Traditional photogrammetry produces polygon models which are completely different from radiance fields that splatting does. I'd compare the difficulty to editing a vector image vs editing raster image.
Photogrammetry creates a polygon soup that is famously hard to edit.
Yes you can mitigate the problem. But the same is largely true for splats as well.
It's fair to say one is easier than the other, but it's a difference of degree rather than of kind. And mesh tooling has the benefit of a few decades head start and several orders of magnitude more investment.
It solves the problem with hand tracking if done correctly (and only Apple has come close).
Hand-tracking is a terrible idea without micro gestures and eye tracking - because your arms tire very quickly and pointing at things is very inaccurate.
There's a reason why brushes are more popular than finger painting)
Of course - we've already invented the optimum solution which is tracked 6dof controllers - but someone decided they aren't ever going to be mainstream so here we are...
(If that's all you want from them, aren't there cheaper solutions? These are "full VR in glasses form factor")
Not sure. I mainly still use a Quest 3 as a monitor replacement and had no strong reason to switch. The prescription inserts I got with these weren't quite right for my prescription and I prefer using my actual spectacles. So I've only used them to port stuff to/develop for - and tinker with existing apps (mainly the google supplied stuff as the embargo only lifted yesterday so there wasn't much released for it specifically).
I haven’t seen any device with the stability capabilities of Hololens 1 or 2. Would love to be proven wrong soon. The HLs use sensor fusion in hardware to adapt the current rendering at up to over 200 fps during device movement.
Hololens was ahead of its time. I have not used the Vision Pro much but suspect it's the closest and other headsets will start catching up in the next year or two
The devil is in the details. Every single UX design choice can have massive impact on productivity and creative tooling workflows. Don't predict anything based on the spec sheet.
If you don't accept this particular argument then there are plenty more. This one is just quick and easy to convey.
There are very solid arguments backing up the same general points that require time to communicate and require good faith attempts to understand from the other party. But you probably don't want those either.
So do it. Everything AI assisted I've made has involved a ton of work. The only difference AI has made is that I've increased the level of my ambition and avoided a ton of things that were previously blockers.
I think whether it requires a lot of work is not the most important factor. The important factor is how easy it is to replace it. If it was made with AI I expect it to be easily copiable with AI. Therefore it doesn't make much sense to spend a lot, if any, money on it.
reply