Baseless Next Generation Rumors with no Technical Merits [post E3 2019, pre GDC 2020] [XBSX, PS5]

Status
Not open for further replies.
New rumor for future game development, source: me
Game developers will make games with all the graphical effects, AI physics etc they want, no need to think of hardware. User buy the game and load it into his/her gaming machine. Go take a cup of coffe. Machine learning will analyse and adjust to best possible performance on the hardware with focus on eye candy, fps etc. User get a list of what he/she prefer, make your choices and the options that will not have enough available power will be greyed, start game.;)
No AI required - MS will process your game with Simplygon running on Azure servers. (how can i make nodding smiley?) Lockhart version smaller download as well.

Gamedevs focusing on scalable tech instead maxing out raw brute force power is not necessarily a bad thing. Could even cause better software progress than 20TF monsters.
 
And in the situation everything is "scaling up" from lower spec, not scaling down.
We could discuss this but even if was scaling up or down not to much of a big deal unless you think their going to auther textures for 1080p / 1440p and not 4k.
It's the CPU and ssd speeds that makes the bigger difference when scaling up or down. In regards to scaling all the way back to XO.

Probably worth saying I give no credence in any report that says Lockhart has half the CPU cores etc. If that's the case then everything that is being said about scaling becomes a huge problem and would make Lockhart redundant. Graphic engines especially considering GPUs will be same architecture are very scalable already.
 
Last edited:
Not releasing lockhart would be big blunder from MS, considering they made it clear all their games will be released on PC anyway, therefore will have to work over dozens of different specs anyway.
 
Not releasing lockhart would be big blunder from MS, considering they made it clear all their games will be released on PC anyway, therefore will have to work over dozens of different specs anyway.
Can raise the minimum spec of pc games, but can't raise minimum spec of fixed XO though.

That's why Lockhart makes sense, if they was simply going to force current gen support then Lockhart wouldn't have a place.
Decent CPU and ssd for a console baseline, whilst not only having an accessible console for start of gen, but getting down to the even lower price points in couple years later is a very reasonable approach in my view.
 
This is not great, if true. Most games are cross-platform and if Microsoft mandate all future Xbox games must run on both configurations that low-ball configuration will hold back both Xbox and PlayStation cross-platform games. I suspect a piece of the puzzle of missing.

Right, this certainly explains why most of the next gen games on X360/PS3 looked far better than Crysis. Crysis was held back by having to support hardware much weaker than both the X360 and PS3.

Wait, that didn't happen? Ooops. :)

OK, maybe they were all better than Crysis: Warhead which pared back some things to support even lower spec'd hardware configurations than Crysis. Oh wait, that didn't happen either?

Sure on a similarly spec'd hardware configuration the settings used in Crysis may have brought it down to a similar level as consoles, but it certainly wasn't something holding back the game.

Control has to support and run on worse hardware than either the base PS4 or XBO, is that holding it back from using graphics that are quite significantly higher than either while also using hardware RT acceleration?

Or perhaps with most developers now focusing so much on console hardware first when making games, they've completely forgotten how to scale games across a wide variety of hardware while simultaneously pushing what is possible in games. IE - Lazy dev. syndrome.

So this OsirisBlack dude who accurately leaked PS4 Pro is back again with some new cryptic wordings as the following:

PS5 stronger than XB2
The weather is a violent windy storm as in tempest referring to The Tempest (Ariel, Gonzalo, Propero) Prospero is PS5 devkit
PS5's GPU clock is 2.1Ghz

From this.

5 > 2
How's the weather?
2.1 = 5

If you're going to use 5 > 2 to mean that PS5 is greater than XB2. Then considering the rumors floating around XB2 is most likely to be Lockhart. 2.1 would most likely be Anaconda.

Then again, all of this is just someone likely trying to feel self important. Especially when we have established journalists making claims that developers are frustrated at MS because they haven't received Project Scarlett devkits yet.

Regards,
SB
 
Last edited:
Some discussions state that current PS5 dev kits are just a 5700XT-based GPU with high frequency,bacause bigger Navi GPU is not ready yet.
Final silicon will use bigger GPU and lower frequency for more reasonable power dissipation.

Do you think this assumption is possible?


If the above statement is true then how do the dev kits perform ray-tracing? Do the dev kits have the same performance of final silicon on ray-tracing?


Furthermore it reminds me that Wired article in April said the current devkit is a "low-speed" version. Why the SSD is "low-speed"? Is the SSD still lack of something important? Or the final SSD may work with other things such as ReRAM?
 
