chrisl

I'm also a little curious as to the mud situation. Ideally, Russia will be struggling to shuffle all those forces to Kherson on trucks in mud. And if mud makes it harder for Russia to bring up supplies and reinforcements to the left bank, so much the better.

If it's real it will be easier to maintain with RU being very short on artillery shells combined with ever increasing UA CB range. If there's no artillery threat, and RU can't put anything in the air that can fly close, UA can put floating bridges across at much lower risk and be able to move a lot more equipment across.

Also the 50th anniversary of the Yom Kippur war and the eve of a historic agreement between Israel and Saudi Arabia. The only similarities with the RU "offensive" are that it's the same week and Iran is providing weapons to both (but I doubt coordinating with either, let alone both).

Create that friction for supplies, and then keep working on gaining land once the mud starts, despite the mud being just as much (if not worse) a problem for the attacker. Russia having to truck all supplies hundreds of km through mud will be a lot of friction. But if I were going to do that, I'd wait til there was actually a lot of mud so they have to sort out both long distance trucking and mud at the same time.

Around the middle of the article it says he’s a British national of Ukrainian origin - that’s probably why he could leave.

It's not unheard of for Israel to attack Iranian weapons plants, and I wouldn't be at all surprised if they did so in response to this attack. So there's some chance that it could reduce availability of Iranian weapons and ammunition for Russia. Even without a direct attack, transportation could get harder. So I don't really see any upside for Russia, and some real potential down side. Particularly since Israel's support to Ukraine has been relatively modest so far.

It's the kind of thing that we'd never hear about being there, at least not for another decade or so, but could be used for city defense where there would be low risk of capture. Make sure there are a lot of small arms in the neighborhood that can take credit for shooting down all the drones and missiles.

Light and radio are the same phenomenon, just different frequencies (or wavelengths, or even worse, wavenumbers if you're a spectroscopist). One of the best optical system designers I know got his start as a radio astronomer and I met him because he was hired to do the fancy things radio people do (with wavelengths of meters) in the optical (wavelengths of half a micrometer). RF is just a bunch photons of a lot lower energy than optical photons. And we can build incredibly sensitive sensors for both, with very different implementations (thanks quantum mechanics!) Most of the discussion so far has been pretty accurate on the practical differences, which are many. A "mirror" in the RF can be an umbrella lined with hardware cloth that looks like trash in the optical. Lasers still have optics, but you can make the beam very narrow with very small sidebands so it's hard to detect, at least over battlefield ranges. There's still beam spread that depends on your wavelength and optical system - most of the lasers I use are attached to optical fibers so that they emit light in a wide cone at low enough intensity that they're eye safe.

The 45 m/s drone that can go around corners gives defensive systems a lot of time to see it and eliminate it before it can go around those corners. We're only at the very beginning of the drone wars, and Russia seems particularly weak in coming up with defensive systems to deal with them, so slow commercial units are very effective. A somewhat more advanced opponent will be able to deploy a variety of anti drone systems. And you really are going to want weapons with a variety of speeds - as MikeyD notes, there are targets you can't chase down with a 45 m/s drone. And there will be targets that get into cover that that slow drone can't penetrate, or whole targets that the slow drone can't carry enough HE for. There will be times when you want the drones to be the eyes for the high velocity gun. You're missing one or two zeros off the switchblade 300 cost, depending on details of how it's configured. But being able to deliver a 40 mm grenate 10 km away without risking yourself is a kind of valuable capability. What's not clear is if it's more effective to make a drone that only leaves the grenade behind, and brings the RC airplane home - in a lot of cases it very likely is, even if you're only getting 3-4 missions out of it before it's destroyed.

That’s still just 45 m/s compared to 1500 m/s of a direct fire cannon.

All of those will still have a role in delivering the smaller stuff into operating range, much like aircraft carriers, but big, slow, and expensive isn't looking like a winning combination in the danger zones. And the danger zones are getting wider and wider.

Speed. Drones like that are effective against static or slow targets, but if you have a target of opportunity and can hit it with something that has a sighting-to-hitting time of a few seconds, you can get it before it disappears/moves out of range/hides somewhere else. Drones are tens of meters/second, rockets & missiles a few hundred m/s (5 to 10x faster than drones), and direct fire guns ~1500 m/s (5 to 8x faster than ATGMs).

