▲ 197 points
back
338 comments
They may kind of suck but even so they are still transforming the war in Ukraine. A month or so ago FPV drones took out much of Russia's nuclear bomber fleet (https://news.ycombinator.com/item?id=44150789), something like 80% of battlefield casualties are due to drones, neither side pretty much can use tanks because they get taken out by drones. It's a huge change in war fighting.
One stat: "In May alone, Ukrainian drones destroyed over 89,000 Russian targets" https://www.newsweek.com/robert-brovdi-ukraine-russia-war-dr...
They've recently promoted the 'Birds of Madyar' guy to run the newly formed Unmanned Systems Forces and are moving to a unified drone line defence the whole way along the frontline. Update on that: https://www.express.co.uk/news/world/2073811/russia-army-ukr...
Also re drones not having the effect of artillery, fair enough but Ukraine has been using FPV drones to destroy Russia's artillery. Here's footage of one of it's most modern being taken out https://youtu.be/DMOjOJnAd8A?t=161 It's kind of asymmetric - the artillery can't similarly take out the drones because they are too small and replaceable.
The article addresses this in the first few paragraphs. The author argues that whilst the 80% may or may not be accurate, they're using a definition of drone that encompasses many kinds of machine and not just FPV drones which is what he's talking about. He also says the number is highly misleading because most of his FPV drone missions were double-taps where the target had already been taken out by more traditional military assets, and that commanders used drones mostly because they were given them rather than because that was the best military strategy.
FPV drones for combat are a hot flash in the pan. They have had a major effect for now, but naturally as these countermeasures evolve, so weakens their effect.
I keep telling people that the terrain and the strategies that Russians use is the primary reason for the effectiveness. Mortars and artillery already handle the same requirements as the author says. The reason they are effective in 2024-25 is that the drip-drip-drip of single soldiers running over vast fields / unarmoed vehicles driving over known routes is the only way Russians make progress. For a moving target they are great, but multiple moving targets would get shredded by competent artillery anyway.
Most nations don’t have flat open fields where signals can reach far away drones unimpeded by line of sight for tx/rx.
By far the best use of drones still is as battlefield recon/fire correction to adjust existing artillery/mortar capabilities.
Source: I’m one such drone hobbyist and I’ve watched way too much footage from the front. None of what i’m writing is in absolute terms. I just don’t see the same way as commenters in the public who think they are a checkmate for any combat situation. The incompetence of the Russian forces caught everyone by surprise, but they have learned. My country’s border with Russia is heavily forested and not as flat as Russia. The drones are not able to go through the canopy. Infrared recon is a way better choice than FPV suicide drones.
The big thing that FPV drones have going for them is that they're ludicrously cheap and easily constructed from relatively basic parts by moderately skilled people.
It's literally cheaper to strap a grenade to an FPV drone and fly it into a tank hatch than it is to fire a single non-precision artillery round, let alone tens or hundreds of them.
Plus, you can deploy your drones remotely from the top of a trunk deep behind enemy lines and fly them into irreplaceable strategic aviation assets with a shot exchange factor better than 1000x.
> Most nations don’t have flat open fields where signals can reach far away drones unimpeded by line of sight for tx/rx.
The drones now are using fibre optic cables with the reel mounted on the drone. Having the reel on the drone avoids snagging issues and the fibre itself avoids EW jamming and line of sight issues.
> I’ve watched way too much footage from the front.
Did you see the videos of a drone dropping a shitload of thermite on a forest canopy? [0]
> Most nations don’t have flat open fields where signals can reach far away drones unimpeded by line of sight for tx/rx.
Most nations have cellular networks that penetrate buildings and forests just fine. In fact, Ukraine used the Russian cellular network for their recent attack deep behind enemy lines.
I'm not saying this will always be possible, but it's not hard to see that line of sight communication is not the end of the line for military drone control. There are many routes for providing an ad hoc line of communication if you don't just use consumer-level tech.
Isn't the drip-drip-drip of single soldiers running around the response to artillery in the first place? Any concentration of manpower attracts artillery and if it's significant, HIMARS gets called in. Naturally, the response is to disperse men and make artillery less effective. The response to that is FPVs chasing down the individuals instead. They're a counter to a counter and can't be judged in isolation.
Exactly. Many are missing the big picture - while Ukraine has managed to hurt the Russsians with drone warfare, how much has all that really helped Ukraine to drive back the Russians or re-take the territory held by the Russians? The simple answer is that it hasn't. Moreover, drones are not going to give an edge to Ukranian any more as the Russian too have mastered not just counter-drone warfare but also streamlined it into their conventional warfare tactics. (For example, Russians now outproduce the Ukrainians in drones and now use WW2 style motorbikes to evade drone defence - https://bulgarianmilitary.com/2025/06/25/russias-motorcycle-... ).
