Next-gen Cross-Platform Strategy [2020]

Sure but it’s a big part of it. Density means over draw. Over draw means depth/ordering. Over draw means culling. These are things we do on the CPU before we issue draw calls as well. As the scene gets more complex so does the process to set the scene up for drawing. The asteroid test wasn’t a draw call test; there is another test for that.

I don’t agree that the idea in the launch of these consoles that we will have AI so great and physics so great that people will take notice of; over what we have today; is unlikely.
AI and physics weren’t pared back because of the lack of available power. They were pared back (like audio) because people didn’t notice. The better AI got; the more people thought it was cheating. Physics didn’t even matter; unless we have gameplay built around it (problem solving) it’s still all very limited.

I don't expect crazy physics but at least physics for all NPC and like I said simple crowd AI. The Jaguar was and is still a problem for design of games. The physics was not stagnant, it was a regression current gen.
 
I don't expect crazy physics but at least physics for all NPC and like I said simple crowd AI. The Jaguar was and is still a problem for design of games. The physics was not stagnant, it was a regression current gen.
sure but that doesn’t sound like anything much better than we have today. Not enough for the concern trolling (in general) about running cross gen games for a launch window.

The core problem with adding physics/destructible physics/ more NPCS, larger crowds et; there are still game design considerations in which you may not want it. I get you could specifically design set pieces to show off tech not possible from this generation, but I’d that doesn’t make a game necessarily better than the games we have today.
 
sure but that doesn’t sound like anything much better than we have today. Not enough for the concern trolling (in general) about running cross gen games for a launch window.

The core problem with adding physics/destructible physics/ more NPCS, larger crowds et; there are still game design considerations in which you may not want it. I get you could specifically design set pieces to show off tech not possible from this generation, but I’d that doesn’t make a game necessarily better than the games we have today.

This is not concern trolling, this is the reality. True next generation games will look better than cross gen games and they will have other things out of the visual being better. There is a reason they showed a title very early in development like Hellblade 2 during Xbox Series X reveal. I doubt the game will release on Xbox One. Like I said wait the reveal of the two consoles and we will see which people will see if there is a difference between cross gen games and games made only for next generation console.
 
Last edited:
This is not concern trolling, this is the reality. True next generation games will look better than cross gen games and they will have other things out of the visual being beter. There is a reason they showed a title very early in development like Hellblade 2 during Xbox Series X reveal. I doubt the game will release on Xbox One. Like I said wait the reveal of the two consoles and we will see which people will see if there is a difference between cross gen games and games made only for next generation console.
Yes but this thread is about next gen exclusives in particular happening at launch. It does mean there won’t ever will be. GPU can scale. Unless there is an introduction of new features that cannot; compute shaders cannot scale back to 360 and RT can be adjusted to turn off. That only leaves a power gap in CPU and a read speed difference on the hard drive. I’m just not expecting all of this at launch and I have doubts that PS5 exclusives when they launch will be head and shoulders over what we have today.

I’m not expecting Hellblade 2 at launch. I don’t even think we will see it as early as holiday 2021.
 
If I'm understanding correctly. Killzone is being used as an example of a launch game i.e not impressive in scope or something?
Think a lot of context is being missed, apart from being simply a launch title.
They went from PS3 to PS4, different OS, totally different architecture, I'm sure the engine took more than a recompile and afternoon of optimization.

My point is what makes cross gen games easier, is actually what makes getting more out of next gen quicker possible.
  • Tools same
  • OS same (at least minimum expect the underlying OS to be the same)
  • Engine same
  • Architecture similar

So the isues with launch game development this time around is largely mitigated.
It wouldn't be years of research to take the same engine and do more on the CPU than is possible on current gen.
Same with SSD.

Before it took years for games because architecture fundamentally would change. Just not the case this time.
As a result, expect graphically at least for cross gen games to look impressive right away.
 
