On re-watching, when the stewardesses are sitting on the ground there seems to be some flat panel they are sitting on. Maybe this is helping hold them down. It’s gone once they are in the air, and the removal is hidden by people in the foreground.
Look at their hands. They are holding on to the straps on the floor.
Also keep in mind, this isn’t a real airliner cabin. It’s a set, built inside a cargo plane. It has all sorts of features (grab bars, straps, etc) to help them with the act.
I’m confused about the issue with the balls. They (every single one of them, the colored ones, the disco one, even the paint from the paint balls) absolutely roll to the back of the plane. The ones that don’t are the ones that get caught on something like a seat or the bars they wedged their feet under.
ETA, in fact, I don’t think there’s anything that doesn’t work it’s way to the back of the plane (other than then people).
Well at the very least there is a lot more combining of multiple passes than they admit to, eg using rotoscoping to combine different people and props from multiple takes. But I still don’t get why the balls and disco balls never roll forward en mass. Remember the plane is going from 45 degrees climb to 30 degrees nose down then pulling up again, it goes from weightless to 2G’s and then back again yet the balls stay mostly in the same place they landed and just wobble slightly. Not buying it.
Yes they roll to the back, but they never roll forward again as they should the next time the plane goes from weight less to pulling out of the dive.
Well, at first I was struck by the way it seemed like the balls were falling due to gravity while other elements in the video were, at the same time, floating as though in free-fall. (I thought this was what the OP was saying as well.)
On rewatching, though, it looks like probably in these scenes there is light gravity, not zero gravity, and this is sufficient to make the balls all roll in one direction while people can, by applying force, make themselves look like they’re floating.
They aren’t? It looks a lot like they are.
What’s a time stamp where you think we should be seeing balls rolling forward en masse?
They go through several complete cycles from weightless to 2G to weightless again after the colored balls come out. At some point during the transition the plane is still nose down but is pulling up, at that stage its not zero g and the balls should all roll forward. Remember they are pulling out of a 30 degree nose down dive, they can’t maintain zero gravity perfectly until they are in level flight again. It doesn’t work that way.
Do you see everything else in the video moving forward at these points?
Nothing would ever move forward. When the plane is going up, things would move to the back, when the plane is going down, it’s going down at (up to) 2g, so things would either free fall with the plane or move to the back because the plane is falling faster than then objects inside of it.
In order for things to move to the forward it would have to be falling, nose down, slower than gravity can pull it down. Or, I suppose, be flying, in any direction other than down, slower than gravity can pull it. A stunt you don’t often see planes doing on purpose.
All I have available to me for thinking through this is visualization, but I’m thinking of a single light ball and nothing else on a plane doing that parabolic arc thing over and over.
Plane is going up–ball is in the back of the plane under 2g.
Plane humps over the top of the arc–ball gently begins to float.
Plane is going down–ball is under complete freefall.
Plane begins to accelerate back upwards–ball moves first towards the front of the plane, but (as quickly as the plane curves upward) curves down to the bottom of the plane, until it begins moving/rolling to the back of the plane.
I’m not seeing (in my mind) a time when it should make a particularly large movement forward.
Why? If the plane is pulling up from a parabolic dive, the acceleration should be more or less straight “up” relative to the plane, i.e. almost perpendicular to the floor. Because that’s the direction the wings generate lift towards. There’s no reason there should be a strong forward (or backward) component to the acceleration vector.
If the plane is accelerating (forwards/upwards), why would an object inside of it be forced towards the front of it. I can’t picture that. It would attempt* to fall back towards earth…the back of the plane. I thought that was coremelt’s problem with the video.
*If the plane is accelerating up at less than one G the object would ‘fall’ to the back of the plane, at 1G if would appear to float, at greater than 1G the back of the plane would get to the object before before it could fall that far. I can’t see any situation in which it would attempt to move forward.
The plane is always moving forward and because of the nature of the vomit comet, I can think of a scenario where anything would would fall to the front.
A couple of other data points. The charter of 21 flights on an IL-76 would put this music video way up there with one of the most expensive ever made. It’s not S7’s plane, based on an estimate of $200,000 US per flight = $4.2 million US just for the plane costs alone. That’s one hell of a sponsorship from S7, and the only other music videos ever made for comparable costs are by Michael Jackson and Madonna. OK GO is not in the same league, they’re popular on the internet but they’re not getting that kind of commercial success.
Second I happen to know (form having worked on similar jobs in the past) that agencies and PR people claiming things are real to get “viral” credability is common practise. The famous Honda Cog commercial is absolutely not one take, I know people who worked on it at the Mill in London hiding the bits when they joined together takes.
You could do this using CGI, wires and green screen for a hell of a lot less than $4.2 million, so simple economics dictates its probably not real.
But they didn’t. Unless you can come up with something better than ‘I think it’s fake’, the answer is ‘but they didn’t’.
And we’re not going to get anywhere if you’re just plain not going to believe anything that we and OK GO tell you.
Where is that $200,000 estimate from?
So, coremelt. What would it take to convince you it’s real? You’ve cited at least seven “impossibilities” in the shoot so far, and folks have produced evidence that you’ve been wrong for each of them…at which point out comes a new objection with even less credibility.
You’re well past special pleading at this point, and accelerating toward No True Scotsman. At what point are you forced to at least consider that just maybe the thing that quacks like a zero gee duck might actually be a zero gee duck?
As for the expense complaint, the bloody thing starts with a “thank you” to the folks who provided help, and folks donate time and resources to these sorts of projects all the time. It doesn’t actually cost $4.2 million dollars for the plane to be in the air for eight takes, heck, there’s a link upthread for a company that would do it for me $40K (eight 15-arc runs at $5K each), and I’m not famous (that you know of).
I’m not going to believe anything OK GO tells me, they have a very good incentive to lie. My saying its fake is based on extensive experience in the visual effects industry, not just me saying “its fake”. And I’m hoping we can find a way of checking flight path records which would answer it one way or another.
Another giveway. If they shot it for real there would be an awful lot of failed takes where things didn’t turn out right and they would have released some blooper reels. Why wouldn’t they? that would be easy extra youtube hits. They haven’t done that, another thing which makes me believe its all visual effects trickery.
And no it’s not eight takes, its 21 flights over 3 weeks with hundreds of failed takes. What it will take to convince me its real? Either releasing blooper reels showing the failed takes, or flight path records.
Are you really saying you can tell from some of the pixels and from seeing quite a few shops in your time?
Seeing as I have spent more than 20 years working in visual effects. Yes.
Also the economics simply don’t make sense. You get $5000-$8000 per million views from youtube. cite:
OK GO’s most popular video to date only has 50,000,000 views netting them approx $300,000 US from that one video. Nice work but its nowhere in the league where a company would sponsor them to the tune of $4 million US plus for a video. S7 is a domestic Russian airline, why are they going to pay $4 million US to sponsor a video which is mainly going to watched by people that aren’t even in Russia?
Its far more likely that S7 paid the much lower post production cost for using Visual effects, and they needed an airline sponsor to make it plausible that it was real.