Microsoft IGN: Cost of Xbox Series X

[Paperclyp]

I agree that speaking about Lockheart as if it’s a foregone conclusion is just a step of speculation too far. 

[Rbk_3]

6 hours ago, JPDunks4 said:

Engines these days do most of that work for you already.  Fortnite is playable from high end PC's down to Mobile Phones or Switch on the Unreal Engine.

 

With the number of updates they push out almost daily to all these platforms, I think its pretty clear the work is largely being handeled by the engine itself to properly scale.

 

Making versions of games to play on both versions of the next gen console won't be the same as porting games to different consoles.

 

* I also don't know shit about developing games, but based off a lot of what I've read the past few years, it sems to be the direction everything is going.  Scalable engines with hardware that is ready to utilize them.

Exactly. If PC games can be developed to play on 100's of different configurations, I don't see how it can't be done easily on a scaled back system with the same same architecture. 

[cusideabelincoln]

19 hours ago, Duderino said:

 

UE4 has been a mixed bag this generation in terms of scalability, at least as it relates to performance.  On one hand Fortnite, Gears 5, MK11, and others seem to scale well.  On the other, you have Jedi: Fallen Order, PUBG, State of Decay 2, We Happy Few, and many more that struggle on the base Xbox and/or PS4.  It's not just Unreal either, look at Anthem, Control, etc.

At least the Lockheart rumored CPU should be less of an issue but we're still looking at a GPU that is closer to the Pro or X.  Even with a target resolution of 1440p, higher poly counts, increased shading complexity, and more demanding games in general are likely going to result in Lockheart requiring special considerations and optimizations.  Possibly even separate lighting solutions from the Series X or PS5.

 

The lesser consoles running at 1080p is the easy solution here.  The base Xbone sees resolution drops below 900p now.  When the new systems are targeting 4k or 1440p, 1080p will be the fallback for the slower variants.

[Duderino]

19 hours ago, Paperclyp said:

I agree that speaking about Lockheart as if it’s a foregone conclusion is just a step of speculation too far. 

I trust that Jason Schreier (Kotaku) has accurate sources, that Digital Foundry has reason to believe it is still in the works, and well... that an outspoken Jonathan Blow (the Witness) is not looking forward to at least one new console that is not the PS5.  Lockhart is by no means a forgone conclusion, but there is enough here to have a discussion about it as a possibility.
 

20 hours ago, JPDunks4 said:

 

I agree, but when Xbox One and PS4 were initially developed, I don't know if they were made with the same mentality of being able to handle these scalable engines.

 

A big messaging point around the One X was that they went to developers and took all the biggest game engines being used at the time, and looked at bottlenecks and other issues that come up, and built the hardware in order to best accommodate those engines to better be able to scale and make life as easy as possible for the developers.  

 

I think this is why my thinking is, when developing these new consoles, they will be accounting for their limitations more-so than they were back in 2012/2013, and try to make them better able to handle the inevitable mid gen refresh and constant improvements to PC tech.  

 

I just don't believe MS would be handicapping their powerful console.  Their entire messaging the past few years is to have creators create what they want to create.  If news came out that developers are being hamstrung by being forced to support the weaker console and its affecting their ability to build the games they want, that'd just shoot themselves in the foot.  

 

But who knows.  We don't even know of Lockheart even exists at this point.

In the case of the X, it's a lot easier to scale games up.  Scaling down next-gen tittles for a hypothetical Lockhart would be a bigger ask of developers.  Much easier to work with less limitations than more.

 

I doubt Microsoft wants to add more work for developers and possibly introduce limitations, but a second SKU is not outside of the realm of possibility if it fits their consumer facing strategy.
 

16 hours ago, Rbk_3 said:

Exactly. If PC games can be developed to play on 100's of different configurations, I don't see how it can't be done easily on a scaled back system with the same same architecture. 

Putting all problematic PC ports aside, I would agree that architecture is not the potential challenge with Lockhart. The power differential is.
 

1 hour ago, cusideabelincoln said:

 

The lesser consoles running at 1080p is the easy solution here.  The base Xbone sees resolution drops below 900p now.  When the new systems are targeting 4k or 1440p, 1080p will be the fallback for the slower variants.

That's one optimization strategy that could help when resolution is contributing to the bottleneck.  Different story though if the key limiting factors come from draw calls, tri counts, and/or compute shader performance.

[crispy4000]

Lockhart at the rumored specs would be a worse investment than buying an og (Kinect) Xbox One at launch.  It’d be much more of a pain in the ass for devs to optimize for well.

