Seems some will deny every thing Microsoft do...

1)Just like the Playstation Now demo Sony showed us earlier this year. Was there so much people telling Sony were lying???

2) Cloud processessing is use with things that could run async so latency should not be and issue. An explosion could be calculated at the moment you throw a grenade or lauch a rocket so at the moment of the explosion you will already have the data to process it.

Think about it this way: the worst latency scenario someone can experience withcloud processing is the best case scenario he may hope with a service like Playstation Now. Because PS now is totaly dependant of latency, cloud processing isn't.

3) Games will not look nor play different than they do right now.

@Bolded:  You are making assumptions on how the tech works without actually knowing how it will be implemented.  The thing about cloud compute is that it was never mentioned if its calculated realtime or not.  Also without knowing how the data is sent from the client to the server and back or what is actually sent make it very hard to base any assumptions unless you are famalier with such work.

 The problem with saying if something will work or not or how it effects a game realy does depend on how the developers and MS tackle the problem.  Your assumption could be going down a path the game is definitely not doing which would make it null.  I believe people should wait until MS give out the details of how their cloud compute is working before jumping to a bunch of assumptions that may or may not play out.

1) Why do you think it would be any different with game that use cloud computation. If the game required internet dev will specify it just like for Destiny, Diablo 3... 

2) Yes and that is actually not really a factor for cloud computing. So it's a win for cloud computing over streaming.

3) Why should it be scalable??? If devs want the game to be able to play offline than yes but nothing prevent them from making the game online only. 

4) No one is forced to use cloud feature in there games, except maybe some first party, and we are not talking about rendering. the rendering is still all done on the system.

5) What happen to diablo 3 players when they are offline? What will happen to destiny players if they want to be offline?

6) Again, if the dev want to make the game online only there is no problem, if not they will probably just turn off the feature that require cloud computing. If you think that cloud compute is there to make every Xbox One more powerfull you're looking at it the wrong way. It is there to add new features and possibility to games. Forza 5 driveatar is a feature that is turned off when you play offline. If the feature that require cloud computing is required for the game like Titanfall than the game will be online only.

7)That's only your opinion. A want to see sources, give me a link where it say's that capcom, turn 10, respawn use to have more headaches when working on cloud feature.

8)Again you're looking at it the wrong way. it is not to make your Xbox One more powerfull but to offer devs more possibilities.

9)Again this is only your opinion.

You are simply ignoring what I pointed out as the fundamental difference between rendering entire sets of operations on a server and paritially rendering an operation (or a subset thereof) on a server.

These are different because they introduce headaches for devs. These are different because the result might now outway the effort. Your arguments aren't appropriate.

I know how the tech works. I don't need to know how the details work to understand client-server relationsips and basic in-game rendering that's referred to the server. I don't need to know details to understand how it works. And the fundamentals of these proposed operations are very vulnerable to serious scrutiny.

1 - PSNow, stupid as it may be, is a proven concept that has worked before. Besides, streaming any steady video content is just a basic primarily one-way street (only thing going upstream is control entries).

Your first point doesn't really compute.  All solutions started a concept.  Then those solutiosn goes into implementation and they those implementation impove and mature as time goes on.  PSNow started from concept then the tech started to mature as the people who development started to improve it.  Basically you are dismissing cloud compute like people did with game streaming.  Why don't you wait until we see the tech and how it works and implemented before making assumptions on its success.

2 - Async = meaningless. Think about it, if you cause a large physics event, and it creates a new environment in process as a result, then you bounce/interact at close range for more destruction of those elements (or a vehicle is driving through it, secondary explosion, wind/water impacts, etc) then it needs to be as syncronous as possible or the immersion will be completely blown. Besides, you're talking about something slow like a grenade or rocket, how about a big tank shell that will take .1 seconds to travel from your gun to the wall when you press the button? And a second shell fired right after that one by a teammate into the same wall nearby? Requiring debris from both events to be calculated against each other for collision detection/new physics results.

This is a theory which has it own flaws.  You seem to forget that the developer controls the world.  Depending on how the game is made there are no random events.  If you take the Build demo, you can see that when the person fires the shot, this calculations are done client side.  The server side calculations are done when the whole building is being collaspe.  In other words, the server only does cloud compute for the bigger events not the direct input by user.  This probably will be no different in crackdown.  Explosions that are direct inputed by the user are done client side.  If those explosions causes a major event, those will be done server side this freeing up the console to mantain client side work and keep frame rates high.  I believe people really are not thinking about the problem correctly and thus going down routes that will not be implemented.

3 - Then what's the point? To spent an asston of cash and put ridiculous resources to something that has little tangible benefit?

