Predict: Next gen console tech (10th generation edition) [2028+]

[function]

The next immediate candidate for comparison is Sea of thieves. Due end of this month, fully optimised for XB. How well will PS5 hold up?

Last gen game with a cross gen update, so I'm going to guess that it'll run very well on PS5. If there are any issues at launch they'll be fixed up pretty quickly.

I think games would not only need to have Series consoles as the lead platform, but also be next gen only games (e.g. mesh shader and lots of compute) to really favour the Series consoles.

Alan Wake, Avatar, multiplatform UE5 stuff all seem to do well on Series consoles, and they aren't even targetting Series consoles as lead. Even with Xbox as lead, a cross gen XBone game is likely to run really well on PS5.

 

[iroboto]

Last gen game with a cross gen update, so I'm going to guess that it'll run very well on PS5. If there are any issues at launch they'll be fixed up pretty quickly.

I think games would not only need to have Series consoles as the lead platform, but also be next gen only games (e.g. mesh shader and lots of compute) to really favour the Series consoles.

Alan Wake, Avatar, multiplatform UE5 stuff all seem to do well on Series consoles, and they aren't even targetting Series consoles as lead. Even with Xbox as lead, a cross gen XBone game is likely to run really well on PS5.

Yea I was also on the line of thinking like Gears 6. The next Halo.

I’m not expecting much from Sea of Thieves either. Any differences will be covered by DRS I suspect.

 

[Seanspeed]

Being the lead console has provided significantly more dividends for them. Feature set wise, the PS5 is the most limited in capability amongst the latest generations of hardware. Imagine how much larger an issue would have been if they didn’t ship with RT.

I imagine they're both valid and overlapping points to the same effect.

It's gotta be frustrating for MS to have made the more powerful console at a higher cost, only for devs to not really give it more attention, which has largely neutralized that extra power.

 

[rntongo]

I imagine they're both valid and overlapping points to the same effect.

It's gotta be frustrating for MS to have made the more powerful console at a higher cost, only for devs to not really give it more attention, which has largely neutralized that extra power.

Its not that devs didnt give it more attention. Its that MS decided to launch two consoles at the same time with different amounts of memory despite console titles having very strict requirements compared even to PC releases. Remember there was only one devkit which was a Series X and developers were having issues when testing the games on actual Series S units from the get go. Sony is doing it smarter and simpler. MS just needed to release one hw system, a midgen upgrade with the same amount of ram and maybe a handheld at the end of the gen. Its really really that simple. Or MS could relax requirements on Series S releases.

 

[Tkumpathenurple]

Its not that devs didnt give it more attention. Its that MS decided to launch two consoles at the same time with different amounts of memory despite console titles having very strict requirements compared even to PC releases. Remember there was only one devkit which was a Series X and developers were having issues when testing the games on actual Series S units from the get go. Sony is doing it smarter and simpler. MS just needed to release one hw system, a midgen upgrade with the same amount of ram and maybe a handheld at the end of the gen. Its really really that simple. Or MS could relax requirements on Series S releases.

I'm not so sure. I think MS handled their dual SKU launch poorly from a developer's perspective, and confusingly from a customer's perspective.

16GB of 14gbps GDDR6 for the Series S and 20GB of 16gbps GDDR6 for the Series X would've made both easy to use for developers and yielded visible results for customers. No farting about with onerous memory constraints or split speed bandwidths. Just ample capacity and easy utilisation.

For the sake of saving tens of dollars per console at the start of the generation and a pittance per console by now, it doesn't seem like that area of penny pinching was worth the drawbacks.

 

[Charlietus]

I'm not so sure. I think MS handled their dual SKU launch poorly from a developer's perspective, and confusingly from a customer's perspective.

16GB of 14gbps GDDR6 for the Series S and 20GB of 16gbps GDDR6 for the Series X would've made both easy to use for developers and yielded visible results for customers. No farting about with onerous memory constraints or split speed bandwidths. Just ample capacity and easy utilisation.

For the sake of saving tens of dollars per console at the start of the generation and a pittance per console by now, it doesn't seem like that area of penny pinching was worth the drawbacks.

The series consoles would have been better without the dual motherboard setup, the squared look, the 320 bit bus and the slow clocks. Just make a big, cheap console to build like the PS5, as easy for developers as possible. The series s should have had 6 teraflops and 12gb of ram for 349€. The fast and slow ram is especially puzzling to me, how much are they saving with that? Just make a 256 bit bus with 18gb ggdr6 at launch, like c'mon Microsoft

 

[Shifty Geezer]

This is next gen prediction thread, not current gen retrospective.

