could bring to their platform as it no doubt ran Live, and maybe, somewhere along the line, an exec pressured the divisions for ways to make Xbox seem better.

"What about all this Azure server power? Surely we can do something with that?"
"We can definitely run game servers and Xbox Live services from it."
"Services schmervices! How about each Xbox connecting to servers to be made better? Can that be done?"
Engineers look at each other.
"Well...conceptually, I guess."
"Awesome. That's our message."
"But we don't know..."
"Hi, Marketing John. New platform message for you - The Power...of the Cloud. I know, great isn't it."
"Latency and bandwidth..."
"Gives an idea of limitless potential. Love it. I'll see what the biggest numbers I can get from the engineers."
Engineer under breath..."He does realise that clouds are just water vapour, right, and have pretty much zero power?"

;)

Honestly though, the lack of anything needing the cloud shows it was never a serious consideration. If it was part of their gameplan, they'd have been working on content from the off and using the Cloud substantially in their first and second party content. It's very much a solution looking for a problem. The benefit of the Cloud over static servers is the ability to spin up jobs, so realistically what it brings to gaming is an economy, which is more for MS's bottom line than any game feature. You could potentially have MMOs bridging servers on the Cloud, for massive populations, but too many players gets unwieldy anyway and doesn't make a difference. MAG had 256 players and it played the same as having 64 because they were spread out.

The ideal of sending off jobs from a machine and having the Cloud return a response - someone in this game has crashed; deform their car and return the model. Someone in a different game has shot a building; model the destruction for them - is a pipe-dream. The Cloud can't be any more than server-based gaming and can't do anything dedicated servers couldn't. Only if there was a cost thing, and MS offered free server power to drive a game that'd cost to much to run on dedicated servers, could Azure enable something unique.
 
Embedding a YouTube isn't a legitimate response to milk's statement. He said the technology came through. It did.

No it didn't.

It's like saying that game X will support 4K 60fps ray traced graphics and then you release a 720p 25fps game that uses raytracing only for low resolution reflections.

Of course the "tech" is there but is nowhere near to what was supposed to do, that was the whole point of the claim.

I don't think MS had any particular aspirations for The Cloud regards XB1. Cloud computing is about making loads of money from services and big corporations, which is why MS built Azure. As an afterthought, they probably looked at what Azure could bring to their platform as it no doubt ran Live, and maybe, somewhere along the line, an exec pressured the divisions for ways to make Xbox seem better.

"What about all this Azure server power? Surely we can do something with that?"
"We can definitely run game servers and Xbox Live services from it."
"Services schmervices! How about each Xbox connecting to servers to be made better? Can that be done?"
Engineers look at each other.
"Well...conceptually, I guess."
"Awesome. That's our message."
"But we don't know..."
"Hi, Marketing John. New platform message for you - The Power...of the Cloud. I know, great isn't it."
"Latency and bandwidth..."
"Gives an idea of limitless potential. Love it. I'll see what the biggest numbers I can get from the engineers."
Engineer under breath..."He does realise that clouds are just water vapour, right, and have pretty much zero power?"

;)

Honestly though, the lack of anything needing the cloud shows it was never a serious consideration. If it was part of their gameplan, they'd have been working on content from the off and using the Cloud substantially in their first and second party content. It's very much a solution looking for a problem. The benefit of the Cloud over static servers is the ability to spin up jobs, so realistically what it brings to gaming is an economy, which is more for MS's bottom line than any game feature. You could potentially have MMOs bridging servers on the Cloud, for massive populations, but too many players gets unwieldy anyway and doesn't make a difference. MAG had 256 players and it played the same as having 64 because they were spread out.

The ideal of sending off jobs from a machine and having the Cloud return a response - someone in this game has crashed; deform their car and return the model. Someone in a different game has shot a building; model the destruction for them - is a pipe-dream. The Cloud can't be any more than server-based gaming and can't do anything dedicated servers couldn't. Only if there was a cost thing, and MS offered free server power to drive a game that'd cost to much to run on dedicated servers, could Azure enable something unique.


They needed to promote Azure.
They needed to give fans some "secret sauce" to believe in because PS4 was more powerful.

I don't think it was hard to do 2 + 2
 
No it didn't.

It's like saying that game X will support 4K 60fps ray traced graphics and then you release a 720p 25fps game that uses raytracing only for low resolution reflections.

Of course the "tech" is there but is nowhere near to what was supposed to do, that was the whole point of the claim.

We both agree that they did disappoint on delivering exactly what shown on the first demo, and simultaneously did do something beyond what the modern standard.
The rest is semantics. Let's not waste time there. I think the same as you.
 
Back
Top