Next Generation Hardware Speculation with a Technical Spin [2018]

[Shifty Geezer]

Streaming's come on leaps and bounds over the last few years. I don't know what PSNow's like, and xCloud is still up in the air ( ) but I can happily play Spider-Man via Remote Play, at my friend's house, which is roughly a 20 minute drive.

Right. It's local. When on the road around the world, accessing servers in Washington, you have an inherent very long delay just in the time it takes the signal to travel thousands of miles, let alone routing. The implication is that MS have latency down to playable across the world, despite the fact we know pings across continents can easily be 200+ ms. It's an implication that shouldn't be made, because it's borderline impossible and why these companies are trying to get servers closer to where they are needed. You can just build one epic installation in some cheap country and serve the whole world from there.

 

[mrcorbo]

Yea passive cooling ? Did I use that term incorrectly?
It doesn't have it's own cooler and relies on the rack

I think once you are required to move a fluid (air or liquid) to promote cooling during normal operation you are no longer passively cooled, but I'm not certain of the textbook definition.

 

[iroboto]

I think once you are required to move a fluid (air or liquid) to promote cooling during normal operation you are no longer passively cooled, but I'm not certain of the textbook definition.

hmm it might be better coined as passive heat sink.

The NVIDIA® Tesla® V100 GPU Accelerator for PCIe is a dual-slot 10.5 inch PCI Express Gen3 card with a single NVIDIA Volta GV100 graphics processing unit (GPU). It uses a passive heat sink for cooling, which requires system air flow to properly operate the card within its thermal limits. The Tesla V100 PCIe supports double precision (FP64), single precision (FP32) and half precision (FP16) compute tasks, unified virtual memory and page migration engine.

 

[Tkumpathenurple]

True, locality does matter. I think it's worth reading that part of the quote again.

Spencer concluded by talking about the issue of latency while streaming games. He mentioned that he’s already streaming games while he’s on the road around the world, and the test servers for Project xCloud are still in Washington, but Azure has a global scale.

The last sentence just merges together a couple of details that deserve more time:

- "already streaming games while he’s on the road around the world, and the test servers for Project xCloud are still in Washington" says virtually nothing, because we've no idea what the quality's like and, as you said, no idea the kinds of games he's playing or the limitations of the tech.

- "but Azure has a global scale" is the bit I'm focusing on. Anywhere that has Azure servers will get some quantity of xCloud servers and my understanding is that Azure's quite widespread.

So, I think it's poor reporting, not a poor response.

All of this game streaming malarkey makes me curious if the pyramids are actually for streaming Assassins Creed Origins across every corner of the universe, but the Sphinx is missing her aerial. I'm going to go there with my Vita and a wire coat hanger.

 

[vipa899]

Consoles future gonna probally be streaming boxes with server centers around the globe to provide the games. This way piracy will even be harder then its now and totally fits the console idea of closed box environment/pay for online etc.

 

[iroboto]

designing our next-gen silicon in such a way that it works great for playing games in the cloud, and also works very well for machine learning and other non-entertainment workloads. As a company like Microsoft, we can dual-purpose the silicon that we’re putting in.

I only care about what this could entail. If this is AMD, then we're looking at some feature sets closer to MI25/60 at least on the ML side of things.
It still doesn't have Tensor performance though.

So they'll have both this and Xbox One SoCs in the cloud? Or just 1 type of blade.

hmm questions, questions. Perhaps XBO is just the test bed for greater investment.

 

[cheapchips]

Its probally not that easy, they would get more performance out of 10TF turing then 10TF out of amd Vega/navi in a console.

I wasn't being particularly serious. I would say though that we're never going to see a Turing without PC tax so it makes flop battles pretty worthless.

 

[Xbat]

2019/2020 will probally be close to the rtx 2000's successor. Amd competing with 2018 close to mid/high end range hw in 2019/2020 isnt all that stunning. We need more competition from amd and maybe intel.

It's all about the watt usage and chip size when it comes to consoles, so anything that gets close to 1080ti performance will be impressive. The thing is in a closed box with devs making games with 1080ti specs as the minimum will be amazing don't think PC when it comes to 1080ti performance.

