Gaikai coming to PS4 2014.

[patsu]

No, cloud computing is definitely not PR or fluff since everyone is using it for real. But you have to use it correctly.

Gaikai is a difficult service to launch. I'm guessing most of the third party CDN and edge server providers don't have GPU in their servers. The latency requirement should be more stringent. Sony may have to do it themselves.
Adobe's Creative cloud probably have GPUs on them for the heavy image/video processing jobs. Most of the MMO and server games don't render on the servers.

Xbox One's cloud servers are CPU centric like everyone else. So MS can share their servers with the rest of the Windows 8 cloud application servers.

 

[dobwal]

Laughing because you're suggesting that service models that seem to be roadblocks for one vendor are being touted as keys for another.

Cloud computing is pr fluff, but streaming the game is somehow better. Not being able to sell digital goods is bad, but never actually being able to 'buy' them is great.

I'm not saying you said these things, it's just the tone on the internet. The biggest payoff for Sony seems to be baffling their audience with bs, but this is moving into business so I'll leave it here.

Yeah, that's the nature of the internet.

 

[patsu]

If the MS executives had been more prepared, they would have given good answers to their audience.

For Gaikai, at least some of us have seen OnLive and Gaikai work on the net. If I hadn't played with OnLive myself, I would have been a lot more skeptical. A hell lot more.

EDIT:
The big question for me is: Can they really make $$$ from it in 2014 ? (Probably not ?) How long before they expect to turn a profit ?

 

[dumbo11]

I think unlike the traditional "large data center somewhere" approach, they may need to host the Gaikai servers right at the edge (i.e., in the large ISPs and cable networks).
Or have their Gaikai backend hooked directly to these ISPs.

http://www.criticalgamer.co.uk/2010...-how-it-will-change-the-world-of-video-games/

So on what scale will you be launching? How many servers are you planning to have? How many do you have? “The way it works is: the closeness to the server is actually critical, it’s a key thing. So my target is to have 300 data centers in the world. It will take 50 or so just to cover the US. Right now, today, we have 11 data centers live and we add new ones constantly. Amsterdam just came online in the last few days, Paris is next and then London. So we just keep adding data centers. Denver is coming online soon and so is Minneapolis. We just keep going one after the other.

All the dots are the Gaikai servers and the area they reach.

“We have a map of the US. The white dots are the data centers, and every time we add one on it starts lighting up green dots around the area. It has weird and strange patterns because of the way the Internet is wired. So it’s not like you think; you can’t just make a big perfect circle or something. It makes a strange pattern and you can start to see where servers are needed. So we’re launching this month with around 30 million minutes of server time. And that’s what we sell publishers; minutes. By Christmas I will be selling 300 million minutes a month of demo time across the US, and then I will try to replicate that in Europe and that would give me a total of 600 million minutes a month. And that’s plenty of inventory. That’s a lot of demo time.”

 

[dobwal]

If the MS executives had been more prepared, they would have given good answers to their audience.

For Gaikai, at least some of us have seen OnLive and Gaikai work on the net. If I hadn't played with OnLive myself, I would have been a lot more skeptical. A hell lot more.

EDIT:
The big question for me is: Can they really make $$$ from it in 2014 ? (Probably not ?) How long before they expect to turn a profit ?

Whats troubling me is that most of us have seen MS's streaming plan for the XB1. The XB1 is basically built for it. Compression/decompression logic sitting behind DMAs on a wide I/O interfaced with a big pool of low latency RAM. Yet, MS didn't to bother to market it with any conviction.

The only thing I can think of is that MS is using a marketing approach that spans multiple e3s. I can see where that motivation comes from. The 360 didn't have a big first year in terms of hardware sales. Every XB1 and PS4 is going to sell like hotcakes until probably the end of Q1 2014 where we will be a few months away from the next e3. A second big splash for e3 with a bunch of features that haven't been readily discussed and not available on the XB1 in its first 8-12 months of life may be meant to give the XB1 a second wind.

 

[Silent_Buddha]

I'm wondering myself how they are going to do this. If it's the OnLive model where you have to pay a monthly fee AND you have to buy the game, I'm not seeing this as being any more profitable than OnLive. For PC games it made sense, gaming PCs can be expensive. A monthly fee could be considered cheaper than buying a PC to game. To play your old PS3 games? A monthly fee could well end up being more expensive than just buying a PS3 after the PS4 launches which is going to limit how many people find it worthwhile. Then again, perhaps that's the point? It allows you to claim "backwards compatibility" while making it too expensive for most people to consider it viable, thus limiting the load on your servers. Consider it a small loss for a marketing advantage.

But without doing that, I can't see it being profitable either as you need to pay regularly for the costs associated with running a datacenter.

I never thought OnLive would succeed (and it didn't) and didn't think Gaikai would either (and it didn't). I'll honestly be surprised if Sony can turn Gaikai into a profit generator. Realistically I think it'll always be in the loss column, but potentially the marketing gained from it would offset those losses.