Even that lost productivity has been quantified by the actuaries. I don't keep close track, but it's in the few $M range - that's what an insurer will pay out to compensate a family for the loss of a person. Last I knew (quite a while ago) it was about $3M. Call it $5M today, and that's the lifetime earning value + loss of companionship. Figure that the economic value to the economy is maybe 3x that because their employer is on the exploitive side, and it's about the cost of one MBT. Coming from the POV of someone who develops technology that is MilTech adjacent, it's not necessarily the direct fire high velocity gun on a moving platform that's dead (although I think its role is going to change), it's the huge pile of junk that you have to pile onto that platform to protect it that's dead. We're at a point that modern infantry-carried AT weapons have a range to the horizon (or at least to the next treeline for NLAW) and the warheads can penetrate *any* amount of armor that you can reasonably slap on, including ERA and maybe APS. If your APS is firing, you probably need to hope you have a fast enough reverse gear, because the cheap AT weapons are going to come in fast enough to overload it. So that pile of armor/era/electronics/CIWS that you're hauling around is mostly just sucking up resources (fuel, maintenance) and not helping you offensively because things are at a point where "If you're seen, one shot will kill you". If you want a high velocity gun on a mobile platform, the system to protect it probably needs to revolve more around keeping it invisible and in motion more than protecting it from things that go boom. Maybe light armor for protection from small arms/shrapnel, but anything more is just reducing mobility. Which is the AMX-10, or the CAESAR for indirect fire. Remove the need to put people in it and you have Steve's UGV mini-tank. Way smaller logistics tail, way less energy consumption, and you can make it electric or PHEV so that its idle power consumption (and waste heat signature) is close to nothing. Back to the high velocity direct fire guns. We're already seeing how direct fire isn't that great against guys in holes - the flat trajectory limits you to hitting the edge of the hole, which is a tough target and doesn't distribute the boom very well. So indirect fire or drones are better there. Trench clearing really seems like it should be done by LandShark Mk I or CandyGram drones that can fly in and go around the corners or blow the doors to the dugouts. Those can be cheap and autonomous and even launched from close outside the trench. Like hand delivered DPICM with some minimal brains. And why send a guy when you can send a grenade with wings? But if you need to hit a vehicle near the front lines, direct HV is effective and harder to defend against than a slow rocket. But in an environment where the average infantryman can hit a vehicle out to his LOS (even at the horizon) you don't want to expose that gun. "Direct fire" will start to include flat trajectory indirect fire slightly over the horizon, like a tank version of the Apache longbow. And to avoid radiating a "send your precision guided arty here" signal with the radar system, it will be directed by targeting systems (drones or radars) that are physically separated and may have at most a minimal two-way connection to keep the radiated signals down.

I don't think it's really a pay issue - contractors can pay whatever they want and throw around equity and if the gov't needs the service and that's the price, it's not that hard to write the contract. Especially for creative development where you're going to hire a roomful (or even a small building full) of mechatronics people to build transformers. They each might cost a lot, but even a decent sized small company isn't going to be that expensive. Way back at the start of my career I watched someone who was only a couple years ahead of me leave a semi-gov't job (FFRDC) to take a job at a small company developing UAVs at ~3x the salary, plus they'd pay to get a helicopter license. The harder part about attracting engineers to that is the working environment - going into the classified world tends to be a one way trip, or at least a deep rabbit hole that people take a long time to pop out of, and then they can't even show off their cool stuff. And the whole working on the classified side of things makes it more complicated to find and pull in outside stuff that's supposed to be air gapped, but that might be useful to incorporate. The big cost comes in paying for the production run and all the costs wrapped around that.

And since most contracts are forward funded with spending authorizations that reach out at least several months, it's usually possible to keep all the gov't contractors working unless the shutdown stretches long enough for them to hit the limit of their authorized spending. If you're on a contract that has schedule problems, it's like a time machine because gov't time stops but you have a chance to catch up.

I started to watch but it got implausible fast and I stopped maybe 1/3 of the way in. That's not a trivial depth to work at for SCUBA divers who have to use mixed gases, and the idea that 6 people are going to do it from a 15 m sailboat stretches plausibility. That the boat will then be left parked on a stand (apparently in isolation?) where it's pristine except for "traces" of explosives (but no other signs that it's been used for a significant SCUBA operation?) is an even bigger stretch. I might see if I can find the Spiegel article outside a paywall later. It's faster to read than to sit through video.