The point of cloud compute is to free up resources that can be used for client side work but allows the game to showcase events that would take down the beefiest of machines.  In other words cloud assistance allows the game worlds to be more dynamic.  The thing is, I believe MS has researched this tech way more than us.  Its not like its something MS thought up the day they announced the X1 but probably something they have had their research teams spending years on different implementations.

---

I'll say it again, for a series of super premium online-only titles, it would make more sense for Microsoft to begin installing a next-gen 'Xbox' server base that streams the ENTIRE game to you, 100% of everything cloud side processed. As long as you have a decent connection with low ping, you have something FAR better than trying to run a game calculated in two vastly disparate environments and then stitching it back together on the fly.

MS has been rumored to be doing this as well.  They actually had a demo of Halo 3 I believe streamed to one of their phones or tablets.  Anyway, there are multiple ways to implement cloud assisted calculations.  What you present I believe MS never stated they were doing.  It probably would be best to wait and get the details from MS as all your points might be totally null.

Arkaign said:
I never thought that game streaming was impossible, because the data involved is miniscule, and all calculations are done remotely. It's no different in many respects to me using PCAnywhere in the 90s over ISDN to support clients in Israel and Japan with remote desktop.

The key not wheather you think something is possible or not, the key would be if you are developing in this space and know if something is possible or not.  Just because you do not have a reference wheather cloud based Physics works doesn't mean there are not may protoypes out there where the tech has been vexxed.  This happens all the time with new implementation of software.  At the software company I work for, customer have thought many things we have done is impossible because our competitors have not done it.

I also don't think you understand what I mean by async. Once you have a big event happening with a ton of pieces moving, all that has to be synced out between client and server, otherwise you have no real interaction with those objects possible without breaking the physics model.

In software, Async means that the client does not have to wait for the server to send back information.  The client can continue to send info to the server and not have to wait for a response.  The server can queue up those transactions and send they back in order to be process by the client.  You can have async and schronous communication at the same time especially if the backend and front ends are designed for such purpose.  I understand what you are saying its just that you only present a one sided solution which does not take into different client server implementations. Is this setup complex yes it is but thank goodness MS development a platform called Orleans to handle the complex interactions behind the scense.  Orleans has been in development for over 3 years so my guess is that MS understand how complex such calculations can be and spent 3 years making it easy on the developer to code against such multi-threaded actions.

Let's get a decent basic hypothetical : you are playing a game where you control a player who can untie a ribbon and release hundreds of various size balls from the ceiling in a net. These balls vary in size and weight. Half of the floor is rubber, half linoleum. The graphics are all rendered by your console. Okay, pull the string, and everything tumbles down and starts bouncing. This is an event being calculated by the server, server tells the client where things are, client draws them. Now walk up and start bouncing the balls against each other. You are now changing the variables dramatically with every action, and that input needs to get back to the server, and there is no longer a 'set piece' in place, but rather a constant variable series of actions that all have immediate consequences. If there is lag, then the physics model won't appear realistic, or worse, it will fail to render and cause client side issues/glitching.

Thats a good scenerio but the key would be if the design of a game would go for such a dynamic situation for the cloud compute part.  The client only needs to render the point of view of the person moving.  Once you get beyound the person point of view, all other calculations can then be sent server side.  Great thing is that information can always be sent to the server without the client waiting on a response.  Data can be plugged in by the server once those calculations are made.  You can easily dedicate CPU threads for just those purposes or with the X1, it already have dedicated hardware just for that purpose without having to waste CPU resources.  Either way, the game designer has full control of what happens within their world.  Having this control allows then to easily predict when to perform server side cal and when to do client side.  What you are looking for is a scenerio where their is chaos but in reality, a game can be tighly controlled to not introduce those scenerios.

Does that make sense? Because a big set-piece action triggered by the user, but that is not interactive beyond the start/end of it, is useless. We can already do that locally. The bigger stuff, huge synchronous cloud to client and back physics events of a continually changing cycle : THAT is both what would be incredibly cool AND is ridiculously impractical.

We probably could argue this point all day.  You really only show one side of a coin where they are multiple different solutions.  I am more than willing to wait until MS give us the data on how them are implementing their solution.  I have plenty of ideals of how they could go about the work but I hate making speculations when the data is really sparse at this time.

It bears repeating that dedicated hardware physx cards start having problems when they're moved from a bus of 2000MB/sec to one that is 500MB/sec. Moving that same physx engine to one that exists on 1MB/sec (8 megabit) is astonishingly unlikely to be impressive or effective.

So how do you explain the Build Demo