dear @ID_AA_Carmack would have any opinions on a technology stack for my wildly overambitious AR project
very large scale graphically intensive distributed persistent ar layer over the world. tbh i'm still working out if it's architecturally feasible at all, the answer: barely?https://twitter.com/davidkw_92/status/1281685411681599494 …
-
-
Replying to @DanielleFong
I'm a big believer in tightly coupling the design of a platform with the hardware that runs it, versus a fuzzy abstraction that maybe hopefully works with things that show up in the future. Given that we don't actually have good AR hardware today, I would say it is premature. \
1 reply 1 retweet 9 likes -
Replying to @ID_AA_Carmack @DanielleFong
Still, there is a strong desire to speculatively build, of course. "Graphically intensive" is likely going to be problematic, because all-day-wear AR gear is going to be extremely power limited, and the latency constraints are even more severe than VR. Smart use of \
1 reply 1 retweet 3 likes -
Replying to @ID_AA_Carmack @DanielleFong
power efficient rendering will likely be key -- images and outlines, not photogrametry. I would lean on video as more scalable and bang-for-the-milliwatt than low end rendering. For the on-device client, I would write a fresh C++ codebase with a small team of fully \
1 reply 1 retweet 3 likes -
Replying to @ID_AA_Carmack @DanielleFong
invested engineers, but if you need to hire large numbers of people or make a demo very quickly there will be significant pressure to do everything in Unity.
1 reply 1 retweet 3 likes -
Replying to @ID_AA_Carmack
wow! thank you. i agree with what you write above. the most likely hardware is probably phones for anything practically shippable today. initial versions will try to stay away from expensive rendering and will probably use a lot of compositing and mixing and filters and effects
1 reply 1 retweet 4 likes -
Replying to @DanielleFong
Making a firm decision to target phones to best advantage, and not just as a stop-gap until headsets arrive, would have a strong focusing effect, but pull-out-phone-and-run-app destroys most of the vision for an AR overlay world. There might be an angle somewhere...
1 reply 1 retweet 6 likes -
Replying to @ID_AA_Carmack
I totally agree that pull out the phone and run has much less of the overall persistent AR capability that the visions we like to imagine enable. i’m really thinking more of an extension of Google Maps and Pokémon GO as AR. most of the interaction will be moving within a space
1 reply 0 retweets 1 like
A lot of whether or not the vision makes sense is if the Reality of the AR layer can be appreciated and perceived without having to have the phone out at all times. The heavy duty AR components may not be the key to the design of the world I want to create.
-
-
Replying to @DanielleFong @ID_AA_Carmack
It would be heavy graphics requirements at first, but transition to AR overlay.
0 replies 0 retweets 1 likeThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