That big drone in your second paragraph is the first target of the little drones (or really their owners, who might have some fancier high altitude thing to take it out.) Sensing has to be distributed, and you aren't necessarily limited to one sensor/drone. It will end up being multiple systems that mesh in various ways. In the Diamond Age concept you don't have to hope your drones can loiter longer - you just have to make them cheap enough that you can afford to run them in randomized shifts so that they have time to recharge and you still have high enough density of them in the air for defense. And they don't all have to be in the air - many can be parked, sensing passively, just as radio silent as the attack drones, with other devices producing the signal that they sense off of. And LiFePO4 are good for *way* more than 200 charge cycles to 100%, but they have lower energy density than Li:ion and Li-Po. You'd probably go for max energy density, since you may not even expect them to last 200 cycles in action anyway. Ukraine gets what, a few days per drone? Whoever has a good near term solution will probably find it classified very quickly, if it wasn't developed in a classified space to start with. We definitely haven't seen anything that's good for defense yet, and given the speed that the western MIC works at, may spend a lot of money for a while without seeing much progress. And whatever the first few solutions are, they won't last long once someone has to face them - a military with the flexibility that Ukraine has shown will come up with solutions on the fly.

there are more places called "Springfield" in the US than there are US states.

The thing that's missing is the anti-drone drones. Along the military aviation timeline, we're in early WWI with pilots tossing small bombs out of the cockpit. The main differences are that the plane is really cheap, has RF comm so that it can transmit realtime ISR, and the pilot doesn't have to be exposed to return fire (because RF comm). And we have suicide drones that are closer to small cruise missiles with FP joystick control. We've seen a tiny bit of sticking anti-drone capability on drones, about the equivalent of WWI pilots carrying sidearms to take potshots at each other. Once people work out anti-drone drones more effectively, the cost of air defense should come down a lot compared to using air defense designed to deal with jets against 5 kg plastic drones.

Normally I don't mind that you don't have a like button for your own posts, but this one needs a like. I've never understood people's obsession with lobster. If you want to eat melted butter, eat melted butter. No need to roast up a bland crustacean to do it.

And Russia already cut off much of the water supply to Crimea itself when they blew the dam.

It's because it's not possible to keep secret - everything in orbit larger than a softball is tracked by radar, and at most you can keep it secret for a few orbits until it passes over someone's radar. Sometimes people (or well funded agencies) can take photographs with telescopes on the ground and figure out what a satellite does. Radio frequencies are similar - there are standard bands for transmission that are determined by the atmosphere, and if you're transmitting enough that your friends can receive the signal, then your enemies can at least get the frequency, too. NATO has an assortment of aircraft bristling with antennas to detect and process all the RF in an area. What you can keep secret is what a satellite is doing. Sometimes you can tell from the pictures, sometimes you can't. You can also keep secret what you're transmitting on a radio frequency - that's why there are encrypted radios. But someone with the right equipment, which isn't all that expensive or hard to get, can tell exactly where you are and what frequency you're using.

Looks like the image sensor is vis/very near IR (like security camera kind of IR) with the gain cranked up high. The couple of bright objects in the sky early on are probably much smaller than they look because of that. None of the light sources are actually very bright - the bright spots are maxing out at the whatever the limiting value is for the pixels, but they aren't blooming from having way too much light. When the drone is close to the ship you can make out some more distinct smaller lights on board that are comparably bright to the two small objects in the sky. Those aren't very bright.

The base stations don't have anything special about them - receiver dishes aren't a big deal. The satellites and launch vehicles are the ITAR restricted part. The restrictions are generally well defined, but they can also be updated pretty easily if the government decides something is a risk, and the government can rapidly change what you can export to where. The switch of everything that goes to space falling under ITAR was very abrupt a few decades ago and had a huge impact on US satellite sales. I had collaborations that were affected when that happened, and I've seen a few cases where people's lives were messed up by export investigations even when they were ultimately cleared.

SpaceX very much at the mercy of the USG - basically everything they do includes things on the US Munitions List (Categories IV and XV), so they have to have a whole bunch of people whose job is to deal with export and launch licenses. Just about anything that can do anything useful in space is in Category XV until proven otherwise. The more erratic Musk acts, the harder their job is going to be. I don't know detailed specs of Starlink, but I suspect a lot of it falls squarely into Category XV, despite being intended as dual-use. And they can't just say "we're going to build and launch them all outside the US" because that requires export.