Yes but this thread is about next gen exclusives in particular happening at launch. It does mean there won’t ever will be. GPU can scale. Unless there is an introduction of new features that cannot; compute shaders cannot scale back to 360 and RT can be adjusted to turn off. That only leaves a power gap in CPU and a read speed difference on the hard drive. I’m just not expecting all of this at launch and I have doubts that PS5 exclusives when they launch will be head and shoulders over what we have today.

I’m not expecting Hellblade 2 at launch. I don’t even think we will see it as early as holiday 2021.

It is a 2022 games probably and Matt Booty told before 2022 we will not see exclusives game on Xbox Series X. The Strategy will be at the advantage of Sony all 2021.
 
So the isues with launch game development this time around is largely mitigated.
It wouldn't be years of research to take the same engine and do more on the CPU than is possible on current gen.
Same with SSD.
Not necessarily. The easiest and most straight forward use case is to just let everything run faster.
Think about the absolute amount of CPU content you have to build to actually take advantage of the increase amount of CPU cycles such that you're now back to 30fps while optimizing the heck out of everything.

It's a simple house analogy.
The smaller the house, the less it costs to fill it with furniture.

Now you've got all this CPU and hard drive - it's gonna cost budget to fill it up. And people have to notice it. They have to notice a difference otherwise it's pointless use of resources to invest all this time in running way more developed AI and physics if no one gives a shit or no one knows the wiser.

But people see resolution differences. They feel framerate differences. They feel load times.
Will studios realistically fill their engine with so much stuff suddenly, break a budget on a game, and flat out remove those advantages as well?

It just doesn't make any sense to me, especially at launch.
It takes years for engines to slowly convert their stuff to newer tech. Content tools need to be overhauled. There's a reason Ubisoft has 5 different engines and refuse to have all their titles running off a single engine (like what EA attempted). It's because each engine, is designed for a specific purpose.

I get not wanting to support Xbox One, because it's a just the most troublesome piece of hardware to work with, with respect to the group. But 3P isn't locked to that rule. Only 1P is, they only will focus on 3 platforms anyway.

If Matt Booty thinks that they can do it; I think they have the metrics that can prove they probably can pull it off.

Its okay if MS restricts itself. But lets be real; unless your game is PUBG, or like 100v100 in a huge 30KM^2 map with all sorts of environments; you're' not likely to take advantage of this all. And MS isn't limiting 3P from choosing their platforms for development.
 
Not necessarily. The easiest and most straight forward use case is to just let everything run faster.
That's pretty close to just running it in BC mode.
If anything the fact you can just run everything faster shows that the level of development issues of making a launch game will now mostly be mitigated.

Worth pointing out I'm not talking about pros and cons of launch exclusive, just about development of them.
 
Last edited:
That's pretty close to just running it in BC mode.
If anything the fact you can just run everything faster shows that the level of development issues of making a launch will now mostly be mitigated.
It's not quite BC if your'e compiling the code for 2 separate platforms. But I see where your going with it.

Game Design will ultimately be the biggest factor in the determination of whether or not those resources are actually required.
 
Not necessarily. The easiest and most straight forward use case is to just let everything run faster.
Think about the absolute amount of CPU content you have to build to actually take advantage of the increase amount of CPU cycles such that you're now back to 30fps while optimizing the heck out of everything.

It's a simple house analogy.
The smaller the house, the less it costs to fill it with furniture.

Now you've got all this CPU and hard drive - it's gonna cost budget to fill it up. And people have to notice it. They have to notice a difference otherwise it's pointless use of resources to invest all this time in running way more developed AI and physics if no one gives a shit or no one knows the wiser.

But people see resolution differences. They feel framerate differences. They feel load times.
Will studios realistically fill their engine with so much stuff suddenly, break a budget on a game, and flat out remove those advantages as well?

It just doesn't make any sense to me, especially at launch.
It takes years for engines to slowly convert their stuff to newer tech. Content tools need to be overhauled. There's a reason Ubisoft has 5 different engines and refuse to have all their titles running off a single engine (like what EA attempted). It's because each engine, is designed for a specific purpose.

