It's probably too early to write off watches thus. Technically they're probably in the 'not ready yet' category: another few rounds of improvement in size constraints, battery life and potential independence from a smartphone (especially as putting in more types of radio becomes more viable) could make a big difference. The aWatch really was too early, or at least too early to create Apple's trademark illusion of perfection in a bolt from the blue. The obvious contrast is the iPod, where Apple didn't enter the game until its distinctive strengths in UI/UX and overall quality could be applied to already-existing components (most famously the Toshiba HDD) and concepts ('Palm-Pilot-like sync', 'WinAmp clone' ...) to create such a lightning-bolt with an obvious 'value proposition' for the user. Or maybe some of the aWatch's faults are actually down to UI or design failings which weren't really forced by technical limitations, but those too can be fixed over time, with MS-style release-until-good iteration.
> VR? Niche market.
No wai! VR/AR is also in the 'not ready yet' bucket; and it's not quite ready yet, rather than being another decade away or something. Between stereo cameras on smart devices, eye-tracking hardware for foveated rendering, and continued increase in flatscreen resolutions, in a handful of years VR is going to reach the 'iPod point' where there are no technical or price barriers to Apple putting together a product that's strongly desirable to a mass market. The initial 'system-seller'/'killer app' probably won't be anything very 'natively' 3D but, basically, 2D screens pulled into the VR space, likely with some modest 3D/stereoscopic enhancements. Think 'VR mobile' rather than 'mobile VR', as it were: VR as a way to escape the screen-is-too-damn-small/device-is-too-damn-big/screen-is-too-damn-close curve that box-with-a-screen-on-it mobile devices are trapped on. Likewise the trade-off between staring down, the gorilla arm and messing with a plastic widget. It won't make box-with-a-screen-on-it devices (including clamshells) obsolete, because being blindfolded by a VR screen is unviable or too scary or dangerous in many everyday mobile-device use contexts. (Mobile AR has a much better shot in those situations, but AR screen technology still faces real hurdles on the way to 'iPod point'.) But it doesn't have to: there are enough wealthy US baby boomers who take long journeys by air that Apple could probably have a hit device if it sold the thing only to them. (And remember, they're all farsighted.) Image sharpness won't be "retina" at first, but it will be good enough thanks to continued progress in LED pixel densities, especially for things that people are happy to do on a now-high-res-but-still-small smartphone screen. (And wealthy retirees aren't the most demanding judges of image sharpness.) Foveated rendering will keep the CPU/GPU burden down to within the abilities of a contemporary Appley mobile device. (Especially since you don't need to put the CPU/GPU on the damn headset! Apple can happily run a cable down from the HMD to an iPad-to-iBook-sized device.) Head tracking will be fully positional and good enough.
