Data Secrets Lox

As usual, this is a crosspost from Naval Gazing, and the original has pictures and links.

Over the past decade, a great deal of ink has been spilled about the F-35. It’s often lambasted as overpriced and useless in the face of modern threats. We should cancel the whole thing and either buy improved versions of existing fighters or replace it with the drones which will soon dominate the sky.

As you can probably guess from the title of this post, I don’t think any of this is true. While I will certainly not defend the F-35 program as a paragon of good management, it has produced an aircraft that is already very capable, and will become more so with each passing year. It may not be exactly what I would have wanted if I was given a time machine and command of the Joint Strike Fighter (JSF) program in the early 90s, but the fighters being built today are fairly reasonably priced, and more effective than any other fighter in the world with the possible exception of the F-22 Raptor.

“But Bean,” a hypothetical detractor asks, “how can you say that when the entire program will cost something like $1.5 trillion, and has produced an aircraft which can’t dogfight and is generally less effective than existing aircraft?” I will start by dealing with the issue of cost. Yes, it may cost $1.5 trillion over the next 50 years, but that’s the entire program cost, including R&D, procurement, operations, maintenance and sustainment. This is a number that is only really relevant before the program has begun, when you’re using it to compare the cost of multiple ways of accomplishing the same goal. I’ve written about this stuff before. Because it’s been about 25 years since we had a clean sheet of paper, it’s worth noting that a lot of that $1.5 trillion has already been spent in developing the three types and building 840 or so aircraft. Even if we shut down the line immediately, we would probably want to make use of said aircraft going forward, which would mean both the marginal costs of flying them and the sustainment costs involved in keeping them combat-capable. Said sustainment costs are pretty flat with the number of airframes involved, and would make up a significant portion of the total budget.

But there’s a second problem here, which is that while fighters are fun, they’re expensive enough that military forces don’t buy them for no reason, and it seems safe to assume that something else would have to be procured to fill the gap in capabilities left by the hundreds of F-35s that we just axed.² Exactly how much that will cost is left as an exercise to the reader, although if you’re proposing “we’ll develop a new plane and do it better”, then I will point out that the track record of those kind of programs saving enough to pay back their R&D cost is not good.

But why would we need to bother with a new airplane? Existing planes are much cheaper and perfectly adequate, right? Again, no. Aircraft costing is quite complicated, and tends to vary strongly with production volume. A lot of criticism uses older figures for the competitor aircraft, when production volumes were significantly higher than they are now, and the popular wisdom on this issue is also based on the period when F-35 production was still ramping up, and the aircraft were correspondingly more expensive. Today, you’re going to get more F-35As for your money than any other fighter currently in production with the possible exception of the F-16, thanks to the fact that something like 2,000 are scheduled to be built over the next decade. For instance, the FY22 buy of 48 F-35As cost only $95 million each, with the B and C models coming in slightly more expensive due to lower volume and higher complexity. Costs of legacy fighters would come down if we turned up the volume again, but we’d want upgrades to keep them viable, which would eat up savings. Also, many of the numbers thrown about are not adjusted for inflation.

Ah. But that only covers procurement costs, and the F-35 is significantly more expensive to fly, isn’t it? Well, yes. As of 2019, the cost per flight hour was about $44k, twice that for an F-16, but efforts are underway to bring this down, with the value having fallen by 18% over the next two years. The exact value they’re shooting for is $25k in FY12 dollars, which is still higher than most other fighters, but entirely reasonable given the capability the F-35 brings to the table.

Because ultimately what we care about is not cost per se but how much we are getting for our money. After all, an F-16 costs a lot more per hour than a P-51, but nobody is suggesting that it was a bad decision to switch to jet fighters. The F-35 has two major advantages over previous fighters. First, it’s designed to be stealthy. This doesn’t mean it’s impossible to detect, but it does mean that various sensors won’t be able to pick it up until it is much closer. In particular, it’s stealth is optimized against relatively short-wavelength radars, the kinds used for fire control. Longer-wavelength radars might well be able to pick it up, but for reasons of basic physics, they generally can’t track precisely enough to provide targeting data. Stealth has often been oversold, but this is a useful capability as our enemies introduce more and more capable air defenses.

