arthur2e5

Thing should get its own submission.

War Thunder kind of has that, except that the "progressive" bit means you need to download the full thing to progress in the game beyond a certain point.

Enlisted, which uses the same engine, is fine though.

The publications all appear to do what the README is describing, except for the part where the README says bugs are preventing some results from being reproduced.

Going through the stuff chronologically seems to help understand how the ideas got put together. The 2017 MS paper is the oldest of the four and mentions a weaker MSPrecise method using summary statistics. I mean… if we really are sure some two things are somehow related, you might as well throw ML at it and see what it comes up. It's just linear, no super fancy models.

Add some agitation device or just do it in ultrasound -- maybe? Still need to figure out automated draining and filling.

Sous vide is long waterbath at around 55 C or lower. Not a typical plastic safety issue temperature.

Cooking in plastic (or any "original package") has a food safety point in restaurants. Keeping it sealed until the outside is sterilized reduces possible contamination. Immersion also makes for easier and faster heating.

1) was a problem in the early days, but what they poured into clang did eventually pay off. It's a reasonable compiler with tons of vendor extensions, many being Apple's, these days. Clang also served as a source for GCC to take new features from. Good to have some competition.

2) XQuartz took a while, but is doing well.

3) was a big one in making upstream wine not work. with wine32on64 you can sure still pull the pinball party trick, but building that thing takes too much.

4) was specifically SIP for me, especially when they started making updates break with system changes. The ntfs-3g automount thing has always relied on replacing mount_ntfs [which is, by the way, not the proper way to make an fs on a Mac -- there's some filesystem bundle business], and now it's too much of a pain to still access the Windows drive.

Telling them to fuck off did not work with Gui Minhai, who was a Swedish citizen at the time. If you're in some way Chinese they can just do the dragging first and make you say the "I wanted to come home" line later.

I mean, yeah, just don’t do phone inspections and go straight for the wire. Poison the DNS, cut any known IP ranges, and snoop on SNI for good measure. That’s the school and enterprise way of doing it. It’s not that evil as long as you don’t use it on unconsenting civilians (see China)!

For what is worth, 256 / log2(10) is around 77, while the observable universe is estimated to have anywhere from 10^78 to 10^82 atoms. An 256-bit address space, if fully utilized, would produce Asimov’s AC from The Last Question.

More realistically though, we would throw away at least half that length like how we are handing out /64 blocks to everyone on IPv6.

That’s the "Patel" programming language used to define Iosevka. See https://github.com/be5invis/PatEL.

Smoking and alcohol are unfortunately too convenient as coping mechanisms go. A stimulant-anxiolytic to keep you going through a shift, and a sedative to slow down and stay numb. And they (nicotine especially) feel just too effective—at the start before you build up dependence at least.

Lifestyle modification can only go so far when the stress of living remains high. Swap the burnt stuff with something that puts less tar in your lungs, maybe, but any further is… sigh.

It’s kept by patenting of the process involving the bacteria and the bioreactor, I think. Google Patent query: https://patents.google.com/?assignee=%22Solar+foods%22&oq=%2...

Now CA3153196A1 is the interesting one as it names Xanthobacter sp. VTT-E-193585 as the strain of interest. You even get a 16S rRNA sequence with it for doing some taxonomy! This is not to be confused with FI129574B involving a similar but genetically modified bacterium.

… Which makes knowing what microbe it’s using even more interesting. I mean, H2-eating microbes are quite common (and are likely how life started around seabed thermal vents), but making these things edible?

(Patent says it’s a strain of Xanthobacter.)

16GB optanes with an ordinary SSD interface are pretty widespread in China and DIYers do use them for swap/pagefile. The problem is that they take up a whole M.2 or NVMe slot with the corresponding number of PCIe lanes…

These drives generally come from budget laptops where the optane is originally used as the system disk. Not a good idea anyways and computer shops usually swap them out. You pay about 28 CNY for 16 GB, which is 2x the average SSD price-per-gig.

The format of GPS civilian signal is public and you could just spoof it with any transmitter like a HackRF SDR box. There’s no cryptographic signature to verify or anything. The military code has had cryptographic anti-spoofing from the earliest days.

Or you could just send any super strong signal on the frequency to cover the satellite signal up. That can also work.

No UTF-8 in DNS. Well, not any more.

