Skip Navigation

Posts
1
Comments
299
Joined
3 yr. ago

  • It's not really the intended use case of Lua to be used in things that will be upgraded after they're released. It's meant for things that need to work exactly the same way in a decade as they do now, and that's best served by sticking to a specific version and either only upgrading to bugfix releases, or sticking to one specific commit forever. Those use cases also are better served by sticking to old versions of any libraries they use that are written in Lua (otherwise they might be hit by regressions or turn out to have been reliant on old buggy behaviour that's been fixed), so probably aren't upgrading to newer versions. That's why the article is making the point that it's not worthwhile for the maintainers of those libraries to keep compatibility with old versions of Lua as the people still using old versions of Lua are probably going to be sticking to old versions of those libraries, too.

  • It's not that uncommon for games to stutter for 50 ms, e.g. when an async task on a worker thread takes too long and holds up the rest of the game, or when something needs loading. For a Unity-based game, garbage collection can block the main thread for that long if the developer wasn't careful, too. It's also only sampling every 10 ms rather than recording every power state change, so a game that takes less than 10 ms to actually do its processing has a decent chance of missing any sampling points for a while.

    The exact quote from the mailing list is:

    The 300 ms decay is sized so that a render thread which is only 50-80% busy from periodic vsync and GPU-fence waits holds the boost across its whole busy period at a couple of CPPC_REQ writes total, while an idle core sheds the boost well before it can matter. The energy exposure of a wide window is small because EPP only influences behavior in C0 and an idle core sits in CC6 regardless. If folks want me to make these tunable I am happy to expose them.

  • I've not seen it so much with Java, but for Lua, it's not an English word, and most things in software are either English words or acronyms/initialisms, so it's usually a safe heuristic to assume anything that isn't an English word should be capitalised.

  • It's Lua, not LUA. It's Portuguese for moon, not an initialism.

    You can't migrate once you've got a plugin ecosystem because you're not the maintainer of all third-party plugins, and the maintainers might have lost interest or died and used a licence that doesn't let anyone else take over.

  • There doesn't really need to be as they've stuck to the same series for decades now.

  • That's not exactly it. It's the game that sleeps (e.g. because it's waiting for the graphics driver to be ready for more work or there's a framerate limiter). When it sleeps, the CPU core running that thread temporarily goes to idle speed, then when the sleep finishes, it initially clocks back to a medium speed and only goes back to full speed after it's run for a while because someone somewhere assumed that processes that sleep of their own volition don't care if they don't run quickly. That's not true for games, so they take longer to prepare the next frame than they should.

    The patch makes it so that if a core goes into the C0 state (maximum speed) and has already been in the C0 state for most of the time for a few milliseconds, it temporarily sets a preference for that core to go straight to full speed after a sleep instead of to medium speed, and doesn't remove that preference until it's not been in in C0 for 300ms. That gets rid of the slow start after each frame's sleep.

  • Technically it was the Free Software movement that was a form of Leftism, and Open Source was Capitalism's attempt to subsume it once it became apparent that it was capable of producing better software.

  • You need to know how to carry out a cyberattack before you can devise a way to stop it, so CS students need to be taught how to do cyberattacks to learn about cybersecurity.

  • If you do want to fully shut down, you can hold shift when pressing the Shut Down button, and it'll do it even if this setting is set to the default. This can be good to know if you ever troubleshoot other people's computers or have a computer that starts up noticeably faster without changing this setting (which is actually pretty common if you try timing it, despite people commonly claiming otherwise).

  • Sometimes things look simple but have complex knock-on effects, and it takes an expert to notice that there's anything that might be a problem. At least none times or of ten, there's no problem and it looks like someone's wasting their time, but it occasionally saves a much bigger headache later.

  • We just don't have much use for data about how frostbite affects people who've already been tortured first.

  • Methylated spirits aren't typically much cheaper than IPA in a lot of the world until you're buying industrial quantities, aren't as effective at dissolving lots of things (which may or may not be relevant depending on the particular resin you're using), and typically contain additives that leave stains when dried. For hobbyists, this doesn't seem like a sensible cost saving trick even if it does work for the resin they have.