"Most nations don’t have flat open fields where signals can reach far away drones unimpeded by line of sight for tx/rx."
This is true, but flat open fields are precisely the places where major mechanized battles usually took place. For the very reason that manoeuvering other equipment in complicated terrain is hard.
Ofc there are significant exceptions, like the Alpine front in WWI, where Austrians and Italians faced each other in mountainous terrain for years, or the Hürtgen Forest in WWII. But a remarkable share of all major mechanized battles of history took place in flat open fields, or something at least resembling that sort of terrain (gently sloping hills with good visibility etc. etc.)
Great insight in this comment - in the 21st century wars are so unique that hi-tech weapons emerge specifically for certain battlefield applications.
I.e. just one country over, just one slightly different conflict between different actors would require completely different looking-feeling weapon systems.
It’s not something you think about, usually - so much of what we see used is made for asymmetrical warfare.
All writing on the success of technology in war follows the same structure.
A new weapon is introduced and finds success, is boasted as the future of warfare. It works and is a significant advantage for the side using it, being a force multiplier.
After the initial succes the other side starts using it too, and there's a scramble for countermeasures. This makes the wonder weapon less effective.
Then articles are written that are the inverse of the hype following the first implementation. Even doubting if 'this is the end of -wonder weapon- ?'
Look at the tank. With every new weapon (take drones) it is theorized that drones would be the end of effectiveness of tanks as a weapon system.
It's not, but it's not longer a wonder weapon, yet a piece of equipment, that's constantly evolving. Is an arms race and it's been like that since the invention of the club by our ancestors.
Don't forget that drones are evolving very fast and their potential is frightening. What about swarm of unmanned AI and computer vision capable drones spreaded across fields and forests waiting for their prey? You can make antipersonnel drones much smaller as you don't need even to kill the enemy - just to wound.
You can place a big batteries across that zones so drones could go recharge theirselves and continue serving.
Eventually you can just drop thousands of such killers above the territory or even some city and they will kill every human they find.
Ok, then we can make unmanned drone hunters and human killer bots will start to enhance their defense capabilities. That will start another round in evolution where humans on the battlefield are just spectators. Or prey if they unlucky.
> Most nations don’t have flat open fields where signals can reach far away drones unimpeded by line of sight for tx/rx.
Citation needed.
Many of the issues sound like issues coming from using improvised civilian hobbyist tech and doctrine being in its infancy.
If current FPV drones are bit lackluster, it doesn't preclude 'next generation' that are purposefully developed for military use won't be useful. Also it sounds like the designation of "FPV drone" is specific to particular family of drones specific in current day and time, which may be something quite else next year. Like, obviously the next stage is a FPV drone with some capabilities of "reusable" drone or loitering munition author complains of (capability to hover easily)? Or "reusable" drone with FPV camera?
Western militaries have things like this: https://greydynamics.com/switchblade-drone-small-spring-load...
More autonomy, but MUCH more expensive. Thousands or tens of thousands of dollars per use. The issue is indeed using mass-produced consumer drones. It's a bit like the widespread use of "technicals" in some conflicts: yes, a pickup truck with a .50cal in the back is inferior to tanks or armored cars, but it's also much, much cheaper.
There's a bit of a "Sherman vs. Tiger" thing that's been going on since the dawn of industrialised warfare. Is it better to have a more effective weapon that you can only afford a few of, or lots of cheaper ones?
The US doctrine approach to the problem would simply be a set of B2 bunker buster decapitation strikes on Russian military HQs, but of course that option is not available to Ukraine. They can't even manage Iraq-war-style wave of SEAD strikes followed by unit level CAS. The air war has kind of stalemated with neither side having conventional air superiority and both being vulnerable to the other's anti-air.
> Many of the issues sound like issues coming from using improvised civilian hobbyist tech
I don't think it's improvised civilian hobbyist tech. They run autopilots that also fly professional drones and can fly planes.
I think it's mostly that it has to be super cheap, otherwise it doesn't bring value (because other weapons are more efficient if you have more money). If your one-way drone costs 10k dollars, maybe it's too expensive even though it can fly during the night.
And then there are fundamental limitations, like flying in bad weather.
> obviously the next stage is a FPV drone with some capabilities of "reusable" drone
But a reusable drone won't go inside a hangar (because at this point it probably won't come out). If your drone can go somewhere, drop something and come back, doesn't it mean that another class of weapons could do this job?
Given that the drones he talks about are even less equipped than a standard hobbyist drone (no GPS, no compass), I suppose it's more a "tradeoff" between keeping the drone cheap enough to be disposable vs having enough functionality to be practically controllable by the operator.
I don’t understand why the author has such a narrow definition of FPV drones.
