Hacker News Reader

Comment by jandrewrogers | Hacker News Reader

It is not just production cost, the average latency between target detection and target destruction has a large impact on battlefield dynamics. More precise weapons can destroy most of the capability of less precise weapons before they are ever used. Additionally, precision weapons typically have a much smaller logistical footprint, and logistics can make or break military campaigns.

Much of the US focus on precision terminal guidance is derived from this calculus in a straightforward way. It may be more expensive in a unit cost sense but significantly cheaper in terms of net expected effect on the battlefield. This "precision versus quantity" argument played out to greatly favor precision in Ukraine.

View on HN

> This "precision versus quantity" argument played out to greatly favor precision in Ukraine.

Yes, but the defining attribute of the Ukraine war seems to be CHEAP precision - the ability of drones to respond and attack car-sized targets in real time is what has turned this war into a slog.

The American-made stuff is great, but I've seen multiple examples of Ukrainian missile crews reacting contemptuously to the idea that they could use a Javelin/NLAW to take out an older Soviet piece of equipment; that kind of task seems to be reserved for Soviet-era weapons or (preferably) drones.

OTOH the highly precise HIMARS played a crucial role many times. Regular artillery has to shoot dozens of shells before it hits the exact high-value target. This betrays the position of the cannon; if it's close enough it will be fired at. HIMARS is precise so it can pack up and leave before the projectiles even reach the target.

FPV drones are precise because they're remotely piloted by experienced pilots. This allows them to inflict large damage with small payloads applied at a critical point, Luke Skywalker-style.

HIMARS is highly effective exactly because it's cheap precision. An air force capable of executing the missions that HIMARS can would cost Ukraine many many billions of dollars. HIMARS clocks in at $5M per truck and $200k per rocket.

FPV-drones can also be precise post-mortem. You record the flight as command-inputs in sim from start location.

Then you deploy the drone from some carrying vehicle, land and loiter, listening for a trigger. Trigger comes, the drone flies only gets a connection for a lineup if any and flies through the "line up" trajectory.

Pilot involvement can be optional.

Should work as long as the drone's inertial and visual navigation stay adequate. (I suppose that GPS and the like gets constantly jammed near high-value targets.)

In theory, you could do loitering mines that go back and forth, every trigger, resting and solar charging in between

HIMARS is precise, but my understanding maybe 25% are actually getting through to their targets these days. But still until recently Ukraine didn't have anything of comparable capability

Much like 30% of Russian drones are not actually shot down but driven off course by EW these days.

Things progress at a ridiculous pace in a high-stakes environment like a war for existence.

"What is the Kelly Blue Book value of a 1989 Toyota Hilux?"

The reason drones are kicking butt right now is because you get both precision and quantity. Advances in electronics, software, communication links, and sensor technology mean that you can make guidance systems as a hobbyist that would be a million dollar missile from a specialized defense contractor just 15 years ago.

You lose range, but urbanized warfare of the 21st century seems to be a very different battlefield calculus from the strategic bombing campaigns of WW2. The vast majority of engagements these days seem to be within easy drone range, probably because they can be produced in quantities that negate the "just destroy everything within 200 miles" strategy of WW2 carrier battle groups.

Also small drones, which is to say smart grenades, are a counter to trench warfare.

The fact that recent engagements have been within easy drone range is an accident of geography. The same situation won't necessarily obtain in the Western Pacific. The quantity of drones you can produce won't matter if the launching platform can't survive long enough to get within range of the target.

Yeah, although given the number of TEU shipped from China to the US, I would not count out a significant number of drones having been prepositioned in US territory.

This gets complicated, because technology usually advances on multiple fronts at the same time. As others have mentioned, we've seen drones primarily used as an air-to-ground weapon in Ukraine because the airspace is not particularly contested. We have not yet seen them used in air-to-air combat.