 

[egoless]

Consoles future gonna probally be streaming boxes with server centers around the globe to provide the games. This way piracy will even be harder then its now and totally fits the console idea of closed box environment/pay for online etc.

Streaming doesn't solve that. It just makes it even easier to...share. http://fortune.com/2018/08/20/netflix-hulu-amazon-video-streaming-password-sharing

 

[anexanhume]

Some independent corroboration of the AdoredTV Ryzen/Navi rumors


https://hardforum.com/threads/adore...on-3000-series-leaks.1973015/#post-1043970615

There is a whole lot of reality in that video. A lot. There is a little wrong, but not a lot.

He’s staff at HardOCP, not a random poster.

 

[bgroovy]

The pull quote about the ability to leverage the silicon being developed for Project xCloud and, by extension, Project Scarlet for non-gaming workloads is interesting. It makes we wonder in what ways that might affect the design and what goes into it. Are we just talking standard x86 CPU and GPU processing units or will there be more custom hardware that maps well to machine learning?

You know what it tells me? MS is constitutionally incapable of focusing on console gaming as a goal in and of itself. It is the same mistake they have made time and time again. It torpedoed the design of the Xbox One and it's sales potential. And now they are planning to make the same mistakes, in a different way next gen. It's the same thing I have warned people about in the past, but everyone wants to believe things are different under Spencer and Nadella. This makes it pretty clear that is not the case. The greater company only cares about Xbox in as far as it can be leveraged for something else. In the past this has always resulted in a compromised experience for console gamers compared to platforms that do not see their hardware as a stepping stone to something more lucrative.

 

[iroboto]

You know what it tells me? MS is constitutionally incapable of focusing on console gaming as a goal in and of itself.

thats one perspective you can take.

The point of innovation is to break boundaries and change the game. If you’re doing neither you aren’t bringing anything to the table. And by default you will lose.

The sign of any successful company is the ability to pivot. Companies that cannot shake their past to become something new are unlikely to last as long or grow as large as a company like MS has.

 

[bgroovy]

The point of innovation is to break boundaries and change the game. If you’re doing neither you aren’t bringing anything to the table.

Not every market wants or needs that level of innovation. Mandatory Kinect was bad innovation. A Share Button was good innovation. One is now the industry standard. One has been relegated to the industry dustbin. If Microsoft wants to innovate themselves to a distant third place again, they are certainly free to do so. But I'm tired of being lectured by people about how MS has changed and really cares about console gaming now when Spencer is literally telling us the limits of that commitment, and where the strings are attached.

 

[mrcorbo]

You know what it tells me? MS is constitutionally incapable of focusing on console gaming as a goal in and of itself. It is the same mistake they have made time and time again. It torpedoed the design of the Xbox One and it's sales potential. And now they are planning to make the same mistakes, in a different way next gen. It's the same thing I have warned people about in the past, but everyone wants to believe things are different under Spencer and Nadella. This makes it pretty clear that is not the case. The greater company only cares about Xbox in as far as it can be leveraged for something else. In the past this has always resulted in a compromised experience for console gamers compared to platforms that do not see their hardware as a stepping stone to something more lucrative.

Well, this is how I see it.

Allowing the hardware that will be powering instances on Project xCloud to also be utilized for other cloud computing tasks will allow them to massively over-provision the initial server rollout for the Project xCloud service without having to worry about that hardware sitting idle while the subscriber base is still low. This should guarantee the best possible experience for this service from day one and will continue to allow for aggressive expansion of the server resources available since MS have minimized the downside risk of those investments.

This also means that when MS solicits bids from suppliers, technology partners and manufacturing facilities the contracts they will be offering have the potential to be more lucrative than if they were for the design and manufacture of just a console design. This should allow them to secure more favorable terms in these contracts as the larger scale will allow for partners to make more money over a comparable period of time even with lower per-unit revenue.

Finally, MS have been able to invest so massively in producing games because those games will effectively be multi-platform across Xbox consoles, PC, and Project xCloud.