Second, it’s designed to be able to integrate information from various sensors, both its own and offboard, in a way never before done by a tactical aircraft. Each F-35 has not only a very capable APG-81 AESA radar, which can serve not only as a targeting sensor but is also capable of electronic warfare and even high-bandwidth communications, but also an integrated Electro-Optical Targeting System, a very fancy IR camera that replaces more conventional targeting pods, and the Distributed Aperture System, a series of IR cameras that give a 360° view of the airspace around the aircraft. Imagery from the DAS is fed to the pilot’s helmet-mounted display, allowing him to literally look through the airplane for other aircraft. Other F-35s can provide data using a special directional datalink which can be used without compromising stealth, while other platforms can contribute through the more conventional Link 16. The fighter’s computers then integrate all of this, potentially overcoming enemy stealth in the process, to give the pilot an unmatched understanding of what’s going on around him. Given that in Vietnam, 85% of aircraft shot down never saw their attacker, this gives the F-35 a huge advantage in air combat. More generally, this sort of sensor fusion is clearly the way of the future, but it’s also extremely difficult, and problems with the relevant software have been a major contributor to the delays in the program.

More broadly, the F-35 is as much a software program as it is a hardware program and the software is still under development, which a lot of people don’t seem to realize. A few years ago, it was common to hear criticism of the program on the grounds that it was only capable of carrying a few weapons. I actually work in weapons integration (not on the F-35) and you’re looking at tens of millions of dollars and a couple of years to fit a new weapon onto a jet. You have to make sure that the fighter’s computers can talk to the weapon successfully, that the weapon and structure can take the strain even if the fighter maneuvers heavily, and that there won’t be weird aerodynamic effects, either while the weapon is on the plane or after it has left.³ This has largely cleared up as this process has been completed for more weapons, and as other capabilities have come online.

This also explains a lot of the fury around the F-35′s dogfighting capabilities we saw back around 2015. First, it’s worth pointing out that one major reasons to invest in things like sensor fusion is to avoid dogfighting in the first place. Add in modern short-range AAMs, which are capable of launching at targets that are 90° or more off the nose of the plane, and you start to question the need for dogfighting in general.⁴ But even if we don’t accept this, there’s still the fact that the reports weren’t accurate. The aircraft in the famous 2015 case was one of the first prototypes, and lacked most of the fancy sensor and software systems, as well as the radar-absorbent coating fitted to operational aircraft. But the most important fact was that it was not capable of pulling the Gs that an operational F-35 can, a major handicap in air combat. More recent reports have the F-35 performing very well across the board.

There’s also criticism from the opposite direction, that we should be buying drones instead. I’ve largely said my piece on that earlier this year, but the short version is that in the areas where they are useful, unmanned systems have already taken over. We call them things like “cruise missiles” and “AAMs”. But in a world where bandwidth can’t be guaranteed to be either plentiful or reliable, the case for sending more sophisticated drones in on their own is dubious at best. The more likely scenario is that drone fighters will be developed to accompany manned fighter aircraft, perhaps best illustrated by the Australian Loyal Wingman program. Moving the human controller close to the drone would greatly reduce the bandwidth headaches, while also reducing the number of manned fighters needed near the front lines. But a reduced number isn’t the same as zero, and the F-35 is an ideal platform to serve as a controller for these new drones.

