Show previous messages
April 16, 2020 at 4:56pm
Currently only single touch input is supported in WebXR mode, so scaling is kind of tricky. Multi-touch is coming along with overlay support, and we have scaling enabled if you try it in Chrome Beta today.
As far as why it isn't working for you, your glitch totally worked on my Pixel 3 so you configured
<model-viewer>correctly. Make sure you are going to the
https://version of the live demo; WebXR requires SSL and a valid certificate.
Looking forward to two finger scale/rotate! Also, do you think supporting occlusion is "on the horizon" or "far out"?
I think if you compare it to the overall timescale of WebXR (which, to be frank, is measured in years), it's "on the horizon"
There is active work going on to implement features like occlusion, lighting estimation and other fancy things like that in Chrome
When these things will land is somewhat uncertain and very much out of our project's control
In particular, it's worth noting that many feature releases in Chrome are being delayed in order to ease the burden on the team during the COVID-19 pandemic. Most folks are being asked to focus on stability / bug fixes for now.
April 20, 2020 at 9:56am
Welcome ! And thanks for joining the chat 🙌 I hope it becomes a useful community resource
April 22, 2020 at 8:02pm
We are continuing to support declarative 3D model rendering..
welcome! The rationale for dropping Magic Leap support can be summarized as:
• Magic Leap support isn't free; it costs some bytes to include it in our library • As far as we can tell, very few people are using this feature in practice • It is still possible to use an
<model-viewer>, and that may even result in a better experience in a lot of cases • The future of Magic Leap the company is uncertain; they are reportedly undergoing a pivot and shedding employees: https://www.theverge.com/2020/4/22/21231236/magic-leap-ar-headset-layoffs-coronavirus-enterprise-business-shift
To review the current status, we aren't dead-set on removing support. Particularly: if we learn of folks using the Magic Leap integration in production today, that would be a strong indicator that we should retain support.
obviously we have a common goal with regards to declarative 3D content on the web. I would love to work closely with anyone on the Magic Leap side to figure out how we can best serve users in that regard.
However, it's not obvious that our Magic Leap integration was useful to users the way we had integrated it. As far as I'm concerned, all possible futures are still on the table but we need more data and user anecdotes.
April 23, 2020 at 3:25am
Thank you for your response. The recent developments have not been helpful. We do have the same vision for declarative models, in an ideal world, we can all agree and standardize a declarative tag
<model>(like how <video>) and all platforms support it natively. But that surely is an uphill and time consuming task at the standards body. We hoped that <model-viewer> would fill that void in intermediate time. At a better time, we could have solicited your feedback and improved our <ml-model> tag to more closely align and provide that consistency for web developers. We still hope we can attract a good chunk of customers and users for the platform and be supported by <model-viewer> in future.
April 23, 2020 at 8:19pm
Also, I'm sad to see ML go through this; I do believe in the future of AR (with all the relevant platforms) and would love to see it succeed in businesses (at least for now - hopefully consumers catch on later).
April 24, 2020 at 4:38pm
Yeah, I would like to second comment. The folks on our side are working in this domain because we are really passionate about the kind of cool things Magic Leap was delivering to us. It's hard for all of us to take in the recent news (although I'm sure it's that much harder for you). We still hope that there is a turning point for the technology if not the current business model.
If you have the time and energy, I would love to have a longer dialogue with you or anyone on your team about the future of Magic Leap holograms on the web. I still believe that there is a ton of potential there, and we would really love to explore better integrations with Magic Leap with a forward-looking perspective.
May 9, 2020 at 7:55am
October 14, 2020 at 5:02pm
November 29, 2020 at 4:20pm