Regards,
SB

 

[ERP]

But without doing that, I can't see it being profitable either as you need to pay regularly for the costs associated with running a datacenter.

They actually just rent space in existing datacenters and locate special machines there, they don't own the datacenters.
MS does own datacenters, but it also has machines in other peoples datacenters.

I never thought OnLive would succeed (and it didn't) and didn't think Gaikai would either (and it didn't). I'll honestly be surprised if Sony can turn Gaikai into a profit generator. Realistically I think it'll always be in the loss column, but potentially the marketing gained from it would offset those losses.

Those are the big questions when and how much.

 

[Silent_Buddha]

They actually just rent space in existing datacenters and locate special machines there, they don't own the datacenters.
MS does own datacenters, but it also has machines in other peoples datacenters.

That makes sense as they aren't heavily invested in cloud computing as MS is. But it's also going to be potentially more expensive than what OnLive or Gaikai was doing per user as unless they have come up with a PS3 emulator, they'll need a PS3 (PS3 blade?) for each user, whereas with PC games, they could potentially run multiple virtual machines (now that GPU acceleration can be used in VMs) on each "box."

Those are the big questions when and how much.

Yup.

Regards,
SB

 

[patsu]

If they do this, they probably have repackaged PS3 (22nm ?) into rack-mounted servers without the Blu-ray drives. It has to be as dense as possible.

Whats troubling me is that most of us have seen MS's streaming plan for the XB1. The XB1 is basically built for it. Compression/decompression logic sitting behind DMAs on a wide I/O interfaced with a big pool of low latency RAM. Yet, MS didn't to bother to market it with any conviction.

The only thing I can think of is that MS is using a marketing approach that spans multiple e3s. I can see where that motivation comes from. The 360 didn't have a big first year in terms of hardware sales. Every XB1 and PS4 is going to sell like hotcakes until probably the end of Q1 2014 where we will be a few months away from the next e3. A second big splash for e3 with a bunch of features that haven't been readily discussed and not available on the XB1 in its first 8-12 months of life may be meant to give the XB1 a second wind.

MS may be more focused in their TV ventures and Cloud compute extensions at the moment. I think one of the execs mentioned that they don't think cloud game streaming is ready for prime time yet.


EDIT:

http://www.criticalgamer.co.uk/2010...-how-it-will-change-the-world-of-video-games/

Ok, so they had 11 out of the targeted 50 US data centers in 2010 ? By now they should be quite close to 50.

 

[AlphaWolf]

For Gaikai, at least some of us have seen OnLive and Gaikai work on the net. If I hadn't played with OnLive myself, I would have been a lot more skeptical. A hell lot more.

EDIT:
The big question for me is: Can they really make $$$ from it in 2014 ? (Probably not ?) How long before they expect to turn a profit ?

My experience with Onlive actually makes me skeptical. I'd say it was awful for any twitch, racing or timing games

It's hard to know how long to be profitable until we know their exact model, but I expect years before they can recoup their investment and that assumes they can convince people it's much better than it has proven in the past.

 

[patsu]

My experience with Onlive actually makes me skeptical. I'd say it was awful for any twitch, racing or timing games

OnLive only deployed a small number of servers.
You need to be "close" to a server, and have a good enough ISP. ^_^
I had to go to a friend's place to try it.

I also love the TV spectating feature. Hope Gaikai expand the spectating concept. It's pretty neat to see so many sessions together.

 

[Kb-Smoker]

Only thing i would like this "service" is for playing game demos. I hate having to download a 7GB file just to try out a game.

 

[patsu]

Yes, that's the main Gaikai proposition today.

Sony offers a full BR game demo every month in PS+. I wonder if they will replace it with Gaikai streaming.

 

[AlphaWolf]

OnLive only deployed a small number of servers.
You need to be "close" to a server, and have a good enough ISP. ^_^
I had to go to a friend's place to try it.

I also love the TV spectating feature. Hope Gaikai expand the spectating concept. It's pretty neat to see so many sessions together.

I'd wager that I was as close to an Onlive server as I will be for any Gaikai. And my isp is 50Mb cable.

If they are going to want to sell this they are going to need convince more than the people sitting on a data center with a dedicated t3. I agree the spectating is a nice feature, but in and of itself I wouldn't pay a dime for it. If you want netflix like money for this kind of service it needs to get close to the relative experience of non streaming gaming, and I'm not convinced they will be there for a while.

 

[Kb-Smoker]

Yes, that's the main Gaikai proposition today.

Sony offers a full BR game demo every month in PS+. I wonder if they will replace it with Gaikai streaming.

That would be great. I try crysis 2 on gaikai the day sony brought them and it ran pretty good. I was really surprised.

But i will believe it when i see it. It hard to take something like this at face value. When you try go roll this out to millions of ps4 owner when a new demo comes out.