In a lot of ways, this post is a few years too late. Most prominent military procurement programs go through three phases, at least in the public discourse. Early on, they are miraculous, and will solve all problems. They’re cheaper than what we have now, and far more capable. Then, as they come closer to entering service, problems show up, and the media turns on the program. It’s now the worst thing ever, behind schedule, over budget, and useless compared to what it’s supposed to replace if not actively dangerous to our brave troops. Lastly, these problems get resolved, and it becomes a perfectly normal, if imperfect, weapons system, eventually celebrated as a fine example of American engineering, and the benchmark against which the next system will be measured. The same process has taken place with the F-111, F-14, F-15, F-16, F-22 and V-22, and it will no doubt happen again. If I’d written this five years ago, when the F-35 was in the depths of step 2, it would have been prophetic. Instead, we’re now seeing the transition into step 3, and it looks far less impressive. All I can say is that I was busy with other things, but I can point to archived internet discussions where I made many of these same points. In any case, while the F-35 isn’t perfect, it’s clearly the best solution available for the threats the US and its allies face today.

2 This ignores both the political ramifications of shutting down the F-35 line, although a minimum of 1% of the House and 2% of the Senate will be gunning for you, possibly literally because they are from Texas, and the diplomatic ramifications, as a lot of America’s allies have selected it and none have all of their planes yet. ⇑

3 This isn’t a joke. There have been cases of bombs coming back up and hitting a plane because of weird aerodynamic stuff. ⇑

4 An assertion that I am sure will provoke a dogfight of its own in the comments. ⇑

[Mod: added tag - a reader]

Quote from: bean on November 28, 2022, 01:28:25 am

Over the past decade, a great deal of ink has been spilled about the F-35. It’s often lambasted as overpriced and useless in the face of modern threats. We should cancel the whole thing and either buy improved versions of existing fighters or replace it with the drones which will soon dominate the sky.

According to my back of the envelope calculation, for the cost of one F-35 your adversary could buy 10 Kalibrs + 1,000 Shahed 136s. What is more capable?

Ok, cruise missiles and loitering munitions are single-use, but so is a F-35 if its carrier group air defenses are overwhelmed and the carrier is sunk.

If.

https://www.navalgazing.net/Carrier-Doom-Part-1

ETA: But also, the answer presumably depends on the overall scale. As in, a fleet of 100 F35s is probably preferable to 1k Kalibrs + 100k Shaheds, but if all you can afford is the one solitary F35, then you might be better of going with the Kalibrs and Shaheds. After all, your one solitary F35 can only be in one place at a time, and needs scheduled maintenance...

There’s also criticism from the opposite direction, that we should be buying drones instead. I’ve largely said my piece on that earlier this year, but the short version is that in the areas where they are useful, unmanned systems have already taken over. We call them things like “cruise missiles” and “AAMs”. But in a world where bandwidth can’t be guaranteed to be either plentiful or reliable, the case for sending more sophisticated drones in on their own is dubious at best. The more likely scenario is that drone fighters will be developed to accompany manned fighter aircraft, perhaps best illustrated by the Australian Loyal Wingman program. Moving the human controller close to the drone would greatly reduce the bandwidth headaches, while also reducing the number of manned fighters needed near the front lines. But a reduced number isn’t the same as zero, and the F-35 is an ideal platform to serve as a controller for these new drones.

It seems like the massive improvements in AI that we're seeing ought to make the need for bandwidth much lower? Every decision that can be made by the drone itself will reduce the need for communication.

This is probably a dumb question, but if the biggest advantage of the F-35 is in avionics why does there need to be a new plane entirely? I'm assuming that there's some non-approptiations-based justification for why something akin to the F-35's computers, various sensors and the software couldn't be rolled out for a retrofitted F-16X or whatever but I don't know what it is.

This is probably a dumb question, but if the biggest advantage of the F-35 is in avionics why does there need to be a new plane entirely? I'm assuming that there's some non-approptiations-based justification for why something akin to the F-35's computers, various sensors and the software couldn't be rolled out for a retrofitted F-16X or whatever but I don't know what it is.

