This guy is really amazing. He did an awesome youtube video series on writing a render engine from scratch. I used his tutorials extensively for learning vulkan.
I think it depends on context. There are a lot of sim games where the environment is very controlled and forums are littered with people who can't tell whether the occasional picture is real or in-game. A forest is hard to render accurately but a plane in flight is pretty trivial
Forests have pretty much been solved. Look at games like Shadow of the Tomb Raider, forest looks amazing. The difficulty now is limited to finer details, like hair. Hair is still pretty much unsolved.
See? It’s almost deterministic. People really can’t accept that we don’t know how to do something.
Plop a nature video next to your forest rendering and it’ll become apparent just how unsolved trees are. And everything else, for that matter.
The precise claim is this: viewers should be able to identify a rendered video no better than random chance. If you conduct this experiment, you’ll see that real videos from actual video cameras wipe the floor.
The motion blur will probably give it away. Accurate video motion blur is computationally expensive but conceptually simple. Just render at about 100 times your target frame rate and average batches of frames together in linear colorspace. You can speed this up by rendering at a lower frame rate (e.g. 10 times your target frame rate), estimating motion, and blurring along the motion vectors before averaging the frames. You can further speed it up by using an adaptive frame rate depending on motion speed and contrast. But a lot of rendered video doesn't even try. Look at a fast-moving bright point of light and you'll easily see the difference.
(But note this is only replicating video, not reality. Truly realistic motion blur requires ultra-high displayed frame rates beyond the capabilities of current hardware.)
A lot of games fake the motion blur so badly. Racing games are often especially guilty.
If you're driving at 200 mph, and there's a car next to you also going 200 mph, it shouldn't be blurry.
Also, the length of the blur should not exceed the distance an object travels on your screen in 1 frame. In other words, if an object moves 30 pixels from one frame to the next, then the blurred image shouldn't be more than 30 pixels wide.
This feel like multiple logical fallacies. It seems funny to me to claim in advance that people are going to argue with something that was intentionally non-specific, vague, controversial, unproven, and at least partly wrong, and then when someone argues with you, claim you were proven right. It's easy to predict that someone will argue with something misleading or wrong, and that doesn't give your CG argument any credibility.
What CG videos are you considering, what specifically have you looked at? Can you show some good faith examples of the best CG forests ever made, compared to some specific nature videos? Are you talking about attempts to match a nature video, and saying it's not possible regardless of what's in the shot?
Are you looking at the best examples of CG forests lately? There are some CG full frame video examples of forests I don't believe people would reliably identify as CG, if they didn't know before hand and you left out the explosions & spaceships.
Yea, this one goes from 0 to 100 real quick. The first half is a valid-ish argument, then the second is just an exponentially growing web of conspiracy theories...
JFC we really need to turn this societal acceptance of "you don't own the hardware" around.
It's always been legal to sell parts off your car. Do not buy anything that attempts to prevent you from owning what you bought through any sort of user agreement.
My understanding is you can't go to a junkyard and bring a Tesla back to life, cause they own the software. Course wonder if someone determined enough could make it work at minimal like a normal car (no self-driving). Reversing the firmware is probably a bit of effort though (I don't think reversing anything is ever truly impossible).
Colorado has vote my mail as the default option and it works fine. If you want to vote, you register, get a ballot in the mail and return it. You can verify that it's counted by following it online and we haven't really had any issues.
If you set up the system correctly it'll work just fine.
I absolutely guarantee this will be anything but less expensive.
You can get an R44 with better range and payload for a couple hundred thousand. This will be an electronic nightmare requiring extensive certification and maintenance efforts. Cessna can't even sell ridiculously old designs for reasonable prices due to certification overhead.
The operating cost per hour of an r44, according to some quick googling, is $190. The cost after ownership is too high, and probably the reason that everyone still has cars instead of helicopters in their garage. Reliability and maintenance will be the most important thing here and will determine their success. And that's notoriously tough in the aviation market.
It might be less expensive due to the electrical powerplant. It will take a while before there is enough of a track record to know for sure. But, for example, electric cars are much lower maintenance than gas cars, as Tesloop as shown.
Disregarding that they have a lot of free "Goodwill" replacements - their post-warranty maintenance costs are about $0.17/km - my ICE Ford is at $0.09/km over the past seven years post-warranty. (And I think I've included more items in my spreadsheet, since I've got the cost of things like windshield wipers and fob batteries, which they omit.)
edit: I thought looking at entire lifespan of the Tesla might be unfair since that includes a much longer period when it's significantly higher mileage than my car, but taking only an equivalent post-warranty period is even worse, it's about $0.24/km in maintenance over that time frame.
