poizan42
Joined 13 years ago
Karma
1,750
Submissions
653
Member since
May 7
[ my public key: https://keybase.io/poizan42; my proof: https://keybase.io/poizan42/sigs/FkOsaGSpCF1HJswaogubPUxO68ewPLKOR7JFPRsF2zA ]
View on Hacker NewsRecent submissions
Comment · 5 days ago
Yup, SawStop do make band saws now[0]
Comment · 21 days ago
They used the xkcd PRNG: https://xkcd.com/221/
Comment · 1 month ago
My guess is that they are dynamically changing the quality of the model to always keep the speed above some floor. So once it gets below that they switch to a worse quant or reduce reasoning level, or some combination of both.
Comment · 2 months ago
How? How can a single absolute number indicate broad success? You can't read from it if failure rate is 1% (completely unacceptable) or 0.01% (probably acceptable as long as problems are quickly mitigated). Also what do they even count as a success? If someone is stranded somewhere for a month and the problem then gets mitigated, they did eventually successfully use it, didn't they?
The strongest indicator that the system is riddled with issues is in what they did not say. If they had less than 100 cases where the system initially failed and the issues got resolved within a few hours I'd bet they would brag about it.
Comment · 2 months ago
> A Home Office spokesperson said: “Over 10 million people have already successfully used eVisas to prove their immigration status. ..."
So how many people did not successfully use it? Could be 100 million for all we know. Do they think just throwing out a large number without anything to compare it to is somehow proving something?
Comment · 2 months ago
It doesn't really explain much about its purpose. Is it better than libavcodec? I take it that the point is to re-use techniques from the h264 encoder for encoding MPEG-2 where applicable?
Comment · 2 months ago
It's crazy, the last time any PRs got merged was back in January 2024 [0]. Seems it has just become a black hole completely ignored by anyone at framework.
[0] https://github.com/FrameworkComputer/EmbeddedController/pull...
Comment · 2 months ago
I have used Opus 5 and some Fable 5 to finally get realtime transcoding of 4K 10-bit HEVC (to 1080p or smaller SDR AVC) working on a Raspberry Pi 4. It was very good at writing optimized NEON kernels. the Argon HEVC hardware decoder outputs SAND30 which is a tiled format that is annoying to work with and not really supported by anything else, the big performance issue has been with converting and scaling it, but as it turned out a lot of it was really with just moving memory around, so by fusing multiple steps into a single kernel it became fast enough. Experimenting with writing the NEON kernels for the different combinations would have taken forever by hand.
I do wonder if auto-research would have reached something similar, it did take a significant amount of steering from me to get it to the point where it was working realtime.
For anyone interested the ffmpeg is at https://github.com/poizan42/jellyfin-rpi-ffmpeg and a shim for using it with stock jellyfin at https://github.com/poizan42/jellyfin-rpi-ffmpeg-shim
Comment · 3 months ago
The proper name for the calendar using the Gregorian system before the start of the Gregorian calendar is "the proleptic Gregorian calendar".
Comment · 3 months ago
Nothing more complex than Hello world is easy to implement in C without risking mistakes in checking boundary conditions if you are not really careful.
Comment · 3 months ago
Maybe there is something else in Gelidium amansii that it needs, if the tokoroten was produced in the traditional way?
Comment · 4 months ago
Windows has its own library built-in: NLS. There have been a lot of back-and-forth about whether to use NLS or ICU in dotnet on Windows because it resulted in different behavior between Windows and other platforms. I think they ended up using ICU by default from .NET 5 on.
Comment · 4 months ago
But wasn't the whole (claimed) reason that it got banned in the first place that it is a logical impossibility? Reviewing code for bugs is legitimate. Writing regression tests for bugs is legitimate. If the bug happens to be a security issue then the regression test may be a PoC or at least a step towards one.
Comment · 4 months ago
m68knommu was merged into the mainline kernel in version 2.5.46: https://www.kernel.org/pub/linux/kernel/v2.5/ChangeLog-2.5.4...
Of course the 2.5 line was the unstable pre-release for the 2.6 kernel. That means that stable mainline linux has had nommu since 2.6.0, which was released 17 December 2003.
So yes, some time indeed.
Comment · 4 months ago
> A statement that the information in the notification is accurate, and under penalty of perjury, that the complaining party is authorized to act on behalf of the owner of an exclusive right that is allegedly infringed.
DMCA Section 3.A v) - https://www.law.cornell.edu/uscode/text/17/512
It is only the part about being authorized that is under the penalty of the perjury. You can be as malicious as you want as long as you don't falsely claim to own or represent someone who owns the rights to something.