To better understand some of the economics, I strongly suggest the Defence economics, and the US production advantage by Perun. Another part is that every aircraft model you want to support has very large fixed costs, and that's before you get to building and operating them. So if you have military requirements for a plane that has capabilities A+B+C+D sometimes, but can also make due with planes which only have features A+B, A+C or B+D other times, you have to evaluate whether the design, purchase, maintenance, and operations are best suited by the single plane which can do everything, but costs more, or 2+ models of aircraft of which some have lower operations costs but still have a huge amount of up-front cost.

Something like the F-35 is needed in the case of a conflict with a near-peer. Until this year that meant Russia and China, though apparently flying disused DC-10s over Moscow and dropping factor second bricks is now a viable war strategy. The question, then, is whether in addition to F-35 there is significant value in having additional fighter-like aircraft. Don't forget that cargo aircraft, the B-2, etc., are vastly different platforms and will still be around.

@Nabil ad Dajjal

add give one example to @bean's response, the F-35 was designed to have a big, powerful radar.  So it has a relatively fat nose, the F-16 does not, it's at least 6 inches in diameter smaller at the radome*.  So you can't actually stick an F-35 radar in an F-16, it won't fit.  Designing a radar as good as the F-35 in a smaller package is probably possible, but it will mean more power, power, more cooling, etc, all of which you'd have to find room for  in an F-16 airframe.  Doing that will probably mean tearing out a bunch of other stuff and upgrading that too to shrink it down.  The whole project becomes a lot more expensive pretty quickly. And even if you're willing to gut the whole thing and replace all of it, well, you're still working with an airframe not designed for what you actually need it for.

*note, this is a rough estimate.

I thought the F-35 was, in fact, the attempt to create a platform that is both generic (able to do lots of different missions) and modular (comes with a variety of different loadouts and configurations), precisely because we want to cut down on the number of times we need to roll out new platforms. And that part of why the f-35 became as expensive as it is is because we decided to try to make it do so many things for so many stakeholders.

Depends on how much you (or more realistically politicians) are willing to trust the AI to not shoot down, say, a passing airliner.  I suspect this will end up with the same problems that self-driving cars have.

Worse, I'd imagine. For military purposes you have to assume that you are operating in maximally adversarial scenarios, and against maximally sophisticated adversaries, which is presumably not something the self driving folks stay up at night worrying about.

To better understand some of the economics, I strongly suggest the Defence economics, and the US production advantage by Perun.

Is there a text version of this?

Quote from: Humphrey_Appleby on November 29, 2022, 02:50:24 am

For military purposes you have to assume that you are operating in maximally adversarial scenarios, and against maximally sophisticated adversaries, which is presumably not something the self driving folks stay up at night worrying about.

https://xkcd.com/1958/

My issue with that comic is that I think it elides a distinction between murder and prankster.  Yes, most people aren't murderers, and aren't going to go and do things that will actually kill people.  But if you look closely, there are lots of not great road markings out there, markings which human drivers generally ignore when it makes sense.  A self-driving car is going to have trouble with that, and I could see a few people deciding to make the road markings worse in non-lethal ways because a stranded self-driving car a lot funnier than a human driver slowing down a bit in confusion.

I struggle a lot to understand things like the F35 because I feel like I don't understand the use-case.

If we're fighting China or Russia then I'd expect the war to be either a proxy war or a nuclear armageddon. The F-35 doesn't seem very useful in either one of those.

If we're fighting the Taliban then it's mostly a matter of counterinsurgency. Of course we want air superiority, but it feels like we could have gotten that with just what we had in the 1980s honestly. Past a certain point, new technology helps you only if it helps minimize casualties. Predator drones seem to have worked really well for this, is there a reason why that program didn't get the F-35 program's budget?

I agree that the VTOL stuff sounds very useful, maybe the best part of the F-35 program. Anything that makes carriers more effective is pretty useful. But again, that doesn't scream "let's replace all our other airframes with this" to me.

But who knows. I am just a humble software engineer. I suspect that these things are self-licking ice cream cones to some extent, but most bureaucracies evolve into that past a certain point.

