Dragon1-1

Or, you can sell at a loss, and sell a product or service that will make up for that loss (of course, for this you do have to make the "killer app" which will be both indispensable and unusable without your hardware). Several examples of that from the tech world, as well. Besides, mass production itself tends to bring costs down a lot. That's what I was referring to, "cheap enough to mass produce" means figuring out how to make it work with a Chinese assembly line. I assure you, consumer electronics aren't sold anywhere near cost. Electronics are typically sold at a huge markup, unless someone is trying to pull the aforementioned trick (which is why it works in first place). Economics of scale mean that once you have mass production going, you can slash the markup. In fact, I think that laser diode arrays might one day scale better than screens do. One nice thing about retinal projection is that it doesn't necessarily need to go to the same extremes of miniaturization that screens have to in order to look good (same reason why you don't get screen door effect when you project a full HD screen to cover an entire wall). Added bulk isn't too bad, and miniaturization is what tends to drive up manufacturing costs.

They still do broadcast, though. You can fool some older ESM systems, particularly automated filters and the like (common on RWRs, which need to avoid adding too much pilot workload), but in the end, a dedicated EW platform like found in large SAM sites or on a surface ship will be able to look at raw emissions spectrum and see that white noise is stronger from a given direction, even if only slightly. The moment you start radiating anything, you risk giving yourself away. Of course, that alone doesn't necessarily mean you'll be engaged, but it does mean the enemy might think to point some more powerful radar at the area where the unusually strong white noise is coming from. LPI doesn't mean undetectable, it just tries to trick automatic filters into filtering it out.

With more emphasis on "digital" and less on "warrior" as the era goes on, too. In fact, given how much of a flying iPad the F-35 is, I suspect it's paving the way to eventually sticking the pilot permanently into the simulator and wiring the cockpit to a broadband datalink, with the jet flown remotely. And that would sneakily pave the way for the beancounters to finally fire those pesky fighter pilots and replace them all with relatively cheap, mostly replaceable drones that don't complain about flying at night, don't get busted for DUI after a night in town, and don't threaten to go work for the airlines every time you "misplace" their promotion papers to save a bit of dough on personnel expenses. The next generation of fighters is already planned to be "optionally piloted" (read: drones with a cockpit stuck in to prevent the fighter community from lynching the designers). I'm not saying it's a good idea, I'm saying someone is going to try. One hopes that by the time someone figures out how to jam or worse, hijack the datalink, there's still someone in active service who not only knows how to fly a plane, but can train others to do it.

Can't something be done with what the AH-64 uses for rocket zones? This is a similar concept, you have a pylon with one of each missile.

Good video. I wonder if you picked up on the irony of talking about simulating "mature, adult professionals", in light of the early part, when you mentioned that early debriefs would often degenerate into "No, I shot you down" back and forth. DCS community and actual fighter pilots who get to handle million dollar fighting machines have more in common that one would hope...

You don't have to add algorithms to DCS. After all, the aircraft models don't change much mid mission, and when they do, they do so in a few, predictable ways. Nobody cares how exactly RCS changes during gear transition, you can calculate RCS with gear up and gear down, then LERP between the values. Hence, the algorithm would be used to precalculate RCS tables for each aircraft (and maybe even for ground units).

It is an S duct, just a shallow one. Either way, you need to be quite a bit under the nose to see the turbine blades. Notice how low is the angle of that picture, and look at the model in DCS if you don't believe me. Also, even without RAM, radio waves are not 100% reflected, and also notice that the interior of the intake doesn't have flat surfaces. This would cause most of the waves to also bounce to the sides, and this would result in them exiting the intake at an angle, not to mention such a complex path would obscure the strobing effect of turbine blades (which is distinctive enough that a mid-2000s Viper can tell which plane it's looking at based on this). You can't really treat an intake like this as a flat surface. MiG-21 doesn't have this advantage because the fan disk on all MiGs prior to -23 is in plain view, or obscured only by the inlet cone, which is transparent to radar anyway because that's where its own radar is. That would have the effect of making the Viper more visible to SAM radars than to other fighters, because the former would have a good view of the turbine blades. In fact, if you're sneaking up on a fighter, you'd probably want to be below it, to force it to look for you against ground clutter. Aggressor Vipers are well known for their bushwacking abilities, and even against blue sky, you'd often see them late (definitely later than you'd want, given the A model was a vicious WVR combatant).