Some discussions state that current PS5 dev kits are just a 5700XT-based GPU with high frequency,bacause bigger Navi GPU is not ready yet.
Final silicon will use bigger GPU and lower frequency for more reasonable power dissipation.

Do you think this assumption is possible?


If the above statement is true then how do the dev kits perform ray-tracing? Do the dev kits have the same performance of final silicon on ray-tracing?


Furthermore it reminds me that Wired article in April said the current devkit is a "low-speed" version. Why the SSD is "low-speed"? Is the SSD still lack of something important? Or the final SSD may work with other things such as ReRAM?

It can be multiple thigs if the april kit was not using a Zen 2 CPU maybe the bus was only PCIE 3 and in final version it will be PCIE 4 for example.
 
If you're going to use 5 > 2 to mean that PS5 is greater than XB2. Then considering the rumors floating around XB2 is most likely to be Lockhart. 2.1 would most likely be Anaconda.

Then again, all of this is just someone likely trying to feel self important. Especially when we have established journalists making claims that developers are frustrated at MS because they haven't received Project Scarlett devkits yet.

Regards,
SB

I don’t believe for a second that developers haven’t gotten a dev kit before Phil gets to publicly take one home.

Osiris has since cryptically clarified that the PS5 is more powerful then Scarlett so he wasn’t talking about Lockhart. I don’t know how much stock I put into a lot of the rumours but I am certainly curious to find out why the PS5 dev kit has the design it does, the cooling seems way overkill.
 
It's to appease the cats of the developers. They demanded something to curl into.
 
Come on, I think it's about time some of our B3D insiders give us a Hmm, Huh or smiley to one of those quotes :). First who does it gets a cookie from me.
 
I think it is better to wait after E3 2020 for "real" information. Actual there is so much crap, disinformation, Fud and Fanboyism around the Internet, that i become sick. So much rubbish information contradict each other in a Cointelpro manner.
 
They try to keep their xbox1 market share as oppose to starting a new generation with zero market share..
If MS's next-gen looks like current gen, its dead already.

Or perhaps with most developers now focusing so much on console hardware first when making games, they've completely forgotten how to scale games across a wide variety of hardware while simultaneously pushing what is possible in games. IE - Lazy dev. syndrome.
Scaling visuals is fine. A new gen means scaling gameplay too (potentially). 10 TFs means 6 TFs for visuals and 4 TFs for world processing in some way, if you want. Between the large amount of floppage and improvements in tech, there should be better opportunity to turn more towards the world rather than the pretties.

It's probably not a big issue especially for 3rd parties making cross-plat games, but some devs expecting a nice, clean break and significant advance for a new gen would disappointed with a half-gen step. And MS's first parties would probably be looking enviously at Sony's targeting 10 TFs and seeing they can't go that far and will always be hampered by hardware targets when it comes to winning best tech and best graphics awards.

How many games fundamentally improve on more powerful hardware beyond just having more pretties?
 
In fact I think xbox1 may become extremely cheap entry model (199 with low-end SSD), and xbox1x is replaced by lockhart.
Does Xbox One X cost more than $200 to produce? If so, that won't happen. It is quite conceivable that by next Fall Microsoft can sell a 'low-end' Xbox, using a technical platform common to their premium product, offering more performance than Xbox One X but costing less than manufacture. If this is the case, there is no reason to continue making or selling Xbox One X.
 
If MS's next-gen looks like current gen, its dead already.

Scaling visuals is fine. A new gen means scaling gameplay too (potentially). 10 TFs means 6 TFs for visuals and 4 TFs for world processing in some way, if you want. Between the large amount of floppage and improvements in tech, there should be better opportunity to turn more towards the world rather than the pretties.

It's probably not a big issue especially for 3rd parties making cross-plat games, but some devs expecting a nice, clean break and significant advance for a new gen would disappointed with a half-gen step. And MS's first parties would probably be looking enviously at Sony's targeting 10 TFs and seeing they can't go that far and will always be hampered by hardware targets when it comes to winning best tech and best graphics awards.

How many games fundamentally improve on more powerful hardware beyond just having more pretties?
I have my reservation regarding this for two reasons :

1) Lockhart or no Lockhart, MS's strat has already been set to stone 2 years ago with inclusion of game pass and "play everywhere".

This basically means, as Phil said many times, you will be able to play their games on top end console or vast amount of other devices (Xcloud - which will not be 4K highest settings, and plethora of PC's). I am 100% sure they put alot of time and effort into making game engines and their own development tools as scalable as possible.

2) Tech is on fundamentally same level.