I get not wanting to support Xbox One, because it's a just the most troublesome piece of hardware to work with, with respect to the group. But 3P isn't locked to that rule. Only 1P is, they only will focus on 3 platforms anyway.

If Matt Booty thinks that they can do it; I think they have the metrics that can prove they probably can pull it off.

Its okay if MS restricts itself. But lets be real; unless your game is PUBG, or like 100v100 in a huge 30KM^2 map with all sorts of environments; you're' not likely to take advantage of this all. And MS isn't limiting 3P from choosing their platforms for development.
They got these with One X so you better have more than that if you want them to buy your new shiny. If all we get are better load times then might as well skip this gen entirely. Luckily I believe you are quite wrong and we'll see it from the start, just like every other gen.
 
They'd have to do better than the first full year of PS4, where the only dedicate PS4 retail titles were Driveclub and Second Son.

Don't be afraid Shu said to Famitsu about this generation in the first party side he considers they failed before 2017.

Like I am 99,9999% sure the Xbox Series X will have a GPU more powerful than the PS5. I think Sony will have a much better first third years than during current generation.
 
It's not quite BC if your'e compiling the code for 2 separate platforms. But I see where your going with it.

Game Design will ultimately be the biggest factor in the determination of whether or not those resources are actually required.
Yea, pretty much compiling with a different profile.
Mainly tweak graphics, and you've now got a cross gen launch game. That just wasn't possible before. Speed of ssd will give instant improvement to the feel of playing on the system also.

But the engine and tools will be compatible with next gen also, so the level of work to just get a game working is a magnitude different if wanted to make a launch non cross gen game.
 
Ahh, you're going to have to recall that DX12 wasn't out when a great deal of many games were designed. Most of them have not yet fully adopted DX12 based engines. Most still support DX11
So even with a super processor you may not have been able to run super rich and dense games on PC either.

All I'm saying is, I wouldn't generalize how CPU and SSD is enough to unlocking all this potential that was locked away by crappy jaguar CPUs. It's not representative of the entire landscape.
Keep an eye on CPU % below 4x more processing power is the same thing as 1/4 the load too ;)

https://www.dsogaming.com/news/dire...proves-performance-greatly-reduces-cpu-usage/

Max McMullen, Principal Development Lead for Direct3D and DXGI at Microsoft, demoed Intel’s Asteroids benchmark in order to show the performance difference between DX11 and DX12, as well as DX12 Bindless and DX12 ExecuteIndirect.

As we can see below, DX11 ran the demo with only 29FPS. CPU 21% - could not submit enough draw calls
DX11-Asteroids.jpg


DX12 CPU 38%
DX12-Asteroids.jpg


Lastly, via Execute Indirect, Intel’s Asteroids demo ran with almost 90FPS. CPU 9%
What’s really great here, however, is that the CPU usage was reduced significantly, meaning that more CPU intense scenarios will be possible via this rendering method.
DX12-ExecuteIndirect-Asteroids.jpg

Damn, that is very impressive.

Same we speak about exclusives games strategy, I speak about what will be possible to do for first party Sony studio, because the game will only be for PS5.

Are you working for Sony, or a first party developer? I see you everywhere, on different forums claiming things that no-one else so far seems to know about. I highly doubt that MS won't exploit their hardware at all. I'm very certain though that we will see games doing that, even on their platform, if that might be at launch or later is another thing. In general, every Xbox since 2001 had games exploiting the hardware. Now with more focus on 1st party AAA games instead of TV and Kinect, i doubt there's something to worry about here.
We already have SC doing just that, taking advantage of even Optane SSD's, massive amounts of RAM and GPU power, and last but not least CPU.

Depending of SSD speed I don't even expect Rockstar to push the PS5 I/O.

Rockstar pushed about everything this gen, and they will again next generation. More so perhaps then even Sony's own studios. Rockstar has the resources and a mountain of a team of experts. XSX having most likely having the same SSD tech as Sony, i doubt what your claiming anyway.
On top of that, i read on another forum someone doing a test, and the conclusion was that Sony's spider-man SSD demo could have been around 2000mb/s. No one can be sure, but this isn't the baseless topic, i doubt there will be anything else then NVME in there, just like the Xbox. Both consoles (and PC) wont be far off from eachother in practital performance.