No (at least the C model and earlier, they might have added something in newer blocks), but it has an S-duct, which, thanks to its angled interior wall, means it doesn't quite count as a flat surface. To figure out how the radar waves actually come out after bouncing around inside, you'd need to use the modeling software, but my intuition is that a lot of them will be scattered, rather than reflected back. Also note that we're comparing it to other jets of the era, in the vast majority of which the fans are directly in view from the front, with fan blades perpendicular to the airflow to within a few degrees, providing a nice, bright strobe. I'd imagine that Viper is not the only jet that would give interceptors trouble in that area, but merely the most famous. I wonder how the F-104 will look with proper RCS modeling, for instance. It's also diminutive, very pointy, has a small radome, S-ducts (albeit shallower ones than on the Viper) and a limited missile load. I haven't heard too many stories about fighting this one, but I imagine that once properly modeled, it'll appear on the radar quite late, as well. The Viggen might, as well, at least when not loaded for bear.

It looks like we might be hitting physical limits of what 2D silicon chips can do. While a 3D architecture is a possible solution, I'm a bit worried about heat dissipation, particularly with multiple layers of logic.

Elon Musk: hold my beer... Besides, quality doesn't matter if you can compete on price, so yeah, you very much can mass produce something that you don't think is good enough, if you can make it sufficiently cheap. Of course, that tech is far from cheap, but there are several other points that it could compete on, like weight or weight distribution. Also, that viewfinder is not that far below what the early Oculus Rift versions could do, which was usable for VR.

IIRC, this was said about the A model, so it wouldn't be festooned with AIM-120s, rather it would only carry four AIM-9s, and probably no bags as well, or just on the centerline. We're also talking aircraft mounted radars of the 80s, so they weren't quite as good as what we have now. It'd still show up noticeably later on modern sets, especially if relatively clean, but it wouldn't be as sneaky as the A could be. That said, if a modern Block 50 Viper drops both the bags and the air to ground hardware, it could still be pretty sneaky, especially if it has just two AMRAAMs and a pair of Sidewinders.

No, just annoyingly hard to detect, according to pilots who flew against it. In absence of the turbine disc, the radar is probably the biggest contributor here. It's not a Nighthawk, but it's still small enough to show up on radar noticeably later than most other jets.

That's true, which is why the exact mechanics of the RAM don't matter, you can just assume some percentage of inbound radiation is absorbed instead of reflected no matter what angle, and this translates to a straightforward multiplication by a ballpark estimation or RAM efficiency. The vast majority of stealth comes from shape, specifically from avoiding presenting a flat surface to the radar antenna. Radar waves reflect just like any other waves, angle of incidence equals angle of reflection. The tricky bit here is hiding the engines and making an intake that will not reflect waves straight back from its inner surfaces while still working well as an intake. This is also why the Viper is a surprisingly good stealth aircraft, despite not really being designed as such: it's physically dinky and its intake has a pronounced S-duct, so the turbine fan (a big honkin' radar reflector) is not directly in view from most angles. The F-35 is different from the Viper in that it also pays attention to reducing side and rear RCS, in addition to hiding the fan disk even better.

Eyelids will get in the way no matter what (that's their job), and while the eye can change shape, shape of the pupil is directly correlated with its optical properties. Yes, it's different per person, but this only means some way of adjusting the HMD to the specific wearer is necessary. You can try tracking the retina, but it's not the only way to tackle this problem. None of them require a major breakthrough, though. It's more of a matter of making the thing affordable for mass production, this tech is still in its early phase, and that means improvements are likely to come fairly quickly. If patent issues don't choke it off, we'll probably see it in a consumer HMD before the end of the decade.

