Next Generation Hardware Speculation with a Technical Spin [2018]

[turkey]

That is supposed to be a PowerPoint?

Who makes a wall of text like that and with a black background? You just could not present that to a room, and it would look garbage printed.

We have seen Sony internal PPT and they look nothing like that. Ignoring the content it just looks fake, too much detail on the please don't leak me info.

 

[Deleted member 11852]

That is supposed to be a PowerPoint?

It looks as though it's supposed to be a draft press release but it's obviously fake. Firstly, it looks nothing like any of Sony's PlayStation press releases and secondly, rookie mistake is the use of PlayStation tm (trademark) instead of PlayStation(R) (registered trademark). Sony insist on PlayStation(R) universally everywhere, on hardware in the OS, in games ("Please don't turn off your PlayStation(R) 4 system while..")

But it's a fair fake. They've even mocked up a pig-ugly box the you could imagine Sony selling!

 

[vipa899]

Also hw specs two years in advance?

 

[Xbat]

I still feel 2019 is a distinct possibility.

 

[chris1515]

I still feel 2019 is a distinct possibility.

Jason Schreier from Kotaku told he is sure at 90% 2019 is not the year fr PS5 because people in the industry said no sign of PS5 in 2019 by Sony gives to third party at all. It would be Apple Job's level of secrecy to hide the PS5 launch in 2019.

https://www.resetera.com/posts/14715382/

I am 90% sure, based on conversations with a fair number of developers, that the PS5 is scheduled for 2020. The 10% is because I haven't yet seen or heard concrete plans for it, and because with hardware launches, anything can change. (It might even wind up coming out later!)

 

[KeanuReeves]

All good everyone. She gets better, remarkably according to the doctors, every day. Just going to be slow going.

I was just making sure that no one thought I was throwing shit at a wall or trying to stir the pot or something. Felt like something I needed to do since it may be a bit before I can ferret out the video clip.

In the video "How nvidia won the GPU war" or whatever it is called, he spoke about AMD being way further ahead on i believe 40nm (?) you may have caught that in your brain and it accidentally merged into the talk of the modern situation.
But who knows. Maybe AMD is like light years ahead of Nvidia in the raytracing space, but we don't have anything to suggest that as of now. And another thing, it would appear that he does have some sources on certain stuff so maybe in the future, if he relays information like that, we would have something interesting to talk about.

 

[Jay]

I still feel 2019 is a distinct possibility.

After seeing the results of vega on 7nm, I'm more inclined to believe it's 2020 now.

I know navi isn't vega, but 2019 just seems to soon to me now.
Consoles generally don't go with cutting edge manufacturing, and 7nm for gpu is very new.

I had started to come around to Sony being 2019, due to all the talk, even though my mind was saying no, the heart was yes

 

[Xbat]

For me it's only where they feel they are software wise if they launch 2019 and what they think the competitions going to be doing. I don't think hardware availability is going to play a big part in it.

 

[iroboto]

For me it's only where they feel they are software wise if they launch 2019 and what they think the competitions going to be doing. I don't think hardware availability is going to play a big part in it.

As much as I want to believe 'what the competition' is doing is a big thing, I've only ever seen it happen at a promotions/sales type level where you can be that reactionary.

These types of launches are not common, so you really have to look at it as the whole company coming together to get this out as a unified front, and there are a ton of moving pieces. The idea of 'its ready' when it's ready, is a big thing. There can be crunch time for your team if you're close to your competitor, but if you were planning a full year out, you aren't going to advance your plans like that.

We can write it from the perspective that MS has nothing to lose, and they can drag this generation out as long as possible to build good will for next gen, thus they expect Sony to announce first, or launch in late 2021/2022. You can write the same thing for Sony though, that they already won this generation, and the money keeps flowing in. and they'll instead spend their time building up their services and architectures to match MS offerings for next gen, so they can drag their heels this gen raking in the cash while they are developing it.

All in all, looking at the competition, probably won't serve us any good here.

 

[Jay]

There's a big cost/risk going with a new process.
Including things like yield, which for a cost constrained product is a big deal.
On the pc side, can charge more of a premium, or in the case of the new vega sell it to workstation class customers

 

[iroboto]

I know there is a big push to just look strictly at AMD.
But don't count Intel and Nvidia out of the running yet.
From a strategic perspective, all three benefit in advancing their agendas by winning these console contracts for next gen.

 

[Deleted member 13524]

3640 shader cores, can't divide this with 64 to get an CU number. Someone put so much effort in this fake, but doesn't have basic architectural knowledge making such a fault.

What do you mean we can't have 56.875 CUs??
.

I know there is a big push to just look strictly at AMD.
But don't count Intel and Nvidia out of the running yet.
From a strategic perspective, all three benefit in advancing their agendas by winning these console contracts for next gen.

Lisa Su sure has very strongly hinted at having won the design for at least one of the new consoles (most probably Sony).

And all three benefit, but none of them benefit as much as AMD, which is why AMD is willing to charge less for semi-custom solutions. I read somewhere that Subor paid AMD something like $80 million for their SoC.

 

[OCASM]

Isn't "crippling" a bit excessive? Something like 10-12TF of performance seems a reasonable expectation, based on assumed die sizes for a console and the known die size of the M160. We'll have to see what performance it reaches at lower clockspeeds before we have a clear indication of an approximation of performance at console wattage, but the current performance of 14.8TF for a GPU that was meant as a 7nm pipe cleaner bodes well IMO. We'll have to wait and see, but hopefully there's truth to the reports of some Zen engineers moving over to work on Navi, and hopefully that pays off in the form of improved perf/watt.

