I do think the gap between mobile and home hardware has gotten very small. The S2 isn't a ps5 nor a 4090, but still impressive.

I still maintain the problem is some view ps4 tier as an insult. I view it as a compliment. A handheld ps4 is crazy. I'm very happy and think it is quite powerful.

Correct me if I'm wrong but the player experience shows a very modest increase from 60 to 90, a fraction of the increase from 30 to 60, and player performance shows no increase at all...

So in Experience, diminishing returns are very much in effect from anything over 60 & Performance is flat out diminished over 60...

I'm not saying you personally can't appreciate a significant difference over 60 but this study shows that most people don't.

If you showed a dev these graphs they'd be very unconvincing evidence in terms of suggesting they allocate more of a system's finite resource to >60 frame rate.

Remember, you said;

'Diminishing returns past 120 fps. Sure. But objectively the average gamer does in fact benefit above 60 fps.'

and

'I'm astounded people think 60 fps to 120 fps is diminishing returns.'

So you're incorrect, as clearly diminishing returns kick in at >60

Where we disagree is the graph between 60 and 90, for QoE, is a jump.  Just because the jump is smaller doesn't mean it isn't statistically significant.  It is statistically significant, thus there a clean benefit of 90 over 60.  120 over 90, not so much, which matched my personal experience.  

The argument that 90 doesn't have benefit over 60 is flawed.  

I think it depends on what you are describing as "PS4-tier." If PS4 Pro is also "PS4-tier" then Switch 2 is indeed roughly "PS4-tier" in terms of raw performance and above "PS4-tier" in terms of supported feature-set. 

If "PS4-tier" means actually roughly PS4 performance, then this is only mostly true in handheld mode, which is only about 55% as performant as Switch 2 in docked mode, which isn't too far from the difference between Switch 2 and Series S. 

Switch 2 handheld is roughly about a base PS4 with a modern feature-set. 

Switch 2 docked is roughly about a PS4 Pro with a modern feature-set. 

For running 9th Generation titles, those modern feature-sets matter. 

Modern features matter, for me tiers is a rough estimate.  And roughly I put it in the ps4 tier.  1080p/60fps with a lots of 30 fps, which for me is ps4.  And a lot of it is visual opinion.  DK looks like Ratchet reboot on the ps4, it doesn't look like Rift Apart.  S2 games look like ps4 games to me.  I don't see a big jump.  I want to emphasize this is fine.  Ps4 games are still excellent in fidelity.

Edit

For me, there hasn't been a OMG this clearly higher fidelity than ps4.  I just haven't had that moment.  Based on MKW, DK and BotW.  

So it is mostly vibe-based then? Because even if we are just looking at internal resolution, the Switch 2 (docked) is averaging about 32% more pixels than base PS4 and has a far more stable frame-rate in Cyberpunk 2077. And that is with DLSS eating up some of its resources. 

DK spent a significant share of its development as a Switch 1 game, and it doesn't really use too many 9th Generation features. It's not using DLSS either. Though I would say its voxel engine probably would struggle on base PS4's CPU, given that the frame-rate drops on the Switch 2 seem to correspond with CPU-heavy loads and Switch 2's CPU is moderately better than the base PS4's (and slightly better than the Pro's.) Plus voxel engines are CPU/memory hungry workloads. 

Edit: 

And of course Ratchet & Clank Reboot is a 30fps game on base PS4. 

Partially based on vibes, but also based on how demanding games are. On paper the 5090 kills my 4090, actual real-world performance, 5-8% improvement.  And 5-8% means nothing.  I don't personally believe a 30% increase in power means much anymore.  I wouldn't upgrade my GPU unless I was looking at a good 100% jump.  Games are just too demanding to get meaningful benefit from 20-30% power jumps.  

We're talking about a >30% improvement in a real-world cross-generation game in docked mode, not "5-8%." 

Again, Cyberpunk 2077 is running with 32% more pixels on average, sticks to its 30fps target more consistently, has higher resolution textures and superior scene geometry/LoD management, and this is with DLSS cutting into resources to provide an even better effective image-quality than +32% (without it) would provide. Where the game has the most issues on Switch 2 is in content that isn't even on PS4/Pro.

We're not talking about something "on paper" like floating point performance here, but the actual real world performance in a benchmark title.

When you take into account that both incoming GPD Win 5 and AYANEO Next 2 both pack Ryzen AI Max 395+, which is in tests around PS5 level, yeah, it is very small difference. That said, it's questionable if one can really call them "handhelds" anymore - I'd say something that comes with that much TDP can be called portable at best, not really a "handheld", due to how heavy (nearly 1kg) it is because of the battery needed to run it (GDP Win 5 even has battery that's not inside it, but attachable to its bottom).

Yeah, Xbox in particular was a beast of a console, but it cost Microsoft a fortune, they were selling at huge losses, and most of all it launched like 21 months after the PS2. Dreamcast is also from the same generation, and was by far the most powerful console until the PS2 came out later. PSP is pretty much a portable Dreamcast.

I used PS2/PSP because it's the most relevant comparison to PS5/Switch2 with launch dates in mind (PS2/GBA is certainly not!). GameCube and Xbox launched 18-21~ months after the PS2. Back then, a 1-2 year gap typically meant significant advancement in technology. Dreamcast launched under 2.5 years after the N64, yet the difference was generational. In the handheld space, technology moved even faster. GBA (2001) and PSP (2005) were like 2 generations apart lol. We're talking 2D vs 2nd gen 3D. The price difference helped of course, but still crazy when you think about it.