We will not really find out how good it is (or not) unless there is some kind of Vietnam-style proxy conflict where both sides are funnelling their aircraft into the hot zone. If that's the case, we could get another Vietnam-style surprise (oops, dogfighting still matters). Or maybe not. Or maybe we'd find out something else entirely.

Biggest surprises of the last few years of proxy wars are: stupid little quadcopters with cameras and/or small bombs are actually pretty useful for infantry! And: tanks are very vulnerable from the top.

Also: the Russians STILL don't issue socks to their troops.

I struggle a lot to understand things like the F35 because I feel like I don't understand the use-case.

If we're fighting China or Russia then I'd expect the war to be either a proxy war or a nuclear armageddon. The F-35 doesn't seem very useful in either one of those.

I'm not sure this is a good assumption, particularly with respect to China.  If they move against Taiwan and our only options are to do nothing or nuclear armageddon, do you think we'll pick the latter?  Do you think they think we'll pick the latter?  If we have the option to fight them conventionally and probably win (in the sense of stopping them invading Taiwan) they're a lot less likely to try, and a lot less likely to succeed.

If we're fighting the Taliban then it's mostly a matter of counterinsurgency. Of course we want air superiority, but it feels like we could have gotten that with just what we had in the 1980s honestly. Past a certain point, new technology helps you only if it helps minimize casualties. Predator drones seem to have worked really well for this, is there a reason why that program didn't get the F-35 program's budget?

I absolutely agree.  For fighting the Taliban, we need Super Tucanos, not jets.  But there are enemies in the middle.  In a war with Iran, the F-35 would be extremely useful, and I don't think we can rule that kind of war out.

Biggest surprises of the last few years of proxy wars are: stupid little quadcopters with cameras and/or small bombs are actually pretty useful for infantry! And: tanks are very vulnerable from the top.

I've known about the vulnerability of tanks to top attack since I was 7. 

If I understood bean and Schilling's analyses correctly, even the `hot peer war that goes nuclear' is less likely to be armageddon, and more likely to be a `broken backed war' in which the nuclear exchange leaves most of the population alive, but cripples enough of the industrial infrastructure that new production of hardware is off the table for a decade or more, so the remaining war will be fought with whatever weapons you went in with and survived the nuclear exchange. And you want to be able to win the broken backed war...if only as a way of deterring the `other side' from going nuclear in the first place.

Maybe if we didn't capabilities for a war with Iran we wouldn't get into one.

It only takes one side to start a war, and lacking those capabilities might be inconvenient if Iran decides it wants to have a war, even if we don't.

If I understood bean and Schilling's analyses correctly, even the `hot peer war that goes nuclear' is less likely to be armageddon, and more likely to be a `broken backed war' in which the nuclear exchange leaves most of the population alive, but cripples enough of the industrial infrastructure that new production of hardware is off the table for a decade or more, so the remaining war will be fought with whatever weapons you went in with and survived the nuclear exchange. And you want to be able to win the broken backed war...if only as a way of deterring the `other side' from going nuclear in the first place.

I don't endorse this.  If we see a full-scale nuclear exchange, I expect that to be the end of that war.  Nobody is going to have much interest in continuing to fight over great distances when a lot of their industrial base is gone, even if the majority of the population survives.  They'll have bigger problems.  (Provided, of course, that the damage is roughly even.)  "Broken-backed war" was a specific set of expectations dating to the early 50s, and was basically the result of someone realizing that in an era with rare and expensive A-bombs, there would be a substantial amount of infrastructure left.  Thermonuclear weapons and growing arsenals killed it off.  And I'm sure that the F-35 isn't built to fight in a post-nuclear war world because essentially nothing is.

