Three things to factor in for the next generation of XBOX consoles:
1- Pixel pimping is reaching a wall. First because 8K is looking to end up a dead end, much like 3D TVs and VR gaming. Second because the development cost of the incremental features doesn't drive enough added sales to justify the cost in time and effort. And third because it takes time for developers to learn how to effectively use the latest and greatest features. XBOX right now has a whole bunch of extra features that are going unused and not just by third party studios writing for PS. Even the best tech-driven studios in the MS exclusivity "zone" don't use them. The generation will be over before we see any games truly exploiting an SX.
2- The next XBOX is coming early. It was likely yesterday, today it is a certainty. MS needs to change the subject from pixel pimping, much like Nintendo did with Wii and Switch, to a new paradigm where the new wave of games will be internally different. Coded differently (cheaper, for one), rendered different (Windows is getting an internal DLSS equivalent, how can XBOX not get it too? PCs are getting NPUs, XBOX *will* too. In a sense they already have one.) Paying different. The model of CPU and GPU is going to change. It has to. Doubling cores, or clocks, or pipelines alone will not suffice.
3- The most obvious change to the next XBOX and the reason it will come in '25/26 instead of '28 as previously planned is AI. We are already seeing AI tools that run local on PCs. Give it another 18 months and a closed configuration on the hardware side and 18 months of SLM evolution and DirectX13 is going to bring in AI driven levels of interactivity impossible without it. Say NPCs you can have arbitrary conversations with, bosses that adapt their moves to yours. And, unless Sony and Nintendo can acquire AI tech of their own in those 18 months, MS will have first mover advantage in those games.
None of this is to say that gamers invested in Sony or Nintendo will jump ship. Platform lock-in is real. Those folks will stay put. But much like GP and xCloud, AI will entice the next generation of young gamers to consider XBOX more seriously. If nothing else, because even at $599 it will be cheaper than a comparable Gaming AI PC. And that has always been XBOX's calling card.
Interesting. I always like your insights,
@fjtorres5591 , and usually agree with them, but not sure I'm with you on all of that. I'm an Xbox gamer (and PC, but not PS since the PS3, nor Nintentdo, which I only played at friends' houses), but I have never seen any indication that MS has a clue about hardware changes that will appeal to gamers like Nintendo did so brilliantly with both the Wii and the Switch. I think MS enters that area at great risk to the success of the next generation of Xbox consoles.
Having said that, I can certainly see them going all-in on AI. That fits their business model. But do game devs really have any idea how to take advantage of that even if it's available? Dialog is not just text based, it's voice acted and with good reason. I could imagine some modest computer-generated speech for certain words, but conversations? I have a hard time seeing that the AI tech is anywhere near providing realistic layered emotional dialog with sarcasm and undertones enough to provide a good gaming experience. As any writer will tell you, dialog is mostly about subtext and what is implied, not said (Bethesda games often ignore this rule to their detriment). This is especially true for the big open world RPGs where there may be hundreds of hours of recorded dialog and where it's critical to game reception (thinking especially about CDPR and Rockstar games). Dialog is as much about emotion as information sharing. AI can fake specific emotional tonality, but not in an interactive storytelling way that dynamically reacts to character choices.
Where I would think devs could find uses for AI would be mostly outside of dialog, such as:
1. Character movement. Instead of scripted animations, AI can craft realistic on-the-fly character movements based on muscle and skeletal structure so that characters can be truly dynamic, especially during combat. This also saves a lot of work by devs to build thousands of custom animations.
2. Related to #1 -- realistic looking combat collisions. Current game combat looks ridiculous. We're all used to it, but there's no physics at work in hitting someone. Blood animations spout out of the air, instead of from an actual impact point. For example, if you shoot a character in the left shoulder, he should spin around from the transfer of momentum unless the bullet goes through him (a through-and-through doesn't transfer the momentum to the target) and hits something behind him. Or chop someone in the leg with a sword and even if it doesn't cut their leg off, they'll be off balance on that leg for a moment from the impact. There is ZERO such consideration for that in games today, where they still function more like a D&D session of roll to see if the enemy has been knocked down. Fixing that sort of thing would go a long way to boosting immersion and a perfect use for AI.
3. Masses of NPCs who are at least a little bit interactive and more than just scenery. And to your point, maybe here there could be some AI dialog, because it's not (or much less) emotionally important to the story.
4. Naturally interacting environments. Bethesda games have almost fully interactive worlds, but they give up a lot of other things to achieve this. I could see AI being used to populate environments with a huge array of generated objects in a realistic way. If these don't need to be tracked as individual objects by the game engine, they can remain "scenery" but become interactive so the locations feel real, even if many of the small changes are lost when the player leaves the area.
I also thought about voice control and movement tracking to give more options than the few buttons on a controller, but after the Kinect failure, and like wearing 3D glasses or VR goggles, even with AI, I don't think many gamers are interested in this.
As a gamer, like you said, I don't care about more pixels past 4K. The jump from 1080p to 4K is notable, but even the difference between 1440 and 4K is modest and beyond 4K, not sure I'd really notice. On the other hand, I very much care about lighting and framerate (up to about 120fps, beyond that, don't think I'd notice). I'd say that graphics hardware is just at the infancy of effective path tracing (top of the line NVidia boards just starting to get there), which is a real game changer not just for how the graphics look in a game but for game developers too. If they don't have to manually construct the lighting models and all the trickery they go through to fake it, that's a huge time savings for their artists and graphics teams. For PC, they will still need to care about that for a decade, because of all the low-end graphics cards out there, but this is the one big advantage a next-gen console could provide to developers (not sure chip costs are low enough for MS to include this yet though).
On graphics tech, what I want and where I think we're heading is to 120fps, path traced lighting to completely eliminate the fakery and dev effort required by lower-end graphics, and with much more realistic and complex interactive worlds. I also care about HDR, but that's more a display problem than a graphics card issue. I think many gamers don't realize how much of a difference that makes if they haven't played on an OLED screen -- HDR on LCD doesn't look much different than SDR (LCD's just can't generate the brightness and contrast needed), but on a good high-contrast OLED it's night-and-day different, almost as much as going from black and white to color. An explosion or crack of daylight in a dark room in HDR can actually make you wince and look away from the brightness. That greatly boosts immersion.
Thoughts?
Facebook
Twitter
Instagram