A normal GrapheneOS installation uses Android Verified Boot with a locked bootloader, and does not give the user root access. Google's Play Integrity cannot be used to verify the integrity of the OS (as GrapheneOS is not approved by Google), but equivalent verification can be done using standard Android APIs and GrapheneOS's public keys.
AVIF supports regular AV1 videos, there is no intra-only constraint. You can try this yourself with any AV1 video:
ffmpeg -i input.webm -c:v copy output.avif
You are right about Firefox and Chrome not using hardware decoding though, as such an AVIF video plays back poorly. And Chrome refuses to load an AVIF larger than 256 MB on my machine.
Are you buying ones that are "Ultra High Speed" HDMI certified? At least in theory, those should all be able to handle the maximum 48 Gbps bandwidth supported by HDMI 2.1
Strange omission to talk about the general anti-AI sentiment, while not mentioning that the #3 most-watched English VTuber (Neuro-sama) is an LLM – who is actually widely accepted as a "good" use of AI by other VTubers and viewers.
Does anyone actually believe the premise of Neuro-sama that she is an AI? The original premise was that she is a lightweight AI model livecoded on streams, but that is obviously not the case anymore.
It's certainly possible someone could fake something like that for views, but I'm curious what you consider to be the tells that it's no longer an LLM behind the wheel?
Maybe it's LLM and human actor splitting work, but she looks too natural, too fast and too error-free to be LLM based since too long ago. Just their dictation -> thinking -> TTS loop is too fast to be LLM. Context awareness and character adherence is too perfect as well.
I think neuro-sama has been accepted because she isn't deceptive. It's very clear that Neuro-sama isn't a real person or even really intelligent. The other big reason is that she collaborates with other vtubers and has brought new audiences to them.
Also, arguably nowadays the main streamer is Vedal, her (for lack of a better term) handler. Part of the fun is seeing them interact
It doesn't let me edit the link, but I'm confused by what even happened here... I posted this from my phone and that wrong link doesn't show up in my clipboard history.
Unrelated: Hey, I sent hn@ycombinator.com two emails. One was May 6th, the other June 18th (UTC+2). The former's subject is "Broken prev/next links sometimes". In the latter, I've asked to let me know if it arrived. It didn't bounce so your email server has acknowledged receipt, but based on fast responses to previous emails and someone else mentioning randomly that you responded quickly to theirs in iirc early June, I'm starting to assume you're not seeing mine. I don't know how else to reach out than via an off-topic comment or a dummy submission or so. Is there a fallback mechanism to use when your email doesn't?
Well, they don't read all the emails -or- respond to @dang, @mod, etc.
Your approaches are, ride a comment (as you did, 9 hours, no response, likely didn't read) or lean on a frequent flyer (the only privilege I have is increased ability to [dead] obvious spam not caught by filters - but it gets my emails seen)
I sent them an email in past minute - Good Luck! (YMMV)
That's the easy part, just use a color space with imaginary primaries (see e.g. ProPhoto RGB), or use one with real primaries that allows for negative values – e.g. Windows uses floating point scRGB for HDR, which is just linear BT.709/sRGB, but negative RGB values can be used to cover the full range of real and imaginary colors.
To be clear, PNG only supports lossless compression, while WebP has separate lossy and lossless modes. AVIF can do lossless compression too, but you're usually better off using WebP or PNG (if you need >8 bpc) instead as it really isn't good at that.
PNG can be lossy before the lossless step. You can take areas of near-matching pixel values and make them actually match, to work better with PNG's near neighbor compression. There are a few encoders that can do that.