Starting a conventional war with a nuclear power would be incredibly stupid but maintaining the capability to theoretically win one at least makes sense. My understanding is that the country who expects to be on the backfoot in a conventional war is the one who has to make the choice about where their nuclear red lines are and which conflicts are worth fighting. NATO was in that position during most of the Cold War, expecting that they would need to use tactical nukes against the Red Army moving West; Russia is in a comparable position now with respect to NATO's crawl to the East. It's not a great position to be in.

Yes.  The strategic logic behind the F-35 could be summed up as "we'd rather not have to choose between Poland and nuclear war". 

But if you want to patrol?

Drones, AWACS and cheaper fighter jets.

Shoot down enemy aircraft?

SAMs

Quote

Ok, cruise missiles and loitering munitions are single-use, but so is a F-35 if its carrier group air defenses are overwhelmed and the carrier is sunk.

Humphrey gave part of my answer to this.

Yes, I've read it before, but in the light of the Ukraine war, I wonder if it is obsolete.

Consider the sinking of the Moskva. At least on paper, it had 40 SAMs and 6 CIWS + various electronic countermeasurers, assuming that they were operational at at least a large fraction of their nominal capacity (which admittedly may have not been the case). How does it compare to the defenses of a US carrier group? 1/10? 1/20? 1/100?
Officially it took 2 Neptune cruise missiles (comparable to US Harpoons) and maybe some diversion by a Bayraktar to sink it. Scale these assets by a factor of 100 and you're still within the price of 3-4 F-35s. Unofficially, a P-8 Poseidon and other expensive US assets may have been involved in the sinking of the Moskva, but I assume these scale superlinearly (e.g. if you need one P-8 to fire 2 missiles, you need less than 10 P-8s to fire 20 missiles).

Attacks on supposedly heavily defended ground target such as Saky airbase and Chornobaivka also show a similar pattern. Of course, it could be claimed that Russians just suck at air defense, which they probably do to some extent because they generally suck at waging war, but it seems to me that this such attacks also imply that air defense is harder than it looks.
Even during the Gulf War the US and its allies had troubles shooting down the Iraqi Scuds, and even though technology has certainly improved in the last 30 years, it's unlikely that it has improved in a direction that favors defense rather than offense.

  The other part is that there are a lot of times when the carrier isn't at risk.  Yes, the current US force mix isn't exactly optimal for hot war roles.  This is not surprising, because it has other jobs to do.

If the carrier isn't at risk, then it means that the enemy can't really shoot back, it's a "goat herders with AK-74s" scenario, so aren't some Reapers + some F-16s + a couple F-22s for the occasional hard targets sufficient?

Depends on how much you (or more realistically politicians) are willing to trust the AI to not shoot down, say, a passing airliner.  I suspect this will end up with the same problems that self-driving cars have.

You could still have a human operator pull the trigger.

@vV_Vv

Drones, AWACS and cheaper fighter jets.

I would point out that the swiss decision to purchase of F-35s came about heavily on cost grounds.  Fighter jets are expensive, and while F-35 operational costs are pretty high, their purchase cost is less than most existing 4th gen jets.   If you save several million in the purchase, that goes a long way to allaying the operational costs.  Especially when you consider that the F-35 comes with a lot of kit built in that older jets would have to buy pods for. To get something dramatically cheaper than the F-35, you need to go to something like the FA-50, which is much less capable.

If the carrier isn't at risk, then it means that the enemy can't really shoot back, it's a "goat herders with AK-74s" scenario, so aren't some Reapers + some F-16s + a couple F-22s for the occasional hard targets sufficient?

There is a big exclude middle between can't strike the carriers and goat herders with AKs. An even more difficult adversary than bean's example of Iraq is North Korea.