I’m just not expecting all of this at launch and I have doubts that PS5 exclusives when they launch will be head and shoulders over what we have today.

It has never happened before, chances are small it will next-gen, in special thinking that tech jumps arent that huge anymore.

The Strategy will be at the advantage of Sony all 2021.

How do you know this? Has Sony announced what games will be available for only PS5 and when?

They'd have to do better than the first full year of PS4, where the only dedicate PS4 retail titles were Driveclub and Second Son.

Exactly, launch lineups never have been spectacular, what MS has said isn't a disadvantage, its reality. Game development has gotten more and more time consuming and therefore more expensive. With smaller jumps in power between consoles/tech, chances are small we will see a huge difference to the past 25 years of gaming.

I think Sony will have a much better first third years than during current generation.

Which means you have a crystal ball which tells you exactly what's gonna happen in the future :) It could be another PS360 generation aswell. PS2 outclassed XB on every aspect regarding software. That didn't really translate into the 7th gen.

KZ: Shadowfall

Well, that is exactly what he ment. I highly doubt the devs are going forward with another attempt.
 
They got these with One X so you better have more than that if you want them to buy your new shiny. If all we get are better load times then might as well skip this gen entirely. Luckily I believe you are quite wrong and we'll see it from the start, just like every other gen.
I get where you guys are coming from. I'm really just focused on the 2020/2021 year here because after that point MS is dropping Xbox One.

I'm just looking at the tech here; The graphics stack has barely changed. We got Ray Tracing (maybe) accelerated enough to use to some degree. We got some ML library stuff. Everything else going into next gen will largely be an evolution in features moving from tier 1 support to tier 3 support.

Whereas The switch from 360/PS3 to this gen had massive changes in feature set.

So in many ways, the graphics should be able to scale down next gen to this gen because the feature sets should support it to run, it's only the muscle that is lacking. The counter argument is that this gen can't keep up with teh cpu and hard drive requirements of next gen. I agree it can't, but I don't agree it's gonna happen in 2020/2021.
 
In terms of business cases for and against gen exclusives.
Cross gen games will have a lick of paint, so will look good
benefit from ssd without any work needing to be done.
sell to a much larger install base

The positives for gen exclusives aren't all as simple
Make use of CPU and ssd in ways not possible on current gen
It's easy to say all consoles will sell out in the first 6 months, possibly longer. But that doesn't mean that mindshare and momentum is the same.
If your going to upgrade and the choice is one system where it has a gen exclusive where that's the only place you can play the game compared to still being able to play it on current system, it may make you choose to upgrade to the system with that gen exclusive first.
If you upgrade that one, would you buy your multiplates on previous gen or the new system you just bought?

So in the short term you've made more money selling to cross gen, but medium to long term it could effect mindshare, perception, momentum, as people will tell their friends to buy a machine because that's the one they've bought first and currently playing on, maintaining momentum beyond the initial launch window.

Do I think all 1P games need to be gen exclusives, no far from it. But I do think having some can be very beneficial.
Start of gen is best chance to also get people into ecosystem, even switch. Even this time around.
 
Just curious since i don't know, but was Nintendo heavily penalized for having Zelda BOTW on both WiiU and Switch?
While i don't know anyone that got it on WiiU, everyone was still very happy to play it on switch.

Were there discussions of limitations and downgrades?
 
Just curious since i don't know, but was Nintendo heavily penalized for having Zelda BOTW on both WiiU and Switch?
While i don't know anyone that got it on WiiU, everyone was still very happy to play it on switch.

Were there discussions of limitations and downgrades?
Not really comparable.
Is the switch better than the wii u, sure. But not to the same degree where talking about here.
Nintendo marches to beat of their very own unique drum.
Do you think people would be as excepting if Sony or MS launched a system with that level of performance?
 
Back
Top