 

[Seanspeed]

Its not that devs didnt give it more attention. Its that MS decided to launch two consoles at the same time with different amounts of memory despite console titles having very strict requirements compared even to PC releases. Remember there was only one devkit which was a Series X and developers were having issues when testing the games on actual Series S units from the get go. Sony is doing it smarter and simpler. MS just needed to release one hw system, a midgen upgrade with the same amount of ram and maybe a handheld at the end of the gen. Its really really that simple. Or MS could relax requirements on Series S releases.

If they do a midgen upgrade rather than 'new gen' console, they are gonna be stuck with Series S, for better or worse. MS is not in a position to be angering their own customer base by abandoning them early.

But I do think lack of attention is definitely a part of why their results haven't been great, and that's an issue they'll face if they do a proper 'new gen' console anytime soon(ala 2026), cuz absolutely nobody is going to make games just for that system. It's gonna be a 'midgen' upgrade in all but name no matter what, except releasing far too late.

This is exactly why I think whether they do a midgen console or not, their actual next gen box needs to coincide closer with PS6, so in 2028 or later. Let them take their time and come out with a more thought through design, maybe focus a bit more on creating a leaner chip so even if they get outsold, they wont have to take the double whammy of both selling less and making less per unit sold than the competition. So long as they dont have an XB1-like performance gap to the Playstation 6, it shouldn't matter too much if they end up a little behind in game performance.

 

[rntongo]

If they do a midgen upgrade rather than 'new gen' console, they are gonna be stuck with Series S, for better or worse. MS is not in a position to be angering their own customer base by abandoning them early.

But I do think lack of attention is definitely a part of why their results haven't been great, and that's an issue they'll face if they do a proper 'new gen' console anytime soon(ala 2026), cuz absolutely nobody is going to make games just for that system. It's gonna be a 'midgen' upgrade in all but name no matter what, except releasing far too late.

This is exactly why I think whether they do a midgen console or not, their actual next gen box needs to coincide closer with PS6, so in 2028 or later. Let them take their time and come out with a more thought through design, maybe focus a bit more on creating a leaner chip so even if they get outsold, they wont have to take the double whammy of both selling less and making less per unit sold than the competition. So long as they dont have an XB1-like performance gap to the Playstation 6, it shouldn't matter too much if they end up a little behind in game performance.

Agreed, a next gen system in 2026 would be blunder. People will just buy PS5 pros. They need to launch closer to the same time as the PS6 and for now just focus on building a library of titles and cross gen next gen titles. 2028 would be a good year. Biggest change for next gen hw will be the amount of memory, hw acceleration for upscaling and RT. disk throughput is already good enough by 2028 we'll have much faster cheaper storage than is available today. But next gen they need to launch one single hw system. They can experiment with a licensed Xbox hw system if they want, but if they try releasing two systems at the same time again, it will be a huge blunder again.

 

[cheapchips]

So long as they dont have an XB1-like performance gap to the Playstation 6

Why not, in a world of diminishing graphics returns and sophisticated upscaling?

A 2026 Xbox is guaranteed to offer 'infinite geometry', good-ish realtime GI with reasonable image quality at high-ish frame rates. It's biggest challenges are software to show it off and differentiating itself from the PS5 Pro.

The PS6 has a similar quandary.

 

[Johnny Awesome]

The S wasn't a blunder. It Saved the Xbox division.

But I do think lack of attention is definitely a part of why their results haven't been great, and that's an issue they'll face if they do a proper 'new gen' console anytime soon(ala 2026), cuz absolutely nobody is going to make games just for that system.

LOL. Except their own 32 dev teams.

2025 for mid-gen w/Gears 6 and Fable
2029 for next-gen w/Doom Next and ES6

By the time 3rd parties support PS6, the Xbox will be out a year later and more powerful and running the same games.

 

[Seanspeed]

Why not, in a world of diminishing graphics returns and sophisticated upscaling?

A 2026 Xbox is guaranteed to offer 'infinite geometry', good-ish realtime GI with reasonable image quality at high-ish frame rates. It's biggest challenges are software to show it off and differentiating itself from the PS5 Pro.

The PS6 has a similar quandary.

Unless Sony decides to put their 1st party stuff on Xbox(not gonna happen), it will lead to a very clear advantage for Playstation 1st party games in terms of impressiveness. And for 3rd party titles, Xbox titles will likely already get less attention from developers, which will exaggerate the actual technical gap(in my opinion).

Diminishing returns are always real, but there's still gonna be plenty of room for improvement for developers to push things, and thus ways for technical gaps to be visually distinguished.

I do think MS might have a bit more leeway to be behind, but I dont think they can get away with being too noticeably inferior. At least not unless they've really got on top of their games output and are releasing a number of games not available on Playstation.

 

[Seanspeed]

The S wasn't a blunder. It Saved the Xbox division.