 

Plus at those numbers, it wouldn’t be surprising for a Switch 2 to come out ahead of it, at least on the GPU side.  I suppose that could be Lockhart’s lifeline...

[Paperclyp]

1 hour ago, Duderino said:

I trust that Jason Schreier (Kotaku) has accurate sources, that Digital Foundry has reason to believe it is still in the works, and well... that an outspoken Jonathan Blow (the Witness) is not looking forward to at least one new console that is not the PS5.  Lockhart is by no means a forgone conclusion, but there is enough here to have a discussion about it as a possibility.

It’s a possibility, but none of those links fills me with confidence that it is for sure coming, or that it’s not possibly another iteration of the XB1, or who knows. It’s too much hearsay, and it’s still a ways until launch. 
 

I have my doubts MS will split its base up from day 1, but I guess we’ll see in less than a year...

[Duderino]

24 minutes ago, Paperclyp said:

It’s a possibility, but none of those links fills me with confidence that it is for sure coming, or that it’s not possibly another iteration of the XB1, or who knows. It’s too much hearsay, and it’s still a ways until launch. 
 

I have my doubts MS will split its base up from day 1, but I guess we’ll see in less than a year...

Both Jason Shreier and DF are clearly talking about Lockhart in the context of the Scarlett family.  Not the Xbox One.

 

I have doubts too, but those have more to do with the practicality of Lockhart than the validity of these rumors and their sources.

[cusideabelincoln]

3 hours ago, Duderino said:

That's one optimization strategy that could help when resolution is contributing to the bottleneck.  Different story though if the key limiting factors come from draw calls, tri counts, and/or compute shader performance.

 

Resolution is the largest contributing factor to performance when there is no CPU bottleneck.  The two systems are rumored to use the same or similar CPU, so draw calls and AI won't need to be downscaled. Any other bottlenecks should be easily adjustable as most PC ports offer them as options.  It will be more work for developers to verify the optimal settings, but nothing most aren't used to by now who have put in the time for decent PC ports.

 

I'm guessing the reality of the situation is the vast majority of 3rd party multi-platform games that will come out for next gen are also going to be cross-gen, especially in the first 2-3 years of the lifecycle.  I suspect multiplatform games will only start becoming next-gen exclusive past the 2 year mark.  Publishers will appeal to the massive player base that is already established, because that's the economy we live in now.  Only the handful of showcase titles will be next-gen only.  We'll probably also see heavily cut down games.  Like Shadow of Mordor getting a dumbed-down Nemesis system and the Switch somehow barely running Doom and Witcher 3.

[Duderino]

1 hour ago, cusideabelincoln said:

 

Resolution is the largest contributing factor to performance when there is no CPU bottleneck.  The two systems are rumored to use the same or similar CPU, so draw calls and AI won't need to be downscaled. Any other bottlenecks should be easily adjustable as most PC ports offer them as options.  It will be more work for developers to verify the optimal settings, but nothing most aren't used to by now who have put in the time for decent PC ports.

 

I'm guessing the reality of the situation is the vast majority of 3rd party multi-platform games that will come out for next gen are also going to be cross-gen, especially in the first 2-3 years of the lifecycle.  I suspect multiplatform games will only start becoming next-gen exclusive past the 2 year mark.  Publishers will appeal to the massive player base that is already established, because that's the economy we live in now.  Only the handful of showcase titles will be next-gen only.  We'll probably also see heavily cut down games.  Like Shadow of Mordor getting a dumbed-down Nemesis system and the Switch somehow barely running Doom and Witcher 3.

The context of what the GPU has to render factors into those resolution costs.  As that complexity increases next gen, so will the challenge of supporting less capable GPUs whether it’s the last gen consoles, older PCs, or perhaps now Lockhart.
 

Not everything that negatively impacts GPU performance is guaranteed to have a convenient dial, a low acceptable threshold, an available alternative method, or a clear and easy path forward with optimization.  At least on PC devs can bump up the minimum and recommended specs.

[cusideabelincoln]

2 hours ago, Duderino said:

The context of what the GPU has to render factors into those resolution costs.  As that complexity increases next gen, so will the challenge of supporting less capable GPUs whether it’s the last gen consoles, older PCs, or perhaps now Lockhart.
 

Not everything that negatively impacts GPU performance is guaranteed to have a convenient dial, a low acceptable threshold, an available alternative method, or a clear and easy path forward with optimization.  At least on PC devs can bump up the minimum and recommended specs.