Yeah, what I'm not clear on is: what's the cost to fly from SF to Lake Tahoe in a helicopter? If it's a lot more than the $250 that Lilium is promising, what is it that makes Lilium cheaper? Is it just the up-front investment in scale and the route network?
For an owner operator, an R22 and maybe an R44 would be able to do SF to Lake Tahoe for well under $250. For a charter service you could probably get a commercial operator to do it for maybe around ~$500 and there are definitely some inefficiencies you could remove there. Lilium is on crack if they actually think $250 is going to be the all in cost for a private flight though. You don't just spin up a new air frame and 135 operation and make money at those numbers...unless you pull an Uber on steroids and open the VC floodgates...
Unrealistic business plan estimations not factoring in the true costs of operating an aerial vehicle...
It’s a bit like “soon to be plane owners” that don’t quite pay attention when they are told the spark plug for their Cessna is 50USD, they need 8 for 4 cylinders and replace them rather frequently ;-)
Changes when they realize that those companies chartering out planes for less than 200 USD/h probably don’t make loads of cash but just keep track of actual expenses, have a high utilization and capable maintenance staff...
> It’s a bit like “soon to be plane owners” that don’t quite pay attention when they are told the spark plug for their Cessna is 50USD, they need 8 for 4 cylinders and replace them rather frequently ;-)
That's supposed to be the promise of electric aircraft: almost all serviceable parts go out the window, just as with electric cars. Construction and maintenance costs are reduced to a fraction of what they are for mechanical systems. And because these are VTOL, ground expenses are likewise reduced.
Batteries are still rather expensive, though, and energy density sucks. The advantages may not be able to compensate, at least not sufficiently to hit a price point that appeals to a wider, non-millionaire market.
Ok, the only certified batteries you can install in a Cessna is like 500 USD and it literally dies after 2 years of (not heavy) usage. It’s worse and less sophisticated that your 50 USD car battery.
I don’t want to sound like an aviation cynic, but for pilots (even in the non-commercial sector) spending 50 USD for a spark plug or 500 USD for a battery is “normal”.
Tiny bit of innovation and disruption = very very expensive in aviation.
Some israeli startup is doing replaceable aluminum air batteries. Also there are probably low-cycle solid state lithium designs that can start to approach the needed density, it would take a really good recycling loop though.
A medium helicopter charter might be $1200/hour, SF to Tahoe is ~130 nm, so ~1.5 hours each way.
Of that $1200/hr, $150/hr is fuel, $50/hr is engine maintenance, and maybe $100/hr airframe maintenance. These scale linearly with time. For fixed costs, a new helicopter is ~$3 million, so figure $300k per year in depreciation, taxes, insurance, and finance costs. Add another $150k/year for pilot salary and training. If you find lots of customers and keep the helicopter busy, say 1000 hours a year, there's $450 an hour for fixed costs.
The real key to reducing costs is increasing utilization, this keeps the fixed costs reasonable. Batteries and motor will likely improve fuel+engine cost, but also hurt utilization because charging takes more time than refueling. A large network improves utilization. But building a network, with various types of demand (leisure on weekends, business commute during weekdays, cargo during off-peak periods?), is the real challenge.
This isn't going to be certified and allowed for part 135 operations inside at least a decade. Boeing can't keep their jets from crashing due to simple trim control software, what makes anyone think the FAA is going to go along with these flights over densely populated areas?
This feels a lot like when everyone was scrambling to start helicopter taxi services which promptly crashed and burned... Helicopters were a mature and well understood technology then, but the realities of operating in urban areas under a variety of weather conditions just doesn't allow for these services to be A) safe or B) economical.
I wonder what would happen if the car (or something similar had never been invented) and someone came out with a modern saloon today.
"It'll revolutionise personal transport but we estimate it'll kill 40,0000 people in a horrible way per year"
Like many things that are harmful in some way cars got grandfathered in (as did alcohol and tobacco) - if someone came out with an equivalent of alcohol with the same side effects it would be banned immediately as well.
In fact the UK did exactly that with the psychoactive substances laws - we didn't ban a particular drug we banned any drug with a set of side effects - largely because the chemists got really good at tweaking the underlying chemical structure enough to evade the law.
The word you are looking for is pleasure. The only types of drugs that are 100% illegal are drugs that have no purpose besides making you feel pleasurable sensations. Horrible side effects (e.g. chemo) and high chance of addiction (opioids, amphetamines, etc) and more are all allowed as long as the purpose of the effect of the drug is not (solely) pleasure.
Alcohol, and increasingly in some parts of the world Cannabis, are the exceptions (I understand both of these substances have real and potential uses in healthcare, but they are perceived as recreational). These are legal or quasi-legal only because they are both already in wide use and getting society on board to enforce a ban is difficult to impossible (depending on the society, a few do) . Tabacco is also on the list, but seems to be falling off somewhat.