  • Because memory bugs are an absolute bastard to investigate compared to logic bugs, Rust makes the tradeoff of making it harder to express the logic of a program in return for making memory bugs impossible. That Should™ make it easier to write code with no bugs, but can make it harder to write code with no easily-encountered bugs. The kind of bugs it's really good at preventing are ones that go unnoticed for years or take years to link to their root cause, and those aren't the kinds of bug everyone encounters every time they run a program.

  • Solar actually overtook nuclear as least-killy-per-gigawatt about a year (maybe even two, now) ago, although obviously killing people isn't the only bad thing a system of power generation can do.

  • There's at least one company that does tweak (iirc used) airliner turbofans by taking the fan part off so they just have the turbojet (which is already tuned to mostly generate rotational energy to drive the fan turbine rather than produce thrust itself) and use that to spin a generator. Obviously, it's a bit more complicated than that in reality, but there are quite a lot of old engines no longer certified for flight out of an abundance of caution but that still work fine, and a market for high-power generators that don't need to be the pinnacle of efficiency (originally as backups just for occasional use, and now because of AI companies caring only about speed and not about cost).

  • Whether or not that's true at the moment (obviously, the status quo has changed because in 2020 UPlay changed to Ubisoft Connect, so the alleged incident happened years ago, and it's alleged that Ubisoft were forced to stop selling something, so they wouldn't still be selling it), the article specifically says:

    Uplay featured a $15 USD Rainbow Six Siege Starter Pack, but this version was not available on Steam, making the cheapest option on Valve's platform much more expensive.

    The obvious way of parsing that is that it was the UPlay version of the game, but even if not, it's generally not viable to sell Steam keys for things not available on Steam. The only time you can is when a game's delisted but you've already generated keys for it, and then Valve can just wait for Ubisoft to run out rather than making the alleged threat.

  • Deleted

    Permanently Deleted

    Jump
  • Bumblebees can sting repeatedly, but generally avoid doing so. Dying after one sting is specific to honey bees.

  • If we're going to pull up other people's pithy phrases that aren't intended to be taken entirely literally, then the relevant one here is machine learning is the second best solution to any problem. In the (approximately, depending on how you define it) century people have been thinking about computers, we've already found better solutions to lots of problems. If a transformer-based neural network can get 99% accuracy in sixty seconds on 92 billion transistors of GPU and billions more for its VRAM, that's pretty useless if we can also do it with 100% accuracy in sixty microseconds on a $1 microcontroller, or even faster on a less constrained device.

    The attention is all you need phrase is specifically in the context of sequence transduction models, and specifically referring to the discovery that they don't need a combination of attention, recurrence and convolution, but actually only need attention if it's used in the novel way introduced by the paper. If you don't need to transduce any sequences, then this isn't necessarily relevant, and it's critically not a claim that you can do everything by transducing sequences. It was a surprise that applying it to generating new text instead of just converting it worked as well as it did, and a surprise that it kept getting better with larger models instead of plateauing around the GPT-1 and GPT-2 era, and a surprise that the text generation could be used to do other things, even ones as basic as addition. These things weren't predicted by the Attention Is All You Need paper.

  • That's not what the article says. It's about UPlay keys sold by Ubisoft through UPlay that have nothing to do with Steam, and Valve threatening to remove a game from Steam unless the UPlay keys sold through UPlay became the same price as the Steam keys sold through Steam.

  • Deleted

    Permanently Deleted

    Jump
  • From the Wikipedia page, it seems like a fairly normal cancer vaccine. The thing limiting these in North America and Europe is that you can't legally do drug trials on people until you've tried all the approved medicines for that condition first, as trial drugs might do nothing and might cause side effects, and the placebo control group would literally not be getting any treatment. By the time cancer patients have tried all the existing treatments, they're usually either already dead, or already cancer-free, so the only trial participants left are already nearly dead, and likely to keel over immediately no matter how good the trial drug. That means things that drug companies have known will work for over a decade still aren't available, and once they are, they'll need to be unreasonably expensive just to break even.

    Either this one has finally got over these kinds of hurdle after years of effort, or Cuba's been doing trials that wouldn't be legal elsewhere. Even if they have, it's not necessarily a criticism - if a trade embargo stops you accessing lots of medicines, then you need to get through fewer of them before you've tried everything available and can start trial drugs. Once it's fully approved, it won't need to be expensive, as the research costs can be offset against the cost savings from treating patients, as everything's state-funded.

  • 3DPrinting @lemmy.world

    What to do with a roll of unprintable filament