I'm not sure what value there is on speculating at this level.
We have no idea what sort of gaming content they are planning for it in addition to the AR stuff. At that price point it's hard to tell what the market or usecases with be.
1. https://www.bloomberg.com/news/articles/2022-12-01/apple-mix...
So for VR, foveated rendering could mean the difference between ~2k and 3k is a wash. And AAA games designed for VR tend to be a little more conservative on the graphics. And the chip will only theoretically have to handle half the other game work and no full macOS.
Combine that with Apple's continued Metal development, which is very capable if you're willing to do the leg work to take advantage of it, and yeah I think it'll get close.
Doing the math to the OPs numbers. 6k vs 2k = 3x. 90 fps vs 30 fps = 3x.
That an extra 900% performance difference right there.
Reality, out the window.
FPS alone is +200%, but games are running above 30fps already so it's less than that. Add a reduction in system load due to not running the full OS, potentially framework improvements as we usually see at WWDC, and slightly reduced graphical fidelity that is typical for VR games, and we're back to the same ballpark.
The M2 doesn't even hit 30fps at 1080p. You also have to consider the power draw limits they set on these 2 chips placed near your eyes compared to a M2 running in a laptop with lots of heat dissipation.
> an M2 chip per eye
first, the definition on an M2 chip is very murky. There is a massive difference between an M2 targeted for small mobile, compared to a M2 found in a laptop.
A mobile can't sustain power draw of 5 watts for very long.
A headset can't sustain power draw of 10 watts for that long.
AAA graphics needs >40 watts of power. You can't have that next to your face without large amounts of cooling. Now, I'm assuming that Apple will offload graphics to phones/tablets/laptops in the style of oculus link. This solves a bunch of other problems like weight, size and battery life.
This is where Apple has the great advantage of course, not only does it control the entire vertical there, but they actually have a QA team, product metrics and leadership to make sure that the link will be seamless, easy and reliable. Meta will devote a team to it for a year, who will build a totally new system, sling it over the fence and move on to make sure that they get promotions. It never really worked of course, but it pushed the promotions metrics, and the leadership don't care about users, only moving a synthetic metric so they can be made VP of cables.
I digress.
However, for remote rendering to work reliably, you'll either need decent wifi(new time machine anyone?) or a decent quality cable. Wifi is a right shit to make work reliably at full throughput, if not impossible. so I'd assume cables would be the way forward (or some fancy 60gig networking stuff, but that seems unlikely)
> Micro LED for a big improvement on the screen door effect, brightness, quality
Nope. The screen is not actually the limiting factor, its the optics. The "barn door" and brightness is almost entirely down to the style of lenses.
My point was that it seems consumers don't care about expensive high performance VR headsets. They want something that is both easy to use and inexpensive: The unusually early Quest Pro price drop suggests as much.
It's like the Nintendo Switch: If it had cost substantially more, much better graphics, screen resolution etc would have been possible, but it almost certainly would have been a flop. The same thing seems to be true for VR headsets.
May be far fetched but, it could be a monitor replacement for offices; got supplied with a mbp and 4 monitors, 2 for at home and on desk on location, some coworkers 6 in total; that monitors alone add up from € 2-3 K fairly quick (fairly nice 27” 4K displays). I’d gladly take the headset over the monitors if it’s not to heavy; The ‘AR’ mode could mean I would be able to see my laptop screen and keyboard and display the other screens using the overlay and switch between them (& other operations) using hand gestures/ moving my head/ looking somewhere. Would also save me a ton of space.
Edit: rxOS instead of iOS, the marketing material seems to mention day to day usage mostly
That's not really accurate. The rumours suggest optics at least as good as the Meta Quest Pro ($1k), possibly with a substantial move forward from there. They suggest resolution of 3k per eye, significantly above the Meta Quest Pro, and using Micro LED for a big improvement on the screen door effect, brightness, quality. Then there's the compute, with 2x M2s, this is quite possibly going to be able to run AAA VR games (if they're compatible with iOS) – the M2 is running some AAA games on things like the MBA, and this one has an M2 per eye.
These rumours might turn out to be rubbish, but they seem to be credible rather than speculation, and they would suggest a high price tag. If this was a Meta Quest Pro "Ultra" it might be >$2k based on Meta's pricing, so Apple charging $3k for it given their "tax" (and high production quality) feels about right.