He talks as if reusable drones are a completely different category, that they are all toys designed for enthusiast racers… Generally he implies that a myriad arbitrary technical details are fundamental limitations of this paradigm, it’s a strange mindset.
Also, as others commenters state, isn’t a 43% success rate exceedingly high? Even if it’s 20% accounting for environmental factors and faults in manufacturing. How likely is it that a mortar does anything? Or a soldier with a rifle? Or anything else?
> When I joined the team, I was excited to work with a cutting-edge tool.
It sounds like he was imagining some kind of scifi adventure, but it’s always been clear that they are using cheap drones with tech that has been commonplace for a decade. And that’s completely fine, it’s intentional.
I assume this is like a pilot in WW1 reporting how finicky and hard to use bi-planes were. No doubt a bunch of weapons manufacturers have seen this and the special operations Ukraine did in Russia and Israel did in Iran and the wheels of progress are turning and the result will be terrifying.
This tracks with the earlier ~12% drone kill efficiency estimates. However drone is a mass deployment weapon. Ukraine did about 2 million frontline sorties in 2024 and aims for 5 million this year. This 1 out of 9 ratio translates into absolutely devastating damage, that artillery and airstrikes (which are also hardly "easy to use") can only dream of.
This is kind of a 'Muskets are cool but they take too long to reload' vibe.
Yeah Ukraine isn't working with the best tech; it's a doctrine of desperation rather than preparation. But they discovered something effective and it will change the way wars are fought in the future.
It is interesting hearing feedback from the frontline. Even with the issues, I think it is clear drones are changing modern warfare when you have companies like Anduril. What most people think is coming next is autonomous drones, although I don't morally agree with it. Sorry you had to have this experience, I wish this war would end, too many lives have been lost and it is senseless.
> I would, first of all, recommend ensuring that troops in the field have well-trained organic mortar support with an ample supply of ammunition.
That would not be possible because it has become basically impossible to bring in vehicles close to 5-10 kms of the front-lines because of the, well, drones. And you need to carry ammunition to those mortars with something, preferably not how the Vietnamese did it in the jungle (i.e. using brute human force).
Just check this snippet from a recent article in the FT:
> “'At this point, you’re a lucky man if you drive 5km from the front line and your car is still operational,' a Ukrainian drone unit commander deployed in eastern Donetsk region told the Financial Times. He said his men now sometimes had to walk up to 15km at night to reach their positions...
> In the past weeks, Ukrainian supply trucks have reportedly been hit by Russian drones on the road linking Kramatorsk to Dobropillia, some 30km from the fighting. On both sides of the front line, roads are being covered with anti-drone nets in an attempt to stop fibre optic drones."
This comes from Ukrainian guys still fighting this war, not from a Western war-tourist like the guy who wrote this article.
> During my time in [...], I collected statistics on the success of our drone operations. I found that [...]
Assuming the writer and their allegiances are what they say, is any of the info valuable to any of their adversaries?
> They are controlled by an operator wearing virtual-reality goggles
They aren't really using VR headsets, right? The FPV goggles I know are just a screen showing the camera image without any virtual reality.
Interesting read. Curious how the author feels re: the attack on airbases using shipping containers/drones that was so successful?
Seems to be a unique case that worked especially well for (higher end I'm sure) FPV drones. Getting artillery in on shipping containers would have a higher likelihood of detection. Similarly, the ability to 'guide' in the drones with munitions seemed to allow for greater flexibility during the attack and its effectiveness.
I imagine eventually these cheap FPV's will be augmented with low-cost GPU's allowing for running smallish models and self-guided autonomy. This would seem the next evolution where a commander deploys them in bulk and overwhelms the enemy in a way that can't be jammed like radio-communication. Similarly, horrifying when you consider their eventual use in terrorism scenarios...
> If this type of pre-aborted mission is included in the total, the success rate drops to between 20 and 30 percent. On the face of it, this success rate is bad ...
I disagree with this premise. I suspect that 20 to 30% success rate is not at all bad, but rather excellent. Compare to artillery with shells costing a few thousand each on the low end, to $100k+ for more advanced rounds, with 100s or 1000s fired per casualty.
The article talks about signal jammers, but as far as I know most drones there are remote controlled using fiber for exactly that reason?
I don’t agree with the conclusions he draws from his own analysis - almost all the issues and shortcomings he points out are related to technological shortcomings he admits - are already being addressed by the new systems - or are primarily issues with how drones are being used in the field - i.e. tactical combat decisions.
These are not inherently valid arguments regarding the effectiveness of drones as a new weapons platform - but with the current state of the technology and with the decisions on the battlefield
It’s early days, the technology will improve and the tactics will be standardized with time and drones will prove to be a dangerously effective tool - which has the additional scary bonus of being cheap and easy to mass produce and deploy
How many artillery shells one needs to expend to hit anything? I bet more than 10. And they are pricier AND endanger people who use them, more.
