Yesterday @jbrownridge brought my attention to this mystery….

These guys were taking photos of air pellets bursting balloons (as you do) and when they looked at some of the photos they noticed something very strange……

“So we were shooting waterballoons with an air gun and taking pictures of it at 60 FPS using the Casio EX-F1. When I looked at the shots afterwards, I WTF’d when I saw the shadow blowing up before the balloon. Does anyone know if there is a reasonable explanation for this, or should I just settle for witchcraft?”

Can you solve the mystery of the balloons?  Try to do it WITHOUT looking at the comments!


  1. Nice one,
    My guess is it’s the shadow of a different balloon? The one on the gun-ward side of the one we can see. Let’s see a looser frame which shows more of the scene.

  2. agreed ! maybe its the balloon is align together so that the shadow can be seen as one. then shoot the balloon at the back and you will see the mystery

    1. P.S. Alignment won’t wor, as otherwis the visible ballon would be in the shadow of the invisible one… However possibly a back-projection

  3. I too suspect tech failure in the camera’s sensor scanning top to bottom. Even in a sports shutter setting a bullet is much faster.

    The reflection on the balloon shows that the light soirce is in the right place for the shadow.

  4. its obvious really, either the pellets were travelling at light speed, or the pellets hit the shadow before the balloon

  5. Is it alcohol in the balloon? Like if you pour whisky/vodka etc from a jug and you see a shadow of the vapour?

    1. Remind me of this song: “I shot the shadow, but i did not shoot the baloon… “.. or so …

  6. Digital cameras don’t expose a frame the same way as a film camera – they scan the image (fairly rapidly) or use a combination of mechanical shutter and scan. iPhones don’t have a mechanical shutter at all so they are really good at demonstrating this effect: take an iPhone and make an exposure as you whip the camera quickly across the scene you’re exposing. Instead of a blurry shot (which you’d get with a film camera) you get a distorted ‘stretched’ image. As commenter Z says above, this is noticeable in aircraft propellers and other fast moving things. There’s a really great shot here:

    What I think has happened in the ballon image is that the chip is scanning downward (as Patrick says above) and registers the image of the unburst balloon a tiny fraction of a second before it gets to the shadow. Very lucky timing.

    1. Actually most film cameras using a rolling shutter as well (a mechanical one), but it goes from side to side.

  7. Its to do with the method the camera scans the light. At 60 fps the scan has done the ballon but not the shadow, in the meantime the ballon has burst and hence the shadow changes instantly, the scan continues and records the burst shadow.

    1. “But that has to be really fast!!”

      Either that, or have the shadow be very FAR from the object….

  8. I don’t think its the rolling shutter idea as the water wouldn’t move that fast for that to happen. Only know that to happen due to a flash, but the rest of the screen would be dark.
    I would put bets on something else causing the shadow.

  9. I suspect that the focus was on the cameras tend to add (pardon the non technical lingo) non jitteriness to the focused areas…it’s usually on the face so if you move around the face is not hazy(less fps on the focused area) be that in this case the the camera is trying to keep the balloon pic stable but the shadow is left as normal.

  10. or, the bullet skimmed the side away from the camera and the photo is so bright, that the water is over exposed and not visible with the white background.

  11. Shadow of another balloon out of frame

    Light Source-> Small Balloon1 & Camera -> Balloon2
    Light Source is very close to balloon 1 to prevent enlarging and also perpendicular to balloon 1 so the shadow is block by the balloon itself
    Camera is almost in line with the light Source slightly below the balloon.

    So my guess is a very well angled shot with 2 balloons

    1. Look at the bright spot on the balloon – it’s clearly consistent with the shadow being from the balloon pictured, not a second one.

  12. Is this shot set up by someone performing a trick though? or is Richard Wiseman seeing if someone can work out why the illusion of the shadow took place from an innocent situation?

  13. Likely has to do with the way cameras work. The CCD is read from top to bottom and because the bursting of a balloon is a fast movement the frame shows the instance where the camera had read the moment right before the bursting of the balloon but by the time it got to the bottom of the balloon, the balloon had already exploded and so it captured the shadow cast by the burst balloon.

  14. I’d say patrick neville has it right. It’s what I thought before looking at the comments. The camera doesn’t take a frame in one instant but reads it out of the CCD top to bottom.

  15. I would agree that it’s due to the chip scanning downwards. The effect has been around for many many years as the same effect could be observed with focal plane shutters on traditional cameras.

