Microsoft shows off potential power of the Cloud at BUILD

[SvennoJ]

Machiavellian said:

thismeintiel said:

OdinHades said:
A better GPU would be a much easier solution and would save millions of R&D costs. But why take the easy road, it's TEH FUTUREZZZ!

I don't know, I still don't buy all that powah of ze cloud stuff, sorry.

Completely agree with this.  Plus, this demo was heavily controlled and limited.  Now, let's move this into a real world situation, like online co-op.  Imagine if there are two or more gamers in a level, all making different decisions, with dozens of AI enemies and destructable environments.  Then, just imagine not having the fastest/most stable internet connection.  No, cloud gaming that shares large amounts of computations between local HW and cloud servers is still just a thing of the future.  Of course, by the time it becomes 100% doable in real world execution (in a decade or two), a beefier GPU will still be the much better option.

I would have to say that Gaikai/Now is a much better solution in the real world, as there is no shared computing, just streaming the A/V and your controller inputs to and from the host server.  There's also the benefit that it is 100% already proven to work. 

This is actually not correct.  This would be no different then having the game process server side with the computations for thee physics ect done by the cloud and only sending the information the users can view at any one time.  The amount of data is not big at all, if you include compression its even smaller.  The Cloud is not rendering the scene its just send the caculated data back to thee client machine.  This type of setup is not dependant so much on the speed of your connection instead of the constant sync.  Any disruption would cause little lag in the game but evven that can be smoothed  out with predictions.  It would be the same type of predictions that is used within MP games to keep things in sync.

As Gaikai being a better soltution we will see.  Both systems have their pros and cons and each solve the problem in a different way.

At the end of the demo the cloud is doing the physics for 37,000 chunks, position, velocity, rotation. Updating all those independently moving chunks at 32fps plus the geometry changes for new chunks, broken up chunks and left behind gaps is a lot more data than a h.265 compressed video stream.
And the client still needs to be able to render all that extra geometry. You're taking the physics calculation away, not the strain of rendering 37k objects.

It's a good example of how a Gaika/Onlive server can deal with high stress situations by utilizing a server farm to spread peak demand. As long as not everybody is blowing up everything at the same time, it will be a more economic solution instead of having the peak power available for each client separately. It's a tech demo of how slow down can be eliminated in a server type situation. I don't see it being practical as helping out a local machine with physics.