yep

Anchor points on the windows!

It looks like a webgpu driver issue - can you tell me what model of graphics chip you're using?

RTX 5090 with 610.62 drivers, Windows 11 25H2, Firefox 153.0b9.

FWIW I tried the other one linked here[1] which also throws the same error and stops. Using the same engine perhaps?

[1]: https://news.ycombinator.com/item?id=48986175

Hmm, today it works all of a sudden. No changes, no reboots, no new drivers, not restarted Firefox either. Just repeated what I did yesterday: open the link in a private window.

Yesterday when it didn't work I did try multiple times, and as mentioned that other site as well.

Guess Firefox on Windows has some flaky WebGPU support?

edit: must be, worked fine a couple of times, then it stopped with same error as before even after multiple attempts.

thanks for double checking, that's very good to know!

Yes, I also added a bit of context on twitter (https://xcancel.com/fulligin/status/2079296985811873796) and will probably blog about the process a bit more. Grace was a lot more work than Sutro. I wanted to be ambitious in combining interior/exterior scenes in a new way, and I also wanted it to be less of a "toy". Grace actually plans to use this to show prospective renters the space, and we spent a lot of time making sure it could run well on a wide variety of devices.

Looking forward to it. I'm also a big fan of https://vincentwoo.com/2023/07/20/tunnel-vision-an-unauthori..., thanks for keeping the internet interesting.

Beautiful work, thank you for sharing it!

In case it's useful, given your last sentence, it runs really well on Chrome on Android (on a Pixel 10) but it's a bit glitchy (although not impossible) on Firefox on Android. The loading indicator on Chrome got stuck briefly whereas the Firefox one was smoother, but that's about the only case there Firefox is better.

FWIW, it works great in Firefox focus on an iPhone 16 Pro.

So beautifully done, its mindblowing. Please blog about your process a bit more.

absolutely beautiful. I'd love to learn this process and attempt to capture a local rock climbing route.

anything you write about the process, or how you learned are very welcome.

Thank you!

I spent a lot of time hand-masking out inconsistencies between shoots. I waited too long between sessions and they actually built out an entirely new section of ceiling. PPISP guidance in splatting also helps a lot in solving sensor inconsistencies between the many different cameras and lighting conditions that went into the source dataset.

Do you see potential for automation based on what you learned from your hand tuning?

The article is very clearly human written. The author discussed on twitter that he used Claude for visualization and feedback. You can simply say you don't like the writing style, but it is not a very useful comment to have written in any case.

Are you implying that the federal reserve bank of New York has an incentive to mislead? Did you examine the study methodology?

Yes, and it was completely crap.

Stopping at 2024 might be reasonable because you don't have data for 2025 and the current state of 2026, but it leaves a massive hole in AI affected time frames.

There was absolutely nothing there that could even imply that AI isn't the current problem for young unemployment. It's like saying that your data only goes up to 1700s, so therefore global climate change is probably more affected by the Boston Tea Party than industrialisation.

It is not by the federal reserve bank of New York

> The views expressed in this post are those of the author(s) and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the author(s).

Also, they studied one company

> Working with proprietary data from a Fortune 500 company, we are able to shed light on the underlying reasons for these labor market changes.

So, yea, going to call BS. Most of their work is on the effect of remote work within a single company and they are tacking on AI as it is topical.

sad of state of all shorts of media lately

It is very amusing to read HN comments that complain about the "enshittification" of free platforms while simultaneously mocking those who would pay for stuff they like. YT is dollar for dollar the best digital subscription I pay for and I pay it gladly.

I don't think one should mock people who are paying for youtube. If they find it's good, then it's laudable to pay for it. But that said, I personally can't relate to that position. There just isn't any content on youtube that I find interesting enough that I would pay for it. It's a time waster for me, not something I seek out because it's a compelling way to spend my time.

Youtube has reached the terminal stage of enshittification:

• the good stuff is VHS-quality TV content that somebody pirated

• the ads, once nonexistent, are typically disreputable and now incessant

• the few 'creators' worth watching are lost in an ocean of audience-captured, brain-dead garbage "hey guys... [product placement disguised as organic content]... misinformation... remember to like and subscribe... [product placement disguised as organic content]"

• access becomes increasingly arcane due to ad-blocking measures

• one of the lowest quality comments sections - largely inorganic, rogue state-sponsored - on the internet

• increasingly just AI slop

The day I can't scrape videos via yt-dlp is last day I permit youtube domains on my network. Personally, I would prefer to eat a rotten cat carcass than pay a single cent to Youtube.

In a better world, youtube would be some kind of a protocol, not a mediocre company serving as a middleman.