You can throw around statistics but the reality is aircraft receive far more scrutiny. Nobody wants to sit helplessly as one of these things flies them into the side of a skyscraper or watch as one falls out of the sky onto them.
Is that fair? Maybe?
Is that reality? Yes.
This isn't like the invention of cars. We have had all manner of airplanes for over 100 years and know how they work. This is like the NYC helicopter taxi boom in the late 70s and 80s where a number of fiery and high profile crashes put an end to the industry.
One big difference is that these EVTOL aircraft are small enough that they can use ballistic parachutes for an extra degree of safety. They usually also have more redundancy due to distributing thrust between many small electric motors. There has been a lot more research into autonomous drone flight than autonomous helicopter flight as well, which is crucial for safety and keeping costs low. I could be totally wrong, but I feel like battery, motor, flight control, and composite material technologies are finally good enough for EVTOLs to start making sense. Just like the tech boom of the late 90s there will be a lot of investment, most of which will be lost, but the survivors will have a big impact on society.
Maybe. To me, the cost is a bigger factor. If you can show helicopters have the same fatality rate, but has the same price of an Uber, I'm sure a ton of people would use it.
It seems to me price is the larger barrier for most people when it comes to air travel.
Generally speaking, even if cars crash into buildings the building itself is not immediately unsafe; injured people and a broken storefront, but the building is not on fire or collapsing. Unless something has changed dramatically, planes crashing into buildings generally start fires, and generally cause concern about the structural integrity of said building in the immediate aftermath.
> On October 11, 2006, a Cirrus SR20 aircraft crashed into the Belaire Apartments in the Upper East Side of Manhattan, New York City, at about 2:42 p.m. EDT (18:42 UTC). The aircraft struck the north side of the building causing a fire in several apartments,[2][3] which was extinguished within two hours.[4]
> Both people aboard the aircraft were killed in the accident: New York Yankees pitcher Cory Lidle[3] and his certificated flight instructor.[5][6] Twenty-one people were injured, including eleven firefighters. An apartment resident, Ilana Benhuri, who lived in the building with her husband, was hospitalized for a month with severe burns incurred when the post-impact fire engulfed her apartment.[7][8]
> On October 13, 2006, two days after the crash, the FAA banned all fixed-winged aircraft from the East River corridor unless in contact with local air traffic control. The new rule, which took effect immediately, required all small aircraft (with the exception of helicopters and certain seaplanes) to seek the approval of and stay in contact with air traffic control while in the corridor. The FAA cited safety concerns, especially unpredictable winds from between buildings, as the reason for the change.
Most car crashes do not result in 2 dead, 21 injured, and property damage to several residences.
It depends on society's tolerance for death honestly. You're not going to be VTOLing a 737s worth of people so you're probably talking about the number of fatalities in a mid size SUV. People would care a lot less if the routes didn't take them over dense urban areas and the chance of one coming down on their head is super low.
It really depends on what it falls on or smashes into and if the VTOL has experienced lithium battery fire or not. If you use it to move between cities it could probably work just fine.
Ok - maybe those hating me for the comment: FAA validation for my German license took a day and I got a plastic card license after 6 weeks with a preliminary one right away.
Changing ownership of a plane took me 6! weeks with the LBA.
I guess that there are potentially 10+% of all planes in Germany operated with an American N-registration (owned by a trust) because the maintenance overhead and paperwork headache is so much lower.
Imagine registering your German car with a German license plate in the US and setting up a German entity so you could pull that off - how big would the difference in pain have to be?
Lots of things are in fairly close proximity to highways, this is not really a solution. Crashing into the tower block next to a highway is still pretty bad.
Lots of things in dense areas also are not necessarily on an easy route near a highway, so if that's the limitation you run into Concorde's old problem of "where can you actually fly this thing?"
I'm skeptical about the rest of it, but the article lightly covers using existing transport corridors (waterways, rail tracks, and to a lesser priority arterial roadways) as routes for these medium-range flights that explicitly avoid populated areas.
My issue with this is that most transport corridors are
1. not very wide in the first place. Any air vehicle could cross the width of your standard rail line or highway or river in seconds.
2. Surrounded by intensive land uses, because the transportation corridor itself provides valuable access. So vehicles don't have to go very far off the right of way to crash into something valuable where lots of people are.
Really they spawned a huge industry of balance-operated electric scooter...things... The branding was much more accurate than I think even they hoped as they were a segue to more cost effective solutions.
I think it's a pretty good cautionary tale for those looking to follow the Tesla model. Don't target the ultra-high end unless you have enough of a technological and legal moat to fend off the vastly simplified and cost effective competitors.
That's a 10^1000 multiplier. It doesn't even matter what unit you use really...