Do any of you people even know how that will play out?
This is just a stealth way of implying the SSDs will power graphics fidelity, as if they are actually rendering the graphics. That's false. Theoretically speaking the faster the SSD can transfer data to RAM the sooner the CPU, GPU etc. needing said data in RAM can access it, but there are many different ways to actually achieve this.
At the end of the day the GPU is what's doing the rendering of visual output, and we know which system has the advantage there (in almost every area outside of clockspeeds, which only affect the rate of cache speeds and pixel fillrate). We know which system has the wider memory bus for GPU-bound data, and faster RAM bandwidth. We know which system likely has the larger L3 cache for the GPU and customizations targeting ML, AI, RT etc.
One system having the ability to load in data assets to/from RAM faster is beneficial of course (and will have its advantages), but there are clever data programming techniques that
both systems can use to reduce the size of even uncompressed data transfers from storage to RAM, and altering that data at run time through GPU programming techniques. And due to the extra GPU headroom, that's one area the XSX seems to have the
overall advantage, and yes that does mean aside from raw TF.
Targeting that advantage, though, will probably require more effort from devs vs. targeting the advantages of Sony's SSD I/O due to the nature of what we're talking about here. But the potential is definitely there, and it'll be easier next-gen than it's ever been in generations past, particularly factoring in whatever software API optimizations (including proprietary solutions) MS is developing such as BCPack.
Hopefully this clears up some confusion; I'm not saying having a very fast or robust SSD I/O complex won't assist in visual fidelity, far from it. But there's a limit to how much that will help when the other system has a more powerful GPU and its own GPU customizations thereof. And no, percentage differentials alone don't tell you everything; the way the systems are right now based on what we know and can speculate, a system like PS5 would need a much larger favoring percentage delta with the SSD than just 125% (and in areas besides simply SSD I/O) to result in visual fidelity that's in any way superior to XSX on a technical/objective (subjective/artistic is an entirely different thing and not really dependent on technical feature sets, or should say not restrained by them vs. in the past) level.
That's just reality. Each system has its advantages, and they both have strengths in areas that will help them "punch above their weight" beyond what paper specs state. In some ways these advantages will even be able to help mitigate weaknesses they have in select areas. But in the role of a system architecture, the GPU is doing much more of the overall work than an SSD or the associative I/O, so the weight of advantages one system has over the other in given areas is actually not 1:1 nor clearly reflected through paper spec comparisons.
It's quite simple