LOL. Except their own 32 dev teams.

2025 for mid-gen w/Gears 6 and Fable
2029 for next-gen w/Doom Next and ES6

By the time 3rd parties support PS6, the Xbox will be out a year later and more powerful and running the same games.

Xbox locking away their own games behind not just their own platform, but their next gen box, seems to go against everything they're doing right now, though. They were all about cross-gen, and now they dont even seem to value exclusivity at all much anymore.

I do actually think going back to exclusivity is a better move for them in the long run, assuming they can actually start delivering on some real must-haves. It's just, they themselves dont seem to think so.
So no, I dont think they'd have anybody whatsoever making games specifically for some theoretical 2026 machine. At least not for like three years.

 

[Johnny Awesome]

Not locking away. Just superior versions. Nothing wrong with launching a mid-gen with Gears 6 and Fable "Xbox exclusive" and then porting them for the PS6 launch 3 years later.

Old news by then...

A lot of the perils of their new approach involve consumer confidence. If Xbox fans can be confident that they're getting the big releases Day 1 on GP on superior hardware 3 years before PS users, they'll continue to buy the hardware and stay in the Xbox ecosystem. Conversely, it's good if PS users realize they aren't getting ES6 until 2032. If they really want it in 2029, they better get an Xbox or PC.

 

[eastmen]

The S wasn't a blunder. It Saved the Xbox division.


LOL. Except their own 32 dev teams.

2025 for mid-gen w/Gears 6 and Fable
2029 for next-gen w/Doom Next and ES6

By the time 3rd parties support PS6, the Xbox will be out a year later and more powerful and running the same games.

Or MS can just be on its own cadence and sony will be on the back foot in terms of hardware. 2025 for a new system and 2030 for a new system. MS has great BC support so there shouldn't be much of an issue from moving on to new systems.

 

[Karamazov]

Unless Sony decides to put their 1st party stuff on Xbox(not gonna happen),

What if...Next Xbox console let you boot Windows OS...?

 

[Shortbread]

What if...Next Xbox console let you boot Windows OS...?

Aww yes, the old trojan horse method to install Steam and gain access to PS games.

I see two possibilities: (a) Sony unofficially supports XB hardware through this roundabout method and gains additional revenue from older titles, or (b) Sony officially doesn't give a f*** since it's essentially a PC at this point.

I just don't see Sony spending resources on a methodology trying to prevent this. If anything, Sony probably wants this to happen. Think about it, Sony doesn't have to share a percentage of sales with Microsoft, only the current agreement between them and Steam. The same can't be said with XB games coming to PS.

 

[QPlayer]

Not locking away. Just superior versions. Nothing wrong with launching a mid-gen with Gears 6 and Fable "Xbox exclusive" and then porting them for the PS6 launch 3 years later.

Old news by then...

A lot of the perils of their new approach involve consumer confidence. If Xbox fans can be confident that they're getting the big releases Day 1 on GP on superior hardware 3 years before PS users, they'll continue to buy the hardware and stay in the Xbox ecosystem. Conversely, it's good if PS users realize they aren't getting ES6 until 2032. If they really want it in 2029, they better get an Xbox or PC.

And what if MS makes a deal with Rockstar, and GTA6 Day1 comes to Gamepass only for the new Xbox consoles?

 

[Johnny Awesome]

I don't think Rockstar would do that.

My whole point about hardware really is that I think MS benefits by being 1 year behind Sony now. Nothing takes advantage of the hardware fully at launch anyway, Sony's going to sell every machine they make regardless of whether or not another Xbox launches at the same time, and digital ecosystem lockdowns give every incentive for hardcore Xbox fans to wait a year to get better hardware than jump ship to Sony. The other advantage of being a year behind is that even unoptimized games where PS is the lead platform will play at least as good on 1 year newer hardware.

Mid-gen 2025 and new console in 2029.

The rumored handheld is on it's own path and just has to deliver S performance in a handheld, whenever that can happen. The next handheld has to deliver X performance which will be a long while from now.

With 32+ dev teams MS never really has to worry about support ever again. Like Nintendo, they can do whatever they want.

 

[Tkumpathenurple]

I keep wondering whether chiplets might somewhat alter the console tiers of the next generation.

Rather than the current gap we see between the Series S and Series X, as well as the purported gap between the PS5 and Pro, might there be scope for launch tiers of similar power?

For example, a 64CU GPU and 16/32c/t CPU on a chiplet design would permit full fat chiplets in one model, and partially disabled chiplets in a budget model. There would presumably be scope for matching those with differing amounts of Infinity Cache and speeds of GDDR7.

They could potentially even afford to be more aggressive on the degree to which they disable CU's and cores, depending on the cost benefit analysis of using more of each wafer.