3 ms·
Everything Carmack says is absolutely correct. You will never win an argument on technical details with him. He knows this stuff front and back and inside out.
by moron4hire 12y ago
Everything Carmack says is absolutely correct. You will never win an argument on technical details with him. He knows this stuff front and back and inside out.
But he is way over stating the problem of latency. It is certainly a problem, and lower latency is always better, but it's an engineering problem, one that will eventually be solved. We have the necessary hardware to do this right now, the problem is our current state of the industry puts hardware and software developers way too far apart, and puts startup programmers and mathematicians too far apart, too.
Of far more concern is the UX issue. NO ONE has any clue how to design intuitive UI metaphors for VR yet. Oh sure, first person shooters are obvious. But what about everything that isn't a game? Speaking from experience here [1], it is a pain in the ass to be constantly taking the head set off to switch between apps, to fiddle with the phone, to have to take it out of the headset rig to use the touch screen. What is first and foremost needed is a VR window manager.
Carmack is an excellent mathematician who understand how hardware works almost intuitively. He is not a designer; just look at Doom 3. And he does all of his work in the dark. If we were talking about web technology, people would dismiss a project for being so closed and inaccessible. We've come to expect open source for web tech, but for some reason we still have a view that default state of game tech is closed source. If VR is going to have the adoption rates necessary to have a healthy app ecosystem for users, it needs the low barrier to entry that open source provides for developers. It needs parallel, competing framework development. I don't want Oculus and Samsung to be the only arbiters of how to design VR code. I don't want users with $500 burning a hole in their pocket to be the only users who can experience VR. We can't even get people to pay $3 for a smartphone app now, adding extra hardware cost is just going to make that worse.
We need creative people hacking away at the problem in their bedrooms and garages. Carmack is pursuing a model that requires startup capital just to get to play around. I'd rather reserve that level of effort for real content, so we can eventually have real content.
[1] I'm doing this open source and I'm interested in collaborations. Please check it out and consider contributing: https://github.com/capnmidnight/VR https://github.com/capnmidnight/VR
- rurban 12y agoYou obviously never worked on VR. Latency is the only problem, and far more important than the UX. Why do you think we still have no consumer grade VR equipment, even if all the technical problems and fancy UX solutions have been all solved 20 years ago already? Latency. With a bit too much latency in VR your users will not only loose immersion, they will start vomitting and suing the maker. Google glass was ready for >20 years, but nobody dared to sell it. 20 years all the VR industry only cared about were the various UX tricks (speech, motion, position, floating windows, tiles, ...) but still latency, i.e. HW brought it to a halt. Even with special GL HW. VR window managers were amongst the biggest projects out of Xerox Parc. Everybody had tons of ideas how to design intuitive UI metaphors for VR then. Creative people will play around nifty interfaces for sure as they did during VR times, but we still need the cheap, reactive and not overheating HW first.
- davorb 12y ago> VR window managers were amongst the biggest projects out of Xerox Parc Do you know where I could read more about those?
- moron4hire 12y ago"Start vomiting and suing the maker" is the type of overstating the problem that I'm specifically referring to.