But let's be conservative and assume a 10TF GPU, which can have any ratio of performance dedicated to rasterisation, ML, and RT. Would that flexibility really hamper it so much?

I can imagine that all 3 might be somewhat compromised, relative to fixed function hardware, but does the penalty to performance still result in a functional RTRT GPU? To my knowledge, that's an unknown.

Does the penalty to the size of the CU's outweigh just having fixed function hardware? The M160 suggests that AMD might think so.

And if the answer to both of those is yes, does it make it any cheaper to manufacture? Potentially so, if AMD are able to just deploy different amounts of differently clocked multi function CU's to satisfy all of their markets.

RTRT is still nascent in gaming, but it can't hurt to have the entire core gaming industry beavering away on solutions to it. It'll just be the way it always is: lower quality on the consoles.

And if any developers don't want to, that's fine, they can act as if they have a 10TF PS4/XB1 with plenty more memory and a way better CPU.

Edit: I also think it's worth mentioning that Sony's new CEO mentioned ease of manufacturing as key for the PS5 - I'll try to find the source. Considering "ease of manufacturing," AMD's recent chiplet Zen 2 design, their continued push for Infinity Fabric, and their multi purpose CU's, I'm convinced that, if the CU's can be modified to be capable of RT, that's the approach AMD would want to take regardless of developer convenience.

Most games don't make use of hardware flexibility. Pretty much all of them use the same technologies and that's even more true now with the move to PBR and the convergence to a handful of engines across the industry. Speed benefits the vast majority of games versus flexibility for just a few.

 

[iroboto]

What do you mean we can't have 56.875 CUs??
.



Lisa Su sure has very strongly hinted at having won the design for at least one of the new consoles (most probably Sony).

And all three benefit, but none of them benefit as much as AMD, which is why AMD is willing to charge less for semi-custom solutions. I read somewhere that Subor paid AMD something like $80 million for their SoC.

I agree. Just probabilities aren't 0.. yet
The flop is shown, but there is still the turn and river left.

 

[Jay]

I know there is a big push to just look strictly at AMD.
But don't count Intel and Nvidia out of the running yet.
From a strategic perspective, all three benefit in advancing their agendas by winning these console contracts for next gen.

Intel is struggling with getting to grips with 10nm, although doesn't mean they couldn't have won the design, and their discreet gpu is a while away. Their igpu isn't good enough, one of the reasons they used vega also.

NVidia, well, don't see that for many different reasons.

Even if either didn't go AMD, don't see it being before 2020 unless willing to risk being first product at 7nm, and i believe console manufactures aren't willing to do that, guess could go 14nm....

 

[iroboto]

Intel is struggling with getting to grips with 10nm, although doesn't mean they couldn't have won the design, and their discreet gpu is a while away. Their igpu isn't good enough, one of the reasons they used vega also.

NVidia, well, don't see that for many different reasons.

Even if either didn't go AMD, don't see it being before 2020 unless willing to risk being first product at 7nm, and i believe console manufactures aren't willing to do that, guess could go 14nm....

It's not cheap to get into the GPU market. They need a 'win'/marketing boost. They have very good adherence to the APIs on the market, they definitely were among the quickest to meet API feature sets. They need a win of some sort and console victory here could get their foot into the door. The ML market will only continue to increase in size. I think failure to get into this space quickly will be detrimental for their future. If they took a similar path like Google, just creating Tensor Flow and tensor core designs, I wouldn't have brought them up. But they got/going into the GPU market instead and not just iGPU, but the discreet market.

For nvidia. They need a win to push their RTX agenda as the chips are not all that much more powerful than the 1080TI line. I think this is obvious.

AMD is still the strongest case of course. And it's looking good right now. But nothing is yet guaranteed.

 

[Metal_Spirit]

A well made joke, or with any truth to it?

Well written, well presented! But 3640 stream processors? Thats 56,8 CUs! Typo and it was meant to be 3840, or just fake?
The worst point seems to be the 14 Gb of 16 available to developers meaning 2 GB reserved for OS. PS4 has 3.5!

It seems fake, but it´s well done!

 

[Shifty Geezer]

Has anything changed for the nVidia option regards CPU provision? nVidia GPU either means 1) nVidia GPU + Intel CPU, 2) nVidia GPU + AMD CPU, or 3) nVidia GPU + nVidia ARM CPU.

1) Expensive
2) Would likely get better deal with all AMD
3) Whole issue of ARM's viability as discussed at length

If MS get Windows running as quickly on ARM as on x64, it's a realistic option for MS. However, they want cross-device compatible with x64 PC more than anything, so as long as ARM performance is going to interfere with that, the next XBox has to be x64, which pretty much rules nVidia out.

 

[Shifty Geezer]

A well made joke, or with any truth to it?

Well written, well presented! But 3640 stream processors? Thats 56,8 CUs! Typo and it was meant to be 3840, or just fake?
The worst point seems to be the 14 Gb of 16 available to developers meaning 2 GB reserved for OS. PS4 has 3.5!

It seems fake, but it´s well done!

Already posted. Already poo-pooed.

 

[Jay]

For nvidia. They need a win to push their RTX agenda as the chips are not all that much more powerful than the 1080TI line. I think this is obvious.

NVidia really doesn't need console to push RTX, you only have to look at gamesworks to see that.

If MS get Windows running as quickly on ARM as on x64, it's a realistic option for MS.

MS already has win10 running native on ARM, including X86 emulation. The biggest issue would be BC, cpu emulation wouldn't be fast enough, would mean they would have to recompile which I don't see happening.
Would've been interesting if didn't have to worry about BC, what an ARM based cpu could've done.