goodcjw2

Guess this really shows how much domain specific knowledge in the Computer Graphics...

Yet still, this post is now ranked top 1 on HN.

But that headline will not be interesting at all to be news worthy...

Great work to publish your work into public domain. I saw a couple more people in the thread trying to do the same thing.

Just curious: is there a reason to create your own site for this? Instead of listing on things like Sketchfab?

They seem to support public domain for a long time already: https://sketchfab.com/blogs/community/sketchfab-launches-pub...

"Okay, I'm focused on getting X, but let's not forget to read the headlines." -> this is probably the hardest part of "being lucky".

Because it's a tough sell to put a camera right in front of people's TV. IIRC, the only product managed to do this is Portal TV from Facebook...

I actually tried something like this and talked to lots of potential users about this (during 2020 pandemic). But the willingness to pay for the camera is quite low.

+1, System76's build quality is terrible. My company spent lots of $$$ on system76 workstations, only found that GPU card falls off during shipping due to the terrible design they have...

This is definitely an interesting test and the motivation is totally valid.

But I'm wondering whether I can pass if I don't speak up at all? Or just passively play along?

Unpopular opinion: Mozilla has some unique advantage to win the 3D Web.

1/ Hubs and spoke (https://hubs.mozilla.com/spoke) are trying to solve the problem of how to composite and deliver 3D assets over web protocol.

1.1/ Unlike jpegs that can be passively embedded into web pages, 3D assets requires active interactions with other objects.

1.2/ Traditionally, we need a 3D game engine to pack all the assets into one big binary, then build and ship the binary over cartridge, DVD, or recently Steam.

1.3/ Spoke + Hubs is the new web-based 3D engine, which can potentially revolutionize how we create and access 3D content.

2/ Being a popular, open source, and independent browser company gives Mozilla the position of leading the effort to standardize the convention of the 3D Web.

3/ This is low level tech and worth investing in. Also Mozilla makes decent money (> $500M each year) and this is something they can comfortably afford.

4/ Virtual meeting space is just one use case. Admittedly Hubs' implementation is less than super well polished, but I don't know whether they want to be the 3D Zoom in the long run. But I will say it's pretty neat if we treat it like a tech demo for their Web-based 3D engine.

In my personal uses, It's actually not too bad once put on a case.

Wow Tony Stark and Cap America finally meet IRL!

1/ There is definitely potential there. Spatial audio, shared white board, avatar in HD, etc.

2/ The biggest obstacle for now is still comfort. VR is now comfortable enough for 30-min sessions for lots of people, but still not for everyone. For those who complains about Zoom fatigue, VR as of today is just 10X more painful. We need better hardware. There are obviously lots of room to improve.

3/ Horizon Workrooms is not the first and spatial.io already moving towards that direction. After some 30-mins trial, I feel Horizon is more polished and feels better than spatial.io already.

My two cents: metaverse towards gaming is like TikTok towards videos (esp. short form videos). It's all about user generated content.

Video -> Gaming

Blockbuster movie/TV show -> AAA Games

Youtube -> Indie Games

TikTok -> Metaverse

The idea is that anyone can create their own game (mini-game) easily, just like anyone can create a 10 seconds video easily. Instead of bet tens of millions into blockbuster movies or AAA games, it makes some sense to look at "short form" video or gaming. With enough users and creators, and more importantly a good recommendation system, it doesn't require too much imagination dream something big. Given that both gaming and digital video content are 100B+/yearly market, it is fair to guess that one of the next big things like TikTok is indeed something in that user-generated gaming content market.

One way to think about AR is that it can do everything your Oculus Quest can do, but in the form of a pair of glasses that you can wear all day (like those prescription glasses that people already wearing).

VR is like the PC you use at home / in office, AR is more like smartphones on the go. There is obviously quite a big technical gap :)

Properly head-mounted AR technology (ideally in the form of a pair of glasses/sunglasses) can have lots of potentials, but does require science-fiction level of imagination to visualize those use cases.

Unfortunately, this is really hard. Mark Zuckerberg calls it "one of the hardest technical challenges of the decade". There are lots of fundamental technical breakthrough required to make it a reality. In different prototype labs, people are excited to see feasibilities being proved out in individual fronts. But moving all of them into production ready consumer grade hardware is a completely different story.

Give it another 5 years, we might be able to see whether it's indeed possible.

[1] https://www.zdnet.com/article/mark-zuckerberg-calls-ar-glass...