There are multiple reasons to believe that drones' advantages over piloted aircraft are even greater than drones' advantages over tanks. Take the pilot away and the G-forces you can pull increase many-fold. Take the pilot away and you have no compunctions against sacrificing a drone for tactical advantage. Take the pilot away and you can field 10x or 100x as many aircraft, since pilot training is often the limiting factor in the growth of your airforce (see eg. Japanese WW2 experience from 1943 onwards, or the need for Top Gun in Vietnam). More aircraft can play airspace denial, since the presence of a bogey creates a kill zone in the area where they can bring their weapons to bear. Computer algorithms can play physics and geometry games where no matter where a piloted aircraft turns, there is always something waiting to shoot them down. Computers can run these simulations instantly, overwhelming the pilot's ability to react. The human becomes the weak link in the weapon system.

The equilibrium I see is drone designs with a range made to just out-range cheap weaponry like glide bombs and common anti-ship missiles, maybe 50-80 miles. For anything fancy (like the supersonic cruise missiles that the Russians have with 200-300 mile range), you want directed energy CIWS instead, but you need those anyway to defend against enemy drones. Then you pack these drones into shipping containers, and launch and retrieve them directly. A single container ship carries its own air force of roughly 10,000 drones, and makes the airspace around it out to ~100 miles completely inhospitable for foes. The convoy becomes its own aircraft carrier, just like the escort carriers of WW2, but the air wing follows the shipping containers and can be packed onto trucks or rail at its destination. Then you bring the convoy to where it needs to be, creating a no-go bubble around it at all times.

> This "precision versus quantity" argument played out to greatly favor precision in Ukraine

With artillery that's certainly true. On the other hand, Ukraine moved from expensive Bayraktar drones in 2021 to primarily drones in the 1k-10k price range today. Cheaper weapon systems allow them to be deployed in more places. Getting 5000 drones for the price of 5 drones might be worse for logistics, but it also means some will always be where you need them, doing wonders for the latency between target detection and target destruction.

Those FPV drones (either of the bombing or suicide variant) seem to be precise enough - there's plenty of footage of them hitting moving vehicles in weak spots like engines/hatches, individual soldiers etc.

As for AI features when the comms are jammed, Russian lancet drones, which have some autonomous capability, seem to be running on Nvidia Jetson boards, and those things cost like $200.

Would you ever see footage of them not hitting weak spots?

It is a fair question, but I have seen plenty of footage where the first drone misses a weak spot. Sure, the explosion is intense, but the tank continues unabated. Then a second or third drone hits the weak spot for disablement.

I'm not just including precision, but ability to get through defenses - i.e. total battlefield effectiveness.

Precision in terms of "circular error probability" isn't the biggest issue now. Its the EW environment.

My point was if system A gets through defenses 30% of the time, but is 10x cheaper than system B which gets through 80% of the time. System A is generally the better choice, except for some very specific circumstances, very high value targets with limited strike opportunity.

> Additionally, precision weapons typically have a much smaller logistical footprint, and logistics can make or break military campaigns.

Only if you exclude manufacturing.

> This "precision versus quantity" argument played out to greatly favor precision in Ukraine.

Until the allies of Ukraine fell behind on artillery manufacturing — allowing Russia to gain dominance and grind down the Ukrainian military.

That mass produced artillery out sustained precision munitions is a major lesson in Ukraine’s defeat.

It's one thing to make a speculative assessment, or a prediction. But the use of the implied past tense in reference to a situation that has many variables and which in any case is far from decided is definitely quite pretentious.

I don’t think anybody serious questions the outcome of the Ukraine war — as evidenced by the international realignment caused by NATOs defeat, increases talk about Ukraine relinquishing ground, etc.

What event do you believe could change the outcome at this point?

And since Ukraine was defeated, we can discuss why: their inability to match artillery exchanges for most of the war.

Since you not only "know" the outcome, but it apparently it's ancient history for you already - why are you asking?

You disagree — perhaps I’m wrong.