Mortar shells (80mm class) are cheaper, but mortars need to be compared to drone-dropped munitions, not artillery shells - because mortars' range is way shorter than that of an FPV drone and is comparable to the artillery - a good FPV drone - although not the $500 one - can cover about the median artillery firing range of this war (16km).
Drones bring about more casualties and are used wider exactly because they are more cost-efficient.
I see a lot of comments saying that "but the technology will improve".
Sure, maybe. Or maybe it will be like Musk announcing what Teslas will be capable of in 6 months. We don't know, and the author doesn't pretend that they do. Don't forget that drones have been used in this war for years, and the vast majority of the drone industry has already pivoted to the military because it's easier to make money there. So it's not exactly "brand new technology".
But my point is that the author just says "from what I've seen, here is how it looks". And it seems like it has value.
Even if the technology improves and the economics of scale reduces the cost, I still don't buy the narrative that swarms of tiny kamikaze drones will radically change warfare.
Aside from radio jamming, I have not seen an actual defense against a strong EMP.
To defend against an EMP wiping out your drone swarm, you would have to invest in shielding etc which would remove them from the class of small cheap drones.
Idk if anyone can speak about this, but to me this doesn't seem like a problem that these types of drones can overcome.
You need to compare this to hit rate with mortars and attrition by counter battery fire on mortar teams. Not to detract from a sober assessment but it's hard to judge without the other parts of the story.
Thr tldr would be "temper expectations"
Don't drones have another advantage not mentioned here -- counter-battery against operators being more challenging?
Most of the problems the article describes are due to using civilian fpv drones on the battlefield or using untrained pilots: bad controls, narrow channel, faulty parts
Most of them would not apply when military finally catches up, starts producing war fpv drones and make good drone pilot training programs
One of the key points of the article is "I feel FPV drones to be mostly a failure because their success rate is low"
Why is that a failure? If one 500$ drone has even only 10% success rate, if the target is a 1M$ equipment it's still a win!
It is only a matter of time until those drones fly into their target fully autonomously with machine learning.
Heck, I could build that with hugginface (I will never do that) in a few evenings if you are ok to blow up the wrong target with a single digit percentage.
> Drones also operate in a cluttered segment of the electromagnetic spectrum. First-person view drones use unencrypted analog radio signals, and in hot parts of the front, as many as a dozen drone teams may be competing for use of a handful of frequencies (a consequence of using cheaper components). This results in the need for sophisticated de-confliction procedures that, quite simply, do not always work. Even when de-confliction works, sometimes a team must wait as long as half an hour for a frequency to become available before takeoff. If it does not work and two drones find themselves in the air on the same channel at the same time, they will interfere with each other’s signals, usually resulting in a crash. On top of that, the enemy’s drones also fly on the same frequencies, which can also result in interference and a crash.
This gave me the somewhat macabre image of Ukrainian and Russian drones doing automated frequency coordination with each other, so they can orderly proceed in bombing each other's soldiers.
I don't think that's what happens though. But I'm surprised flying drones in the same area as enemy operators is even possible. Wouldn't both sides try to jam or take over each other's signals, deliberately blocking channels, etc - so that in the end, no one could control anything?
Or, if the signals are really unencrypted, what keeps anyone from setting up a radio beacon that just spams the "detonate now" signal on all channels at maximum power. Instant drone-free zone?
I guess I'm missing something, but why isn't the problem of finicky steering solved by adding auto-stabilizing software? Would that take away too much of the maneuverability?
The author offers no major new insights on the effectiveness of drones. His counter argument is against the maximalists who claim quadcopters are revolutionising warfare and render armour and artillery obsolete. But nobody serious ever suggested that. The same way tanks are relevant despite rpgs. They simply represent a new element in the battlefield, and a useful one as well. The fact that they are not ideally suited for dropping bombs doesn't matter. They are great for surveillance and giving units situational awareness, and the fact that they can occasionally be used to attack targets that otherwise would be impossible simply augments their usefulness. The article is interesting,
but it's attacking a straw man. I have a great respect for the ukrainian armed forces
but to be perfectly honest their combat effectiveness is not exactly world beating. The suggestion that NATO should be taking lessons from how Ukraine is fighting Russia is odd.
FPV drones don’t suck if you know what you’re doing. If you don’t have proper training, you’re going to suck at it.
Soon they'll be using CV and won't need FPV.
International volunteer? I think mercenary sh*t would be more appropriate.
A guy from Slovakia is fighting for Ukraine? Don't tell that your prime minister and supporter of Russia Robert Fico.
I've seen some of these FPV videos of kills of unarmed Russian soldiers. I honestly don't know why the pilots are not prosecuted for war crimes.
(and I'm sure Russia does the same to Ukraine, I just haven't seen those videos).