North Korea has a massive and dense air defense network. It is dug in deeply. Sometimes that protection is large amounts of reinforced concrete. Other times it is literal tunnels into granite. They make very difficult targets for even the most modern of precision weapons. Much of it is older than Ukraine's air defense network at the start of this war. Elements of it are in the same class or more modern than anything Ukraine started the war with. That includes recent purchases of Russian S-400s. They also have built heavily protected, buried communications to network their air defenses. Even with an entirely 5th Gen US fighter force they are likely going to be able to continue to contest the air for at least weeks. They are simply too hard to kill more quickly than that. The Iraq or Libya models where we can destroy most of the air defenses in 1-3 days just is not feasible over North Korea. It's like they were on the receiving end of copious amounts of American airpower and spent the best part of 7 decades investing heavily in preparing to deal with it the next time.

Then there is the issue that the carrier groups currently have zero 5th Gen fighters onboard to deal with that level of air defense. The F22 is not carrier capable; the carrier air groups are going to be reliant on the land based F-22s to operate in the highest threat environments.  While the land bases are unsinkable, they are static making them easier to target. North Korea has massive amounts of conventional ballistic missiles to enable overwhelming air defenses to damage those bases. Since they withdrew from the Chemical Weapons Convention they likely also have the ability to "slime" airbases with persistent nerve agents as well. There's a real risk of both fighter losses on the ground and reducing how many sorties those bases can support regardless of fighter numbers. We can reduce the ballistic missile threat by operating from Japan, where their short range missiles cannot reach, but that reduces sorties too.

Another issue is that airframes simply wear out. When Trudeau was elected in Canada they backed out of the previous administration's plan to buy F-35s and instead wanted to relook the issue. Unfortunately, their F/A-18s were nearing the end of any reasonable service life. As a stopgap they agreed to buy newer built Australian F/A-18s freed up by Australian modernization plans. ("Owned by a little old pilot in Sydney and only driven to the grocery store and church!" /s)  It can be easy to get confused by things like the B-52 still being in service. Those bombers have almost no original parts left. We've essentially torn them completely apart and built custom replacements since the assembly plant no longer exists. The data plate (with information like the serial number) is one of the few original parts still in service. The last time I rode on a UH-1 helicopter was just a couple weeks before an entire tail assembly fell off of one killing the entire crew. (Not the tail rotor ...the entire tail assembly just cracked and fell off.) That grounded the whole fleet, pulled a bunch of older production airframes out of service, and was a pretty clear signal that the system was nearing the end of it's life. The debate is not really about keeping the fighters we have. It's mostly about whether we upgrade tech when replacing the current crop of fighters.

The best example of an air campaign with state of the art materiel used by both the attacker and the defender are the US bombing campaigns over North Vietnam; the Vietnamese weren't able to stop the attacks and the damage they did inflict doesn't appear to have been a major limiting factor.

According to Wikipedia, the most advanced SAMs used by the North Vietnamese were introduced in 1961. S-200s and Kubs were both introduced in 1967 and were never used in Vietnam. The Soviets were almost certainly holding back their most advanced systems in order to avoid potentially compromising their military position in Europe if those systems were to fall into the hands of the Americans for study.

In lower intensity but more modern conflicts, Israeli F-35s have been able to attack targets in Syria defended by Russian air defense systems, and non-trivial air defense networks in places like Iraq and Libya (in the 80s) likewise had only marginal mitigating effects on air attackers.

I haven't looked too much into Iraq and Libya in the 1980s, but the one time that a Russian SAM based in Syria fired on an Israeli jet, the Defense Minister of Israel dismissed the incident as a "one-off." Not hit (the missile missed and the Israeli jet reportedly wasn't even in the area at the time), fired. Which almost certainly means that Russia has been allowing Israel to bomb Syria on occasion, which is almost certainly not unrelated to Israel refusing to send advanced military equipment to Ukraine. Russia's policy in the Middle East seems to be to try to have decent-ish relations with and make deals with all of the major players in the region.

The thing that most makes me think that more might be going on here than the Russian air force just being unusually bad is reading up on the performance of Yugoslav air defenses during the 1999 Kosovo War. While the 41 1960s-era short-to-medium range mobile SAMs of course failed to stop the NATO air forces, the Yugoslav armed forces reportedly used them a lot more skillfully than the Iraqis had used their air defenses back in 1991, and they mostly survived the war, tied down a non-trivial amount of NATO resources, and managed to shoot down 2 jets (and damage several more) including a stealth plane.