Both Lockhart and Anaconda will provide same base architecture and technology (as well as PS5), but clear distinction will be made as to which resolution they target. We know for fact that with every generation, consoles have sweet spot they are shooting at in terms of resolution. For 360 and PS3 it was 720p, which was more or less hit for 30fps games (10% +/-). With PS4/XBO it was 900p/1080p.

For PS5 and Anaconda, I expect it to be 4K, or in certain cases middle between 2K and 4K with good up-scaling. Everyone on this forum is aware that by far biggest performance hit comes with increase of resolution, therefore a console which is targeting 4K, and provides 10TF GPU, will reserve alot of rendering budget for increase in resolution, vs one that targets 1080p/1440p and has 4-5TF.

So from that stand point, I don't see what console with 10TF Navi, 8 core Zen2 and 16GB of usable RAM can do, that console with 4-5TF Navi, same 8 core Zen2 and 12GB of usable RAM can't at 1/3rd of resolution.

People are mentioning PS2/Xbox and Wii/360/PS3 years, but that is completely misplaced comparison. Not only where tools back then archaic (PS2 documentation was bizzare even for early 00's), compared to what we have now, but difference in architecture and technology of individual consoles where gigantic.

In MS case you would have 2 consoles from same manufacturer, with same tools and exact same design. Only one would have 1/3rd of GPU performance, but would also be shooting for 1/3rd of resolution. In the day and age we have game engine scaling across top end PCs running RT effects, to mobile phones, saying this will result in development headache's is reaching IMO.

My final point would be regarding your comment about Sony's first party enjoying pushing one single console to the limit while, MS studios have to fiddle around with two, resulting in big difference in visual quality. Don't agree with that either for several reasons :

1) MS studios are already targeting more devices then it would be the case in potential Anaconda and Lockhart scenario. They are targeting everything from X to S and One, and to every single PC there is (along with future Xcloud)

2) 3rd party games this gen have come closer to 1st party then it was ever the case. Some of the best looking and running games this gen have been 3rd party (RDR2, Doom, RE2, DMC, COD MW etc.) and difference between 1st party and 3rd party is much more in terms of asset quality and art direction, then it is in tech.

3) Scalability is becoming more important then hard generation resets. This is because IMO we are seeing graphics being more then good enough for most of the time, and differences between great looking games is minimal (to a point where art style play bigger role). All that while tech is no longer advancing at rate it was in early 2000s.

So my point is, MS not releasing Lockhart would be huge, huge mistake. For MS, they already have bunch of Lockharts, with even bigger and wider gap between top and the bottom, so not releasing something like Lockhart with GP to counter Sony's one console strat would be a missed opportunity.

Lockhart will give people an option to play next gen games at lower resolution. If Sony pushes their exclusives in sub 4K territory, MS can easily do the same and still lower Lockhart to 1080p, not like many people buying ~$299 console will care about if its output is native 1440p or 1080p.

We are in entering the age of convergence and MS is smartly reacting to that. I can see Sony doing the same with Ryan at helm. Starting with PSN Now to racks of "lower end console" that acts like streaming platform for millions of people not caring about 4K.
 
So from that stand point, I don't see what console with 10TF Navi, 8 core Zen2 and 16GB of usable RAM can do, that console with 4-5TF Navi, same 8 core Zen2 and 12GB of usable RAM can't at 1/3rd of resolution.

Anything that uses a heavy amount of compute that is not tied to resolution (say for example, physics). If your using 2-3TFLOPS (terrible measurement I know), of compute on a 10TFLOP console for anything that doesn't scale with resolution your a bit shit out of luck on a 4-5TFLOP console.
 
I have my reservation regarding this for two reasons :

1) Lockhart or no Lockhart, MS's strat has already been set to stone 2 years ago with inclusion of game pass and "play everywhere".

This basically means, as Phil said many times, you will be able to play their games on top end console or vast amount of other devices (Xcloud - which will not be 4K highest settings, and plethora of PC's). I am 100% sure they put alot of time and effort into making game engines and their own development tools as scalable as possible.

2) Tech is on fundamentally same level.

Both Lockhart and Anaconda will provide same base architecture and technology (as well as PS5), but clear distinction will be made as to which resolution they target. We know for fact that with every generation, consoles have sweet spot they are shooting at in terms of resolution. For 360 and PS3 it was 720p, which was more or less hit for 30fps games (10% +/-). With PS4/XBO it was 900p/1080p.

For PS5 and Anaconda, I expect it to be 4K, or in certain cases middle between 2K and 4K with good up-scaling. Everyone on this forum is aware that by far biggest performance hit comes with increase of resolution, therefore a console which is targeting 4K, and provides 10TF GPU, will reserve alot of rendering budget for increase in resolution, vs one that targets 1080p/1440p and has 4-5TF.