But that you responded with rhetoric rather than pointing out where I’m wrong suggests you don’t have an answer to what would change the outcome.

Not even his allies believe in the “Victory Plan” — and several started talking about calling Putin after reading it.

If you think I’m wrong, stop engaging in empty rhetoric and explain what you think will allow Ukraine to win… because nobody seems to know, not even Zelensky.

Sorry, but I can't engage you on this. You have to understand that your original statement, in both content and tone, simply precludes any follow-up or debate. And then when this ws pointed out, you doubled down on the same weird, time-traveling, debate-terminating formulation.

So okay, fine: HN is a big place, and there's plenty of other people you can debate on this particular point if you want. I don't think we'll convince each other of anything anyway, which is also perfectly fine.

    > caused by NATOs defeat

To clarify: Is this a hypothetical statement or is there a specific NATO defeat that you have in mind? As I understand, Russia has not directly attacked any NATO states since the state of the Ukraine war. (Leave aside the sabotage of Nord Stream natural gas pipelines.)

Haha, since when is Ukraine even close to defeat?

Defeated enemies aren’t making regular incursions into the ‘victors’ land which the ‘victor’ can’t stop, or launching attacks agains the ‘victors’ capital.

This just reads like Russian delusion/cope.

What are you talking about?

Kursk was stopped after small gains and prior to any major captures — with Ukraine losing their best units. That loss has led to cities along the line of battle being captured, including the fortress city of Vuhledar.

There’s now increased talk of Ukraine giving up territory — which is their defeat.

Edit due to rate limit:

You’re citing areas Russia withdrew from during the Istanbul talks as “lost” while ignoring that Russia posses 18% of Ukraine and is advancing.

Russia isn’t losing “more and more control” on any front, they’re forcing Ukraine back — including driving Ukraine from places like Vuhledar they’ve held for the entire war until now.

To use your Canadian analogy:

It would be like if the US seized the 20% of Canada closest to the continental US and then proceeded to shell Canadian army to dysfunction from there — which would be seen as a sane and effective strategy.

Oh talk.

[https://en.m.wikipedia.org/wiki/Russian-occupied_territories....]

Looks like Russia has lost everything but a tiny portion they gained, at immense cost - including the near total collapse of their economy.

And are going to be locked in trench warfare on land they don’t control, with uncertain supply lines, with no air superiority - going into winter.

And is losing more and more control of the little they have left.

This is Russia’s Afghanistan writ large, and will lead to the total collapse of the Russian gov’t (and society) soon.

It’s already nearly destroyed an entire generation of Russian men - in the middle of an already epic demographic collapse.

Don’t get me wrong, this has wrought terrible damage to Ukraine too. But with Russia’s economy (previously) and population being 10x larger, this whole debacle is a huge embarrassment to Russia. Even bigger than the collapse of the USSR.

It would be like if the US went to invade Canada, and couldn’t even hold Ottawa.

Edit to answer your edit: maybe if the 18% was the land near Alaska. And they’re at almost the same amount of land they had control of when this whole mess started. All the major economically productive areas of Ukraine are still under Ukraine’s control.

    > including the near total collapse of their economy.

I am not here to shill for Russia, but this is certainly not true. The Russian economy has proven much more resilient than anyone expected since the start of coordinated global sanctions by the world's most developed economies. It is currently growing about 4% per year.

You're not, but given that the IMF is saying 2.6 percent (current), 3.2 percent (projected) -- how does one obtain 4 percent for "current" growth (other than from Russian government figures)?

This is my source: https://tradingeconomics.com/russia/gdp-growth-annual

Regardless, even if 2, 3, or 4%, none of those is "near total collapse of their economy".

So Russian government figures, then.

Agreed that comment above yours was out to lunch, of course.

Zelenski (which presidential term ended) complains every day that they can't properly fight with what they have. This looks like losing for me

That’s called fundraising.

No, it is called a losers fundraising

Reply on news.ycombinator.com