There's actually public software that can be used to determine RCS of any given shape. The only thing that's classified here is how well the RAM coating actually works, and that would just multiply the RCS from shape by a certain factor. The physics behind radar wave propagation are not classified, nor even particularly complex (though heavy to calculate in real time).

Retinal projection is already a thing, however it seems to me that it's going to be more of an AR solution, unlikely to displace screens completely. It's already a commercial technology, though, with Retissa viewfinder using it. For what it's worth, to make a VR HMD you don't actually need to track the retina (which, after all, is inside the eye), you need to track the pupil. It is an interesting proposition, since unlike with the screen, you could potentially locate the emitter (and thus most of the weight) somewhere else than in front of the eyes, and use fiber optics to deliver the beam. Fiber optics don't like bending, so it'll probably still be on your head, but it could be somewhere more comfortable.

That may be because exact simulation of your vehicle's systems is not necessary when you're training in tactics. Tank tactics are similar enough regardless of whether you're playing WT, ArmA, or DCS:CA. The same concepts apply. There's actually professional software, that is VBS, which is developed by the same company ArmA was and which can be used to set up elaborate training scenarios with combined arms and highly accurate weapon performance. The reason to use WT is probably more prosaic: it's free to play. That means you don't have to deal with procurement. Those who ever tried to procure something via official channels in any big organization will appreciate how much of an advantage it is. If the soldiers are already WT players (and many will be, soldiers love those games), setting up a dedicated session for the platoon is as easy as ordering them all to boot up WT and join the server at a specified time. Procuring a VBS license is... less straightforward, even if you technically have the budget for that (itself not a given). Assuming that a given tanker even knows about VBS and that it's an option, everyone knows about WT, but actual training tools have a much smaller audience. Many of us use DCS to experience aircraft and situations which are simply no longer a thing IRL. The F-14 no longer flies in the US, and even Iran is phasing them out. The F-15 is still around, but in a very different form from what it was during the Gulf War. That means our standards are higher than required for many forms of training. Not for the sort of sim training that's meant to stand in for actually flying the jet, but rather for training in general tactics used (though its worth noting that games using realistic BVR and energy tactics are very scarce). The point there is to make you understand a tactic and its use cases, once you do, adapting it to different hardware and circumstances is much easier. I think this is very much underappreciated. In fact, it's the Israelis who are more comparable to the Soviet/Russian military, with their widespread use of conscription and certain parts of their mindset (in fact, US observers a while ago said more or less exactly that, contrary to their expectations). Content of the training programs are one thing, but culture, education level and the troops' mindset greatly influence any given military's effectiveness and level of professionalism. Many people in the West forget that between countries, people are neither born or raised equal, especially not to people with access to Western education and nutrition levels. This directly affects what a military can or cannot do, and modern battlefield in particular is extremely unforgiving to a poorly educated soldier.

It's more or less same deal as INS vs. EGI in the Viper. The specific Viper we have has the INS, but many jets in that time period had an EGI already. It would be nice to be able to remove Link-16 and JHMCS. GPS can be removed by setting your mission to an appropriate date, AIM-9X can be restricted in the loadout, while JHMCS can be swapped for NVGs by the player, but I don't think this can be forced.

I rather suspect they'll laugh at our clunky and heavy VR headsets, hooked by cable to a furnace with a USB ports. A screen on your face is basically the most space-efficient way to do a flight sim, though I suppose some sort of holographic display, or an implant talking directly to an optic nerve (unlikely to ever be the primary interface option) could give it a run for its money.

Still, they had it, and that's a fact that is rather difficult to dispute, unlike the true unicorn that the Eaglejet with air to ground weapons is. They might have been rare, but so was the 4xHARM loadout on the Viper. Given the status of the actual Mudhen, I wouldn't mind being able to turn the Charlie into an ersatz Mudhen at all. That said, CFTs do have their uses on very long range air to air missions. The size of the maps is increasing, we'd have plenty of places to use these. Never say no to more gas.

