HN user

YurgenJurgensen

1,404 karma
Posts0
Comments664
View on HN
No posts found.

That’s not the problem. These machines only have so much thrust, and are only aerodynamic in one direction. Sudden high-speed lateral acceleration is very costly from an energy budget perspective. Randomly dodging around bleeds off speed and may actually reduce survivability by increasing the time taken to cross the danger zone. This is made worse by most of the time the drone needs to aim for a weak spot or critical system to do damage, so terminal guidance is already fairly constrained. You don’t see drone pilots trying to jink around because the best move is to punch the accelerator and try to minimise the time your opponent has to react.

There’s a pretty big gulf between “very quick” and “capable of dodging air bursting autocannon rounds without sacrificing forward momentum”.

Incidentally, modern drone footage is actually not nearly as bad as stuff from earlier in the war. Increased lethality of the FPVs means much fewer conscripts deciding to take a 5.45mm retirement plan.

“Attack drones” have been around for basically as long as civilisation itself. It’s just a recent development that they’ve been made out of metal and not meat. So we never got to this point, we’ve been at this point the entire time.

Modern dedicated AA gun systems claim multi-km engagement ranges against small drones, and 60 degree/second slew rates. A drone capable of beating that engagement time would be better described as a guided missile, and will certainly not cost $2500. The flaws in gun systems are in the cost required for good coverage, not their effectiveness against targets that do enter their engagement ranges.

You’re talking about dodging bullets or shrapnel here. Source on that a multi rotor has that level of manoeuvrability while also carrying a lethal payload and enough battery for a viable engagement range?

Culture is also an expression of real-world concerns and values held by those who create it. AI-vangelists will say “all art is copied” over and over again to justify generating their slop, but no matter how many times you ask them, you’ll never get an answer to the question “What is a single great work of art that did not represent a single real-world value of its creator?”

There is a fairly effective counterpoint here that information is worthless if nobody can find it. Generating a handful of useful designs in a sea of pointless ones doesn’t count as ‘coming up with it’. Remember that a description of every idea already exists in the Library of Babel. I do not think that AI generation without curation really counts as ideation.

If you can’t explain it to a human; you can’t explain it to a bot. I’d rather have people just send messages in their native language than correct their broken English with LLMs or send machine translations. At least that way I’d have a record of exactly what they intended to say.

My Japanese isn’t good enough (I feel like I could pass N3 if I wanted to, but I do not find exams fun, and it won’t benefit my career, so I don’t) to comment on how the MtG rules text reads in Japanese, but I can say that English MtG rules text is so grammatically constrained that I’d say it barely qualifies as English at all, so I could easily imagine someone who could read MtG English rules text perfectly but be totally unable to even hold a simple conversation in English.

And if anything, Japanese isn’t even worse for this. Natural Japanese is a highly contextual language, and so I would expect card rules text to stray even further from natural language due to requirements for total unambiguity.