Baseless Next Generation Rumors with no Technical Merits [pre E3 2019] *spawn*

Status
Not open for further replies.
Same perf doesn't mean same tfops
True enough. Or perhaps they're just harnessing the full benefit of 7nm and bumping up the ever living soul of the clocks ever more?
* Also video states Navi is the final entry to the GCN architecture along with 7nm+ and the limit is still capped at 64 CUs. Basically upping the clocks is the only way to raise performance no? I highly doubt they'll increase more shader operations per cycle.
 
Last edited:
Yes, I know it's wildly open to interpretation it's just the way information flooded onto my monitor and the content description sure made it sounded it's ultra high end:). I'm very susceptible to such way of marketing.
:LOL:

I'm trying to be careful with my words because I'm far from calling Cerney a hype man, he was very precise with what he said and doesn't sound like he's trying to pull the wool over people's eyes.
The albeit appropriate buzzwords and no actual specs does make it sound super crazy powerful I guess. $10k RT hardware and all that (not Cerney, interviewer), 8k...
 
Most of the info in that Wired article didn't come first hand from Cerny, but second hand from the author. I think everyone, me included, took the words as being out of Cerny's mouth, but if you look only at what he said exactly, there's very little info. The rest is reading between the lines beyond the explicit points like Zen2, 7nm, and 3D audio customisation. eg. The only mention of 'SSD' was from the author, and Cerny never says there's an SSD in PS5. He talks about an SSD in his laptop. He talks about putting an SSD in a PS4P. He says what they have in PS5 is faster than a PC. He never says, verbatim, there's an SSD in PS5.

I think a lot of people have heard what they wanted to hear because it's suitably vague. There's too much ambiguity to trust we have a good idea of the final product.
 
Yes and no. Take a typical 2.5" SSD for example. To access the flash the computer has to go through OS calls, device drivers, southbridge, the physical drive interface electronics, the drive controller on the SSD itself then the flash. Now look at what Apple does with flash storage in Macs and iPhone: OS calls, device driver, custom controller to flash. Android phones vary depending on the cost and chipsets.
so overall, if we're talking about making tweaks to performance, within context of PS5, breaking out the SSD isn't enough. We're looking at something akin to PCIE4.0/ NVMe type differences, or something else entirely like loading a ton of stuff into memory. Major differences in which we can access the data - but not necessarily through just traditional I/O means.
 
Most of the info in that Wired article didn't come first hand from Cerny, but second hand from the author. I think everyone, me included, took the words as being out of Cerny's mouth, but if you look only at what he said exactly, there's very little info. The rest is reading between the lines beyond the explicit points like Zen2, 7nm, and 3D audio customisation. eg. The only mention of 'SSD' was from the author, and Cerny never says there's an SSD in PS5. He talks about an SSD in his laptop. He talks about putting an SSD in a PS4P. He says what they have in PS5 is faster than a PC. He never says, verbatim, there's an SSD in PS5.

I think a lot of people have heard what they wanted to hear because it's suitably vague. There's too much ambiguity to trust we have a good idea of the final product.
yea while I love the idea that they can speed up say fast travel times.
something to note here: The Division 2, fast travel loading is faster in the same zone. Once you fast travel to another zone, it's a completely different load time.

Usually when we measure loading time performance, we're talking about booting a program up from cold. That gives us a way to put some consistency behind the experiment and we are directly measuring the performance of how quickly the I/O is working, the hard disk as well as the CPU.

So we don't really know what Cerny accomplished by showing fast travel, it's not necessarily the same as a cold boot.
 
so overall, if we're talking about making tweaks to performance, within context of PS5, breaking out the SSD isn't enough. We're looking at something akin to PCIE4.0/ NVMe type differences, or something else entirely like loading a ton of stuff into memory. Major differences in which we can access the data - but not necessarily through just traditional I/O means.

I seriously don't think we should take the SSD part as literal as a tech guy would. The implied performance would cost far too much and require a certain size. It's simply not economical. So he might just meant the theoretical bandwidth of the interface or that they allow >256M(Desktop "limit") for people to change the NVMe card. Then they can say it's more than AMD Desktop PCs:)

IMHO they will provide a 32-256GB SSD cache with AMD StoreMI like Intel Optane around 1.xGB/s. Can most games even use more performance efficiently as they need to process the data in general which takes time?

Edit: Removed some past artefact. Damn text caching:)
 
Last edited:
If navi 20 has RTRT but the rest don't then that could mean a couple things, not specifically hardware based but navi 20 is beefy enough to run it through standard compute.
Or
Hardware optimised which could mean Sony asked for it to be included same way they got RPM from vaga included into 4pro. Which may mean it didn't need to be as powerful.
So that doesn't actually tell us anything about power as we don't know about navi yet even as a basis.

2TB SSD would definitely add a lot to cost. If that's what you think it will have. Not that it was said, but fair that's what you think.

Support for 8k graphics just means it supports hdmi 2.1, it doesn't give any indication of power at all, if every navi gpu doesn't support hdmi 2.1 I'd be very surprised. Did he say what level of graphics it could do at 8k, he didn't even say what it could do at 4k.
So 8k graphics could be very cheap because it doesn't mean anything by itself apart from hdmi spec cable.

The only thing that was demoed was spiderman. Was that at 4k60, 4k120, 4k at all? We don't even know that, what was demoed was not really about the power of the console but about its data streaming capabilities.
Which their very proud of, and maybe for good reason.

You say this based on not knowing anything about how powerful the ps5 is though?
Simple example how many TFlops is it?
Speed of CPU?

Whilst I'm happy for Cerny's interview, I think people are putting words in his mouth.
Most of what he said wasn't even ambiguous, which is nice. But he never gave any specs at all.