B12 is basically never made by eukaryotes. We humans use bioreactors or traditional fermentation cats to have bacteria do the job. Ruminants like cows get B12 by fermentation in the gut. Animal with smaller guts like chicken need it in food like us, generally via fermented food. There is no significant difference in bioavailability that I know of for B12, but there’s an argument to be had that meat is more palatable than pills or weird fermented acquired-taste foods.

Iron is probably the only nutrient I can immediately think of where there’s a significant (~2x) bioavailability difference between plants and animals. Not big enough to make the animal source "more efficient" from a resource use perspective considering the feed conversion still.

If what they’ve done with ECC memory is any indication, this sort of market segmentation would only slow down or even kill adoption of AVX-512 in the consumer sphere. Welp, at least there’s no reliability issue here, so Torvalds is unlikely to get angry anytime soon.

The way ifunc (well, actually language-level FMV) works in GCC and clang is that the input source code, not the command-line switches, specifies what ISA extensions to build for on a candidate function. This naturally means that packagers and other vendors are not using it as much as you hope: they would need to have a separate patch to add these attributes for each architecture and that’s just not maintainable.

Even Intel’s Clear Linux does not bother to patch individual libraries. They just use the glibc library multi-versioning feature to load from different directories depending on the cpuid.

In my opinion, GCC and Clang could make the whole thing more ergonomic. Ideally you declare a function as "interesting for multi-versioning" in the source using an attribute, then in the command line define what -march's to actually clone for. Kinda like ICC’s /Qax. (On second thought preprocessor defs are sufficient, duh.)

The 3D models are different! Extremely cool.

Dense cities consistently produces lower per-capita carbon emissions. This has to do with transit, with transportation of material, with houses (yes, townhouses too!) not having as many walls to lose heat from, etc.

We can and have designed cities that properly handle rainfall. The issue is not density, but the lack of water-permeable surfaces. Suburban parking lots and wide freeways cause the same issues and on a larger scale due to the low density.

The lack of nature is, again, solved not just once. In Amsterdam they have dense urban areas right next to actual farms with cows, free-range ones. Again, you save space for nature by going high density. Compared to sprawling out with cul-de-sacs a you have more people getting closer to nature more often. Same argument for farmers traveling to town.

(I do sense a bit of mismatch between conceptions of non-dense-city living spaces. You seem to be talking about something that combines the environment of rural areas with the convenience of suburbia, but these two seldom co-exist without wasting a lot of tax money on maintenance…)

Horner does not conflict with vectorization in this case. If you read the code, it actually uses Horner — the vectorizability applies to when you call the online function in a loop.

It’s an old article, so… guess I won’t beat a dead horse as much.

* The range reduction is, well, probably worse than acos. Look for Payne–Hanek–Corbett. Not an issue for games, I assume.

* The LERP is nice and all, but Qt actually has an even better one: instead of doing a LERP, you use the local derivative also available from the lookup table. https://stackoverflow.com/a/52841086

Others have already mentioned the Remez and the Horner stuff. Won’t repeat here.

It’s x86-64 only. A quick look at the https://github.com/tenderlove/tenderjit/blob/main/lib/tender... linked from the README reveals heavy references to x86-64 registers. The “Fisk” library used appears to be a x86-64 assembler in Ruby.

I guess that’s to be expected with “pure ruby” — all the cross-insn backends you can use (Cranelift, LLVM) are written in not-Ruby.

Per sibling comment https://news.ycombinator.com/item?id=28614443, just the OS level.

The “AD” name isn’t that far off. Chop the presentation off an advertisement framework… isn’t the tracking and C&C code already spyware?

I have a bit of minor frustration with how many different XML formats there are for scripture. We have OSIS and USX/USFX, both supposedly universal but really just used for the Bible and Torah.

I cannot access the tanzil XML file right now due to MobileSafari layout issues (boo). https://github.com/ceefour/qurandatabase has its own consistent DTD, but I doubt this is the only format I will find.

I vaguely recall go’s safe but unusual TLS cipher suite default being one of v2ray’s early flaws. V2ray was a tunnel designed to look like normal HTTPS traffic to censors, and they were getting very easily sniffed out until they copied the suite preference from a popular browser.

https://github.com/v2ray/discussion/issues/704 is the actual issue; my memory could be wrong.

In elementary school it was just 余数定理 without the Chinese. Well, academia sounding more foreign is normal here.

On the original OKlab Euclidean will do. Perceptually-uniform color spaces are designed for that (at least for close colors).

On these slightly squished and stretched ones… I think converting back to OKlab is the better idea. You can just turn the polar stuff into Cartesian directly too, but it will be distorted.