Wikipedia's description of the "cat and mouse game" sounds an awful lot like the recent reports of the air war over Ukraine, and when you look at the effects and try to mentally scale them up to Ukraine's January 2022 air defenses: ~250 1980s-era long range SAMs (S-300s), 72 1980s-era medium range SAMs (Buks) and 89 1960s-era medium range SAMs (Kubs), then scale down from the paper strength of NATO's 1999 air force to the paper strength of Russia's 2022 air force, I'm not sure that you wouldn't expect results pretty close to what we've seen in Ukraine since February.

It can be easy to get confused by things like the B-52 still being in service. Those bombers have almost no original parts left. We've essentially torn them completely apart and built custom replacements since the assembly plant no longer exists. The data plate (with information like the serial number) is one of the few original parts still in service.

For the B-52, this isn't really true.  Structurally, they're mostly original, just with a terrifying number of doublers.  (I once spoke to the B-52 structures chief, and I've repressed everything but the terror.)  The engines are also still original, although they're working on new ones now.

According to Wikipedia, the most advanced SAMs used by the North Vietnamese were introduced in 1961. S-200s and Kubs were both introduced in 1967 and were never used in Vietnam. The Soviets were almost certainly holding back their most advanced systems in order to avoid potentially compromising their military position in Europe if those systems were to fall into the hands of the Americans for study.

True.  On the other hand, the North Vietnamese got a lot of SAMs.  I've heard that Hanoi had more defenses than Moscow in 1972, and that the Soviets were spooked by the results of Linebacker II.

I haven't looked too much into Iraq and Libya in the 1980s, but the one time that a Russian SAM based in Syria fired on an Israeli jet, the Defense Minister of Israel dismissed the incident as a "one-off." Not hit (the missile missed and the Israeli jet reportedly wasn't even in the area at the time), fired. Which almost certainly means that Russia has been allowing Israel to bomb Syria on occasion, which is almost certainly not unrelated to Israel refusing to send advanced military equipment to Ukraine. Russia's policy in the Middle East seems to be to try to have decent-ish relations with and make deals with all of the major players in the region.

The thing that most makes me think that more might be going on here than the Russian air force just being unusually bad is reading up on the performance of Yugoslav air defenses during the 1999 Kosovo War. While the 41 1960s-era short-to-medium range mobile SAMs of course failed to stop the NATO air forces, the Yugoslav armed forces reportedly used them a lot more skillfully than the Iraqis had used their air defenses back in 1991, and they mostly survived the war, tied down a non-trivial amount of NATO resources, and managed to shoot down 2 jets (and damage several more) including a stealth plane.

Wikipedia's description of the "cat and mouse game" sounds an awful lot like the recent reports of the air war over Ukraine, and when you look at the effects and try to mentally scale them up to Ukraine's January 2022 air defenses: ~250 1980s-era long range SAMs (S-300s), 72 1980s-era medium range SAMs (Buks) and 89 1960s-era medium range SAMs (Kubs), then scale down from the paper strength of NATO's 1999 air force to the paper strength of Russia's 2022 air force, I'm not sure that you wouldn't expect results pretty close to what we've seen in Ukraine since February.

It's fair to credit Ukraine's IADS with being effective against the Russians but that network also includes fighters no more than a generation behind what the Russian air force has (mostly the same airframes, but minus some upgrades which may not even be universal in Russian service).  A bunch of SAMs and multiple squadrons of MiG-29s or F-16A/Bs is fine, SAMs and MiG-21s or F-4s is a dicier prospect. 

You wrote an entire Naval Gazing post all about the F-35 and didn't even bother to tell us whether it floats?

Otherwise very enjoyable.