I'm speaking to the 8 core zen 3 , Navi + HBM ram. What else does AMD have coming down the pipe that would fit in a 2019 launch or 2020 launch for MS that could be more powerful ? Either MS uses the same stuff and ups the clocks , paid for a custom gpu or put out enough money to allow AMD to move forward their next gen gpu .
 
Most of the info in that Wired article didn't come first hand from Cerny, but second hand from the author. I think everyone, me included, took the words as being out of Cerny's mouth, but if you look only at what he said exactly, there's very little info. The rest is reading between the lines beyond the explicit points like Zen2, 7nm, and 3D audio customisation. eg. The only mention of 'SSD' was from the author, and Cerny never says there's an SSD in PS5. He talks about an SSD in his laptop. He talks about putting an SSD in a PS4P. He says what they have in PS5 is faster than a PC. He never says, verbatim, there's an SSD in PS5.

I think a lot of people have heard what they wanted to hear because it's suitably vague. There's too much ambiguity to trust we have a good idea of the final product.

The article is based on a much larger interview, and likely vetted by Sony for accuracy. It's not that "vague" unless you want to assume incompetence in the author or malfeasance from the source. When people tried to act like Cerny "only actually talked about ray tracing for audio" the reporter made it clear he also talked a lot about visual ray tracing in the interview, he just didn't include those direct quotes.
 
I'm speaking to the 8 core zen 3 , Navi + HBM ram. What else does AMD have coming down the pipe that would fit in a 2019 launch or 2020 launch for MS that could be more powerful ? Either MS uses the same stuff and ups the clocks , paid for a custom gpu or put out enough money to allow AMD to move forward their next gen gpu .
https://www.mcvuk.com/microsoft-cloud-makes-xbox-one-four-times-more-powerful/

A possible way to claim "more powerful", so this is why "source" told the insider definitely that next-gen x1 is most "powerful".
 
Well, as ironic the cloud power comment was I wouldn't be that surprised if their design is a downscaled byproduct of their Azure collaboration with AMD. Project Olympus probably doesn't just mean some EPYC CPU but also future GPU blocks.

If they are really ambitious they might even consider extending their Azure cloud on future consoles based on some license agreement like get free XBox Live/Points if you agree to share your console if it's idling.

Obviously I'm just speculating here:)
 
Well, as ironic the cloud power comment was I wouldn't be that surprised if their design is a downscaled byproduct of their Azure collaboration with AMD. Project Olympus probably doesn't just mean some EPYC CPU but also future GPU blocks.

If they are really ambitious they might even consider extending their Azure cloud on future consoles based on some license agreement like get free XBox Live/Points if you agree to share your console if it's idling.

Obviously I'm just speculating here:)
I’ve been on this wavelength for a while for some reason. Probability is low, but I don’t want to scratch it out just yet. They made mention that Scarlett development team was getting input from the azure team and that’s never really left my mind since.
 
I’ve been on this wavelength for a while for some reason. Probability is low, but I don’t want to scratch it out just yet. They made mention that Scarlett development team was getting input from the azure team and that’s never really left my mind since.

Like the boy who cried wolf, Microsoft will need to manage any communication relating to 'cloud powered' very carefully since they did that at the start of the Xbox One generation and we know how that turned out.
 
Like the boy who cried wolf, Microsoft will need to manage any communication relating to 'cloud powered' very carefully since they did that at the start of the Xbox One generation and we know how that turned out.
yea, I get that part, but the way they're talking about Azure isn't so much in the same context that they tried with XBO. That was weak attempt at trying to use the cloud as this augment for a lack of graphical performance.

The azure discussion seems based more around setting the soc up for datacenter usage, ie they want to be able to stream xbox games to clients, but when it's not being used they want the idle compute for their azure business.

I would believe this scenario much more than cloud augmentation of games because it decreases their risk and cost in data centre builds.
TLDR; it's not something they really should be selling, or marketing.

still referencing this quote
“The thing that’s interesting for us as we roll forward, is we’re actually 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.

We have a consumer use for that silicon, and we have enterprise use for those blades as well. It all in our space around driving down the cost to serve. Your cost to serve is made up by two things, how much was the hardware, and how much time does that hardware monetize.

So if we can monetize that hardware over more cycles in the 24 hours through game streaming and other things that need CPU and GPU in the cloud, we will drive down the cost to serve in our services. So the design as we move forward is done hand-in-hand with the Azure silicon team, and I think that creates a real competitive advantage.”


interview: https://twinfinite.net/2018/12/phil-spencer-buy-ea-next-gen/

This is part of why I think it may not be navi. I think internally this is a bigger factor for MS then competing with Sony.

Stadia chose to run with Vega. And that's a pure cloud based streaming service. And so as per my commentary on the other thread, implying MS has something different from Navi doesn't imply they're getting a higher performing console; Sony chose Navi perhaps because it was best for console, but Google chose Vega, perhaps that was best for pure server builds.

Perhaps MS is somewhere in-between or something entirely custom because the product they want doesn't quite exist.
@Jay
 
Last edited:
My baseless speculation is that the xbox design will be chiplet based, it seems that approach maybe more flexible for scaling the silicon across consoles and server.
 
My baseless speculation is that the xbox design will be chiplet based, it seems that approach maybe more flexible for scaling the silicon across consoles and server.
a good point i didn't consider. Are you thinking like, flexible in terms of binning the chips? Or looking to serve a variety of different configurations by mixing different chiplets ?
 
But do any of the chiplet orders come with queso or salsa?


I would love for the nextgen consoles to be user-upgradable beyond just the storage.. nice modular pieces parts. Oh, you want the midgen cpu and gpu upgrade? Just slide out the main board model and slide in this upgrade board. All similiar to older style laptop modules.
 
Status
Not open for further replies.
Back
Top