 

Like I said, there will be the handful of showcase titles that cannot use those convenient dials.  But the vast majority will be scaleable for the simple act of supporting the previous gen consoles.  When cross-gen fades out, the lower-tier next gen consoles will slot in perfectly as 1080p machines compared to the "4k" machines:

 

Just bumping resolution alone from 1080p to 4k requires reduces performance of the same card by over 60%.  Or in other words it takes about 2.5 times the raw GPU power just to push the extra pixels while maintaining the same framerate.   So far Lockhart could be 4TF, while the PS5 and XSX could be 9-12 TF.  2.5 times 4TF fits just nicely in that range.  Resolution is the most important context concerning GPU performance, assuming next gen is targeting 4k.  And I don't see why we can't assume that when the X1X is already targeting games at 4k.

 

(I used this source to get some average framerate performance on a per-resolution scale over a large number of games.  Focusing on the midrange cards, as the very top end cards would be CPU bottlenecked in a few games at 1080p, skewing the averages)

 

[crispy4000]

5 hours ago, cusideabelincoln said:

 

Resolution is the largest contributing factor to performance when there is no CPU bottleneck.  The two systems are rumored to use the same or similar CPU, so draw calls and AI won't need to be downscaled. Any other bottlenecks should be easily adjustable as most PC ports offer them as options.  It will be more work for developers to verify the optimal settings, but nothing most aren't used to by now who have put in the time for decent PC ports.

 

I'm guessing the reality of the situation is the vast majority of 3rd party multi-platform games that will come out for next gen are also going to be cross-gen, especially in the first 2-3 years of the lifecycle.  I suspect multiplatform games will only start becoming next-gen exclusive past the 2 year mark.  Publishers will appeal to the massive player base that is already established, because that's the economy we live in now.  Only the handful of showcase titles will be next-gen only.  We'll probably also see heavily cut down games.  Like Shadow of Mordor getting a dumbed-down Nemesis system and the Switch somehow barely running Doom and Witcher 3.

 

2-3 years into the lifecycle for cross-gen ports is longer than last gen’s transition, which was sustainable for closer to 1-2 years.  Why would should we expect this time to be longer?  

 

We’d have to see the desire for that come from business trends, IMO.  There will be plenty of excuses between the SSD and CPU to design something beyond the means of previous gens.  I don’t think it’ll take long for it to feel like genuine port jobs for devs when they try to scale back.  It will take added resources, and port house developers in several cases, to accommodate for that.  Which again, isn’t so different from this last transition. 

 

@Duderino is also right about some rendering effects being more taxed by resolution than others.  I don’t think the GPU will be quite as big of a factor, but when you think about base Xbox One and PS4 needing to handle next-gen port downs, it’s still a tall ask.  So is optimizing for 4 last gen SKUs...

 

I actually think the AAA industry might be ready to push on ahead sooner for that reason alone.  There’s too many SKUs.  If Lockhart is a thing, that’s even more reason.

[cusideabelincoln]

9 minutes ago, crispy4000 said:

 

2-3 years into the lifecycle for last gen ports is longer than last gen’s transition, which was more like 1-2 years.  Why would should we expect this time to be that much longer?  

 

We’d have to see the desire for that come from business trends, IMO.  There will be plenty of excuses between the SSD and CPU to design something beyond the means of previous gens.  I don’t think it’s going to take long for it to feel like genuine port jobs for devs when they try to scale back.  It’s going to take added resources, and several port house developers, to accommodate for that.  Which again, isn’t so different from this last transition. 

 

@Duderino is also right about some rendering effects being more taxed by resolution than others.  I don’t think the GPU will be quite as big of a factor, but when you think about base Xbox One and PS4 needing to handle next-gen port downs, it’s still a tall ask.  So is optimizing for 4 last gen SKUs...

 

I actually think the AAA industry might be ready to push on ahead sooner for that reason alone.  There’s too many SKUs.  If Lockhart is a thing, that’s even more reason.

 

I hope Lockhart is a thing for that very reason: Being really cheap and enticing in order to get more people to cross over to the new generation.  This kind of strengthens my 1080p to 4k argument, but weakens my cross-gen one. It would be really hard to develop for the base PS4 and Xbone.  The X1X and Pro wouldn't be as bad, but still rough.  Perhaps it's possible some "next gen" games could run on the Pro or 1X and not the base consoles.  I think the only precedence for this type of thing would be with a few games that came out for the New 3DS that the old 3DS couldn't run, right?