So from that stand point, I don't see what console with 10TF Navi, 8 core Zen2 and 16GB of usable RAM can do, that console with 4-5TF Navi, same 8 core Zen2 and 12GB of usable RAM can't at 1/3rd of resolution.

People are mentioning PS2/Xbox and Wii/360/PS3 years, but that is completely misplaced comparison. Not only where tools back then archaic (PS2 documentation was bizzare even for early 00's), compared to what we have now, but difference in architecture and technology of individual consoles where gigantic.

In MS case you would have 2 consoles from same manufacturer, with same tools and exact same design. Only one would have 1/3rd of GPU performance, but would also be shooting for 1/3rd of resolution. In the day and age we have game engine scaling across top end PCs running RT effects, to mobile phones, saying this will result in development headache's is reaching IMO.

My final point would be regarding your comment about Sony's first party enjoying pushing one single console to the limit while, MS studios have to fiddle around with two, resulting in big difference in visual quality. Don't agree with that either for several reasons :

1) MS studios are already targeting more devices then it would be the case in potential Anaconda and Lockhart scenario. They are targeting everything from X to S and One, and to every single PC there is (along with future Xcloud)

2) 3rd party games this gen have come closer to 1st party then it was ever the case. Some of the best looking and running games this gen have been 3rd party (RDR2, Doom, RE2, DMC, COD MW etc.) and difference between 1st party and 3rd party is much more in terms of asset quality and art direction, then it is in tech.

3) Scalability is becoming more important then hard generation resets. This is because IMO we are seeing graphics being more then good enough for most of the time, and differences between great looking games is minimal (to a point where art style play bigger role). All that while tech is no longer advancing at rate it was in early 2000s.

So my point is, MS not releasing Lockhart would be huge, huge mistake. For MS, they already have bunch of Lockharts, with even bigger and wider gap between top and the bottom, so not releasing something like Lockhart with GP to counter Sony's one console strat would be a missed opportunity.

Lockhart will give people an option to play next gen games at lower resolution. If Sony pushes their exclusives in sub 4K territory, MS can easily do the same and still lower Lockhart to 1080p, not like many people buying ~$299 console will care about if its output is native 1440p or 1080p.

We are in entering the age of convergence and MS is smartly reacting to that. I can see Sony doing the same with Ryan at helm. Starting with PSN Now to racks of "lower end console" that acts like streaming platform for millions of people not caring about 4K.

Do you work for Microsoft or why do you spread "Lockart is so good" Idea here?
Why we need lockart when MS already say that Scarlett games also run on the 7 old years Xbox one ? When we have the Xbox one X ? So for what did we need Lockart? For MS Ecosystem? I dont buy a gaming Console in point of an uninteresting Ecosystem, i want to play innovative Games , Exclusives , and a interesting Hardware to have under the TV. You cannot replace good Exclusivgames , innovative Hardware with a chaotic Ecosystem MS have. This is not for me. And you cannot use the full Hardwarepotenzial of Scarlett when the Games must also running on Xbox Lockart, X, One. So its like a PC where the Games must run on different Configurations. Nobody need that ,only Microsoft. This also the reason why the original Xbox Creator from 2001 leaving Microsoft , because he critisize it that MS had the original Xbox Concept destroyed and Microsofted it.

So no , nobody need such a gaming Device that only benefits the interests of a Company , where the Customers are slaves to pay for an stupid Ecosystem and uninteresting play Anywhere Games.

My Money goes to Ps5 and Switch 2 and eventually a good gaming PC.
 
Anything that uses a heavy amount of compute that is not tied to resolution (say for example, physics). If your using 2-3TFLOPS (terrible measurement I know), of compute on a 10TFLOP console for anything that doesn't scale with resolution your a bit shit out of luck on a 4-5TFLOP console.
Its still tied to resolution if you are wasting your cycles on putting out 2 or 3 times more pixels. Only way to avoid it is to render PS5 game in 1080p and do the same with Lockhart, then you are left with bunch of rendering budget and only way for Lockhart to keep up is to severely tone down details.

Still, what history has shown is that any additional rendering time left on GPU side will be spent on nicer pixels, and not anything else. Next gen CPUs will be bring big increase over last in terms of CPU power, and less so on GPU side of things, therefore I doubt you are going to see devs spending GPU cycles on additional physics that cannot be done on 4-5TF machine.
 
Status
Not open for further replies.
Back
Top