    I first came across it in an article by the golf Journalist Peter Dobereiner. He set up four Cameras set to fire in sync, Two in Portrait mode, but one with the shutter travelling upwards and one downwards, one horizontal and one with an Iris shutter. Then he took a photo at the point of impact of a pro hitting a ball . The Horizontal and Iris shutters, both showed a blurring of the club shaft due to the movement during the exposure. But the camera with the shutter travelling upwards showed a marked backwards bend in the shaft, and the one travelling downwards showed a forward bend. He explained that these illusions were till then (Late 60’s?) used to illustrate either the whippiness of “Modern” Clubs as if the club had sprung forward and imparted more force into the ball, or the famed, “Late Hit” timing of the top pros. In actual fact it was purely down to whether the cameraman turned the camera to the right or left to get into portrait position.
    It’s explained (though not as well as Peter did it IMHO) here:-

  16. You took the picture with the camera shutter open. The balloon before the one in the picture burst. You put another balloon and you got two shadows. The one bursting and the new balloon one.

    1. The main issue with that would be if the shutter was open that long, you’d get motion capture of the new balloon moving around before it settled, the old balloon being removed, peoples’ hands in the way, etc.

  17. Straightforward really!!

    The pellet enters the far side of the balloon, so the front appears intact.
    The water IS coming out but the sun is shining from the front and above The light reflects off the water, making it harder to see against the pale background, but the shadow proves it is there. If it was filled with a coloured liquid there would be a mystery. It would be more surprising if the balloon had burst and the water had no shadow.

    You need a lot more information to be certain of anything.

    1. I almost thought I had to type it out myself, but you perfectly worded my thoughts.

      I’d love to see the math on the sensor scan suggestions though.

  18. The retarded people who still comment with (seemingly serious) incorrect explanations when #2 got it right are retarded.

    1. Oh, I’m sorry, contestant – your score was too low to register on the trollmeter for this forum, so we can’t include you in the all-time list. Keep trying, though – everyone can improve if they try really hard.

      Tip: try to vary the insults within your post – using ‘retarded’ twice, for instance, betrays a woeful lack of imagination, even when it does echo the ‘serious cat is serious’ meme…

  19. I would use a second ballon and a light arrangement sutiable to hide the shadow of the visilbe ballone and to project the shadow of a second non-visible ballon to the white wall..

  20. Maybe you missed and hit it on the rebound from the backboard and then got that shot because the shot had hit the other side of the baloon but not reached the camera side yet.

  21. Richard.

    Can you enlighten all of us with the correct answer tomorrow.

    My head hurts now reading all the comments.


  22. Sometimes, programs in the matrix do not do what they’re supposed to do. Obviously, the shadow program is malfunctioning and must return to The Source.

    The Oracle explained this all quite succinctly in part II, and I don’t understand why you’re all avoiding the obvious. Too much blue pill?

    1. Haven’t seen it in a while, but I’m one of those very rare people who liked all three installments. Even wackier, I thought the best part of the series was the Architect’s speech at the end of the second movie… beats all the special-effect kung-fu.

  23. Either the solution in comment #2 or fine droplets of water that the light from the sun is refracted through, deflecting the light and leaving a shadow behind the fine droplets.

  24. Hmm, too bad this is a bit late but: yes, Rolling Shutter is undoubtedly the explanation. A bit of googling let me know that this particular model of camera has a CMOS sensor. CMOS sensors almost always feature a Rolling Shutter (unlike CCD sensors, which do not) so I would say this is case closed.

  25. *I haven’t looked at the comments*

    I don’t know much about the software/mechanics of the camera in question (or any camera, for that matter!), but my guess is that the exposure ‘scans’ downwards, i.e. the image is ‘processed’ line-by-line from the top down, and the time this takes is enough for a frame taken at just the right moment to capture this image.

    Maybe if you turned the camera upside down, and were lucky enough, you could get the balloon popping before the shadow?

    *Now to see if I’m even close…*

  26. I’m thinking it might just be a rush of air being expelled from the balloon, before it bursts.

    In the shadow image, the balloon hasn’t burst yet – it’s only gas/air escaping.

  27. Eh, don’t know… the water is actually escaping but you can’t tell it from the background wall color?

  28. Hot air rises, so the air in front of the balloon is less dense than the air in front of the shadow. This means the light travels faster from the balloon than from the shadow. Therefore shadow we see is slightly older than the balloon. With the bullet travelling faster than light speed (as Richard commented above) the balloon actually bursts after the photograph has been taken. It’s all very simple really.

    1. PS I flipped the image vertically in Photoshop and the reverse effect is now visible: The balloon has burst, but the shadow has not.

  29. I’m going with the “rolling shutter” theory, especially since the shadows of the splash seem “smeared” downwards, exactly as if some of the water drops were keeping up with the advancing line of the scan. In game development you could see a similar effect called “tearing” where the video render wasn’t locked to complete a frame