MS have never demonstrated a stronger commitment to gaming at any time in their history then they are right now. And this commitment is enabled by them not just making a new console.

 

[eastmen]

Not every market wants or needs that level of innovation. Mandatory Kinect was bad innovation. A Share Button was good innovation. One is now the industry standard. One has been relegated to the industry dustbin. If Microsoft wants to innovate themselves to a distant third place again, they are certainly free to do so. But I'm tired of being lectured by people about how MS has changed and really cares about console gaming now when Spencer is literally telling us the limits of that commitment, and where the strings are attached.

I think your very wrong. What has share buttons on it ? The ps4 ?

Voice control as in the mandatory Kinect is every wehre , cable companies have it in their boxes , phones have it , alexa enabled devices are selling like hot cakes.

Video cameras are every wehre of course but everyone is now using them to stream .

Kinect like cameras are used to log into windows devices , iphones , android phones and so much more . Kinect like tracking is used in vr and ar and even in some games for facial tracking like star citizen.

If you want to say mandatory Kinect was done at the wrong time I can get behind you. However it wasn't a bad innovation and its something I wish would come back next generation .

I rather a company take chances and fail than a company to play it safe. I was playing online with my sega Saturn and having a blast . Then they pushed and made their own isp with dreamcast .

 

[Tkumpathenurple]

I wonder how similar the server hardware will be to the home console hardware. And if they're similar, I wonder if Microsoft will be cheeky and utilise idle consoles for its own purposes? Running some ML algorithm, for example.

It's reminded me of something I'd like to see come back to consoles though: folding at home. It was on the PS3, and I didn't use it an awful lot, but it was cool to start it up every now and then, overnight, knowing that I was helping the scientific community just by leaving my console running.

And with it being winter right now, I could do with a console kicking out heat all night.

I was playing online with my sega Saturn and having a blast . Then they pushed and made their own isp with dreamcast .

You could play online with the Saturn?!?

 

[turkey]

You could play online with the Saturn?!?

https://en.m.wikipedia.org/wiki/Sega_Net_Link

I think the Daytona netlink release is one of the top collectors items for the Saturn.

 

[ultragpu]

Don't we get to take off some AMD flop penalty for better console flop usage.

Nar mate, console flop usage is just a buff that can be stacked on either NV flop or AMD flop, you thus become super saiyan flop.

 

[Jay]

Allowing the hardware that will be powering instances on Project xCloud to also be utilized for other cloud computing tasks will allow them to massively over-provision the initial server rollout for the Project xCloud service without having to worry about that hardware sitting idle while the subscriber base is still low. This should guarantee the best possible experience for this service from day one and will continue to allow for aggressive expansion of the server resources available since MS have minimized the downside risk of those investments.

A lot of that feeds into why when xcloud was announced and said it was based on 1S hardware I said I suspect that it means Scarlett apu was far from being ready, hence next Xbox being 2020 at the earliest.
I think that they will do a very limited test and roll out of the current xcloud.
They just needed to show they was in the streaming game.

With Scarlett apu's in the cloud they can probably run multiple XO's on one apu, and now use it for other workloads (which I considered but had no basis to throw it out there), and get a solid base performance and resolution prior to compression that you won't get with the 1S.

It's also one of the other reasons I think there's a higher chance of MS being chipplet and possibly mcm based. Putting the apu's in both consoles and racks will help spread the costs and even more bining options of the apu's.

All this could have a very good impact on the cost to perf of the console, or even what becomes viable in a console.

Anyone thinking this shows less commitment to console gaming, just refuses to see the benefits of spreading the costs, especially when you consider everyone else will be doing streaming also.

 

[mrcorbo]

I wonder how similar the server hardware will be to the home console hardware. And if they're similar, I wonder if Microsoft will be cheeky and utilise idle consoles for its own purposes? Running some ML algorithm, for example.

EU power regulations say no. The home user could opt-in FaH-style, but that removes the cheekiness,no? Besides, there's no need for hardware parity to make this work. Any connected console with compute resources could be used this way with the proper software running.