By using this site, you agree to our Privacy Policy and our Terms of Use. Close
EpicRandy on 19 February 2024
Edit Thread
shikamaru317 said:
Wman1996 said:

Xbox is now claiming that the next Xbox (it's now confirmed Series X/S won't be the last) will be the largest generational leap of all-time. I do not understand that strategy. Xbox already has broad support for PC gaming, so why worry about a powerful console that will still fall short of some very powerful gaming PCs? Software and Game Pass will sell Xbox consoles, not big specs.
If we find out that Nintendo's upcoming platform is one of their biggest (if not biggest) leap in specs of all-time, what is the point? Is Nintendo really going to benefit from a docked Switch 2 that would outperform a Series S and get closer to a PS5? Nintendo customers (whether adults buying for themselves or for family members) typically come for the Nintendo software, a little bit of the third-party, and some Nintendo charm/quirks. Nintendo will just pass the costs onto us. A Switch 2 that is more powerful than Series S would probably be at least $500 USD. I really don't want that, especially if it's the only way to get Switch as a hybrid without a cheaper hybrid SKU.
Gaming PCs long ago outpaced consoles, there's no use in trying to get way closer to them if the price is too high.

I went into this earlier in the thread, but I think the whole "next Xbox is the biggest leap of all-time" is largely marketing spiel. If you look at all console generations before now, there were some absolutely massive leaps that have happened, some of the biggest being SNES to N64 and PS2 to PS3. Even if Phil didn't mean across all console makers, but simply that it would be Xbox's biggest leap yet, that would still be highly unlikely as OG Xbox to 360 was a pretty huge leap, the GPU leap alone was from 14 gflops to 240 gflops, a 17x increase in theoretical GPU performance. Xbox One to Xbox Series X was only a 9x increase in theoretical GPU performance by comparison, and that was a 7 year generation, we're looking at a 2026 release for the next-gen Xbox if the leaks are accurate, only a 6 year leap in technology. 

On top of that, even if on-paper specs were somehow the biggest leap yet for gaming, due to the perceived phenomenon of diminishing returns, graphical leaps looking smaller and smaller each year the closer we get to realism, this biggest leap yet in gaming technology wouldn't be perceived as such by the layman, not compared to those biggest leaps of the past like SNES to N64 or PS2 to PS3.

Realistically, I just can't see Xbox releasing a console for more than $600, and the best specs Xbox is likely to be able to manage for $500-600 in 2026, the leaked release year for the next-gen Xbox, is probably roughly on par with the current Radeon RX 7900 XT in rasterization performance and above the 7900 XT in ray tracing performance (due to AMD planning big ray tracing improvements for their upcoming RDNA 4 and RDNA 5 GPU's, with Xbox supposedly using RDNA 5 for their next-gen GPU). To give an idea of what kind of leap the next-gen Xbox will likely be in realistic terms over the Series X, the Series X GPU is roughly equivalent to the Radeon 6700 10 GB on these charts, and we'd be looking at roughly 7900 XT tier performance on the top chart (rasterization) and roughly Geforce 4070 ti or 4070 ti Super tier performance on the bottom chart (ray tracing) for this next-gen Xbox releasing in 2026. 

So basically about a 2x leap in framerates over Series X in rasterization performance and a 3x leap in framerates over Series X in ray tracing performance. Not exactly a leap large enough to be worth Xbox making a big deal about it in their marketing next gen. 

So, I see one of two things happening here in regards to why Xbox is claiming this will be the biggest technical leap ever for a console generation:

  1. This biggest leap ever will be a perceived leap not due to actual on-paper specs, but rather from AI techniques. The next-gen Xbox will be able to make good use of AMD's FidelityFX Super Resolution AI upscaling to not only render games at much lower native resolutions to free up more of the GPU for boosting other graphics settings (which is already being used in many games even on Series X), but also for AI frame generation. Frame generation in the current FSR iteration, FSR 3, can already push about 80% higher framerates in many games, and we will see further improvements in upcoming iterations like FSR 4 and 5 before the next-gen Xbox releases. With AI frame generation, those above 2x leaps in real-world rasterization performance and 3x in ray tracing performance, should roughly double to a 4x leap in rasterization performance and a 6x leap in ray-tracing performance over Series X.
  2. In the slides about the next-gen Xbox that leaked out during the FTC vs Microsoft-ABK merger trial, Microsoft suggested the next-gen Xbox could be a hybrid cloud console, with native hardware handling some of the graphical load and a cloud server handling the rest of the graphical load. If Microsoft is still planning that hybrid cloud console for next-gen, and can make the tech work by 2026 (doubtful imo as they were aiming for 2028 before and even by 2028 may not be able to make it work), the next-gen Xbox could be perceived as being the biggest technical leap ever thanks to cloud boosted rendering exceeding the graphical capabilities of native hardware alone.

I think Sarah said, the next hardware will be the largest technical leap and did not precise what she was referring to.

A technical leap can be defined in many ways, not just by performance but by features also. if they make a console that has an open OS and that bridges the gap with PC, it doesn't need to be any more powerful to be qualified as a large technical leap for instance.

And IMO, if it is about the power output, I think it makes more sense to view things in absolution terms and not ratio. Just like politicians claiming every term that there hasn't been any prior time with as many people on the job market as now and attributing the merit to themselves, while completely ignoring that the pool of people available for work is always growing and making this fact quite an obvious and predictable outcome. Here when I see the biggest technical leap in console space, in absolute terms it was from Xbox One X to Series X with ~6tflops of added performance, so a new console to have to be considered the biggest leap, using a politician approach, only mean that it need to be at least ~6.01+tflops of added performance.  

Last edited by EpicRandy - on 19 February 2024