Yeah, that's the part I don't quite appreciate what FB is doing on VR. They heavily subsidize the hardware, pushed lots of competitors out (e.g. Vive is really struggling), then start selling ads once they got dominant position.

It's probably true that they cannot resist cloning existing feature that's already working.

But to cheer you up, FB is putting lots of efforts (rumor goes 10000 folks) working on AR/VR. That's quite a lot innovation :)

I was wondering if https://www.jefit.com/ was going to pop up. Sounds quite similar and has been around for quite some time.

There are probably more similar solutions out there. Fitness apps are definitely an interesting space, where there are tons of them, but they seem to rarely marketing themself. Probably the customer acquisition cost would be higher than the custom lifetime value, i guess...

In terms reps counting. Yes, it works quite well :)

This is pretty cool to have more generic GPU-based support.

Might also be related: Tensorflow Lite Core ML delegate enables running TFLite model on CoreML enabled iOS devices.

https://www.tensorflow.org/lite/performance/coreml_delegate

Would nice to see performance comparisons on M1 Mac/iPad for which way is more performance and efficient. (Admittedly TF vs TF-lite is a 100% apple to apple comparison, pun-intended).

+1000 to focus on mobile first. Totally understandable :)

That being said, I think "more gated and complicated process" are the key here. Guess it's easy for web/backend developers to hold to the key and deploy as they see needed.

But for mobile, there are typically only a small set of gatekeeper at the entire company had the key to publish a new release to AppStore/PlayStore. Thus the more process is required.

Neat! Looks like a great release tool.

I'm wondering whether there is a specific reason to mobile apps?

Would be nice if all releases (web, mobile, etc) can be going through the same pipeline. After all, those "people-centric" tasks are generally applicable to other projects as well.

$3~4B is indeed quite big budget. However, it's dwarfed by big tech's R&D budget. FB alone spends $20B+ each year.

I guess one key difference is that unlike FB, DARPA money doesn't need to be confined to make quick and visible returns, thus they can focus on more long term problems.

[1] https://www.macrotrends.net/stocks/charts/FB/facebook/resear...

Disclaimer: I used to work and knew a lot Fuchsia developers very well.

No, Fuchsia is not a pet project. Instead, it's solving a very realistic problem: building a clean driver interface, while keep it open and gain hardware vendor support.

Anyone have shipped a Linux-based embedded system (including Android smartphone) knows the pain: there are tons of ad hoc driver source files need to be manually merged/rebased against the Linux Kernel. Want to upgrade the Linux kernel from 4.x to 5.x? Good luck redoing lots of merges and rebasing. There are tons and tons of git branches floating around, crazy examples like: "Linux-4.11-some_soc_vendor-some_oem-random_device-random_project-rev-1.3". It soon becomes unmanaged and device manufactures just gave up on keeping up with OS upgrades.

Overall, the monolithic natural of Linux Kernel and its strict licensing terms made it hard to get code upstreamed back or to write properly modularized driver extensions. Even when people managed to get it work, you end up something like AMD's GPU driver takes 10% of Linux kernel [1].

The term of OS is heavily overloaded. But Fuchsia's goal is NOT to build other Android, but mainly to replace the Linux kernel.

Is it worth it? Can Google pull it off? I don't know. But it probably worth a shot.

[1] https://www.phoronix.com/scan.php?page=news_item&px=Linux-5....

A benevolent guess would be: It's probably related to their AR/VR effort. They are basically betting on AR/VR (incl Quest, Portal, etc) to be the next computing platform and voice input (also output) is a key user interaction model. IIRC, they are building a team to replicate Google Assistant and Siri.

How about just using Apple Watch as a display? The display has similar size. Maybe being able to have the display already there is useful?

Nowadays games assets are just crazily huge. AAA games tends to ship at 50G. Now think about the raw assets size and saving all the intermediate revisions.

It's quite easy to burn 1000G of storage on GitHub/GitLab (again, don't forget all the revisions). That puts just the storage cost at $6000/year. At this price point, it's really worth hosting on your own.

As Danieru mentioned they are forcing ppl to manually do cleanups, that probably indicates the storage costs are even higher which worth manual interventions.

Exactly. For us we are dealing with lots of large gaming assets and those are burning through those $50 data packs like butters :)

A side topic: is there a concrete reason why github's LFS solution has to be so expensive?

IIRC, it's $5 per 50GB per month? That's really a deal breaker to me and wondering whether people actually use LFS at volume will avoid LFS-over-GitHub.