[Emblazon]

I just really hope that MS updates the Elite v2 software so that the profile button (which is in the exact same spot as the share button on the X controller) can be customized:

 

Tap: Take Screenshot

Double Tap: Record That

Long Press: Switch Profile

[crispy4000]

25 minutes ago, cusideabelincoln said:

 

I hope Lockhart is a thing for that very reason: Being really cheap and enticing in order to get more people to cross over to the new generation.  This kind of strengthens my 1080p to 4k argument, but weakens my cross-gen one. It would be really hard to develop for the base PS4 and Xbone.  The X1X and Pro wouldn't be as bad, but still rough.

 

I’d be surprised if Lockhart even maintains 1080p a few years into the next gen if these are the specs relative to the Series X.  By a mid-cycle launch, it’d be a much bigger burden on devs than the One S is now.

 

I’d rather the industry not go there.

[AbsolutSurgen]

Assuming that next gen games will all target 4k is not a very good assumption.  Many developers know that the benefits of resolution are lost on a tremendous amount of users, as the human eye can't perceive 4k using the screen sizes/distances many people sit from their screens.  Even though base XB1/PS4 still make up the majority of the console user base, there has been a number of high profile games that have had performance problems on those base consoles.  Imagine how some developers will behave if the "low power" console is in the minority.

When Lockhart was first rumoured, it was frequently called a "streaming box".  There is still a fairly decent likelihood that streaming is a big part of MS's strategy for Lockhart.  It may not need to run these games natively -- it can run indie/backward compatible games natively -- and needs to stream more demanding titles.

 

If we assume a 50% increase across the board for Nvidia's Ampere cards launching late summer/early fall -- that equates to the rumoured SeX rasterization/shading performance being roughly equivalent to a RTX-2170, and PS5 being roughly equivalent to a RTX-2160.  HOWEVER, it is unclear how the consoles would compare in terms of Ray Tracing -- and Ray Tracing was a significant contributor to the cost of the current gen NVidia turing cards.  If they don't have something equivalent to Nvidia's tensor cores, then their value relative to Nvidia's PC cards is somewhat diminished.  Combined with the fact that there is profit margin in video cards for NVidia, the AIB and the retailer that would be eliminated/significantly reduced -- people may be pleasantly surprised at the cost of these consoles.

 

Edit:  Clarifications to second paragraph

[crispy4000]

5 hours ago, AbsolutSurgen said:

Assuming that next gen games will all target 4k is not a very good assumption.  Many developers know that the benefits of resolution are lost on a tremendous amount of users, as the human eye can't perceive 4k using the screen sizes/distances many people sit from their screens.  Even though base XB1/PS4 still make up the majority of the console user base, there has been a number of high profile games that have had performance problems on those base consoles.  Imagine how some developers will behave if the "low power" console is in the minority.

 

That's a good point too.  I think we're likely to see a good amount of temporal upscaling next-gen.  Not just with raytracing employed either.  Good news is that the tech is already there, should be improving, and has even fooled the likes of Digital Foundry in some instances (notably, Battlefront II).
 

5 hours ago, AbsolutSurgen said:

When Lockhart was first rumoured, it was frequently called a "streaming box".  There is still a fairly decent likelihood that streaming is a big part of MS's strategy for Lockhart.  It may not need to run these games natively -- it can run indie/backward compatible games natively -- and needs to stream more demanding titles.

 

That's not a next gen console then, IMO.

[Duderino]

Something else to consider, 3rd party cross-gen games will need to support PS4, PS4 Pro, Xbox One, Xbox One X, PS5, Xbox Series X, likely PC, possibly Lockhart, and maybe even Stadia.

 

That’s an extremely tall order that I very much doubt will be sustainable for 2-3 years for the vast majority of developers.

[AbsolutSurgen]

2 hours ago, Duderino said:

Something else to consider, 3rd party cross-gen games will need to support PS4, PS4 Pro, Xbox One, Xbox One X, PS5, Xbox Series X, likely PC, possibly Lockhart, and maybe even Stadia.

 

That’s an extremely tall order that I very much doubt will be sustainable for 2-3 years for the vast majority of developers.

"Most" games won't support PS4/XB1 for much more than 1 year.

[crispy4000]

2 hours ago, AbsolutSurgen said:

"Most" games won't support PS4/XB1 for much more than 1 year.

 

And that includes the Pro and X.