To be fair, China has plenty of people who might legitimately want a 4090 for gaming. This is a huge market, after all, and it's pretty easy to imagine sellers inside China buying up the stock and hoping to sell for outrageous prices once the ban drops.

I confess I haven't been up to date on the crypto scam. This was a concern with 40 series, I guess it isn't anymore. Then again, if it's not crypto, then it might be AI instead, though I don't know how many are actually buying GPUs for that, since the big players don't use PC hardware for this. Either way, it's certainly not the infinite moneymaker many thought crypto was. And that is the real problem, lack of competition. A monopoly is bad, and a duopoly isn't great, either. Nvidia can do whatever they want with artificial scarcity, price gouging, and so on, and people won't go anywhere else because they can't. A healthy market keeps the prices low and the margins reasonable. GPU market is far from healthy. Lowering the settings isn't the answer with UE5 games, BTW (dunno about others). It's not that it looks hideous, it's that it doesn't help. You're practically forced to have a high end GPU for those, or it just plain won't run. That's why Jedi Survivor, for instance, didn't do as well as it could have. It's an awesome game, but too heavy for many otherwise up to date rigs even on minimum settings. It seems some of that is UE5 itself, but I have a feeling every game dev is running a 4090 or something on their dev PC. It seems like every damn thing that comes out these days has performance problems. My 3090 means I'm mostly exempt from that, but it's not a good look for the studios.

Only when you disregard the entire rest of the crypto ecosystem. Crypto miners are making things worse by introducing additional demand, but this market could saturate as well, because it's limited to those who can access cheap power. Days of GPU racks stacked along the side of a house are long gone. Also, there's the fact that NVIDIA locked down its lower end GPUs to make them less attractive to crypto miners, in order to push them towards the top of the line offerings. So at best, your arguments would apply only to the top tier GPUs. Others are very much subject to market saturation, though obviously supply is currently so low (thanks to NVIDIA throttling it) that this is not a going concern. Mutuals aren't necessarily based around holding any given company's stock long term and collecting dividends. They can do this (some are even based around it), but they can also short them if the going gets though. There are ways of making money off stock that's taking a hit, and I'm under impression most investment funds prefer to trade stock than hold it anyway. Either way, mutuals are typically diversified, so you get the benefits if all companies are doing well, but if it's just one of them squeezing its customers, those with a large share of that particular company (typically rich investors) will be the only ones that'll notice, unless your fund manager had the foresight to go long on its stock, that is.

In the end, I see the F-35 as a way for ED to pay for core dev. Presumably, lowered standards will only apply to aircraft that they know will make them a ton of dough, but are too classified. Also, the F-35 shouldn't be very difficult to develop compared to a modern module, all its systems are hidden behind a touchscreen UI and all aerodynamics behind the elaborate FBW. Plus, it'll help develop the tech for other aircraft, maybe even an FF Stinkbug at some point down the line. https://store.steampowered.com/app/476530/Children_of_a_Dead_Earth/ It's pretty awesome, actually, far from boring. No fighters, though (there are gun drones), the ships are quite large. Very grounded in physics, too, though it has its limitations (and bugs). Interestingly, it turns out ranges are far from infinite, and in fact, a thousand kilometers is quite a long distance to begin a gun and laser engagement in space. The gameplay is supposedly completely emergent, and you can make your own ships and ship components. It's been out of active dev for quite some time, but what is there is really cool (it even has a campaign).

I'd very much like that, radars had been getting overhauls lately on the emitter/receiver side, we need an overhaul on the target side, as well. The Viper, for instance, can be surprisingly stealthy simply because its frontal cross section is so dinky. Meanwhile, if the F-35 goes into a steep bank and exposes its whole upper surface, it won't be very stealthy no matter what it's painted with. OTOH, the F-15 is pretty easy to lock from any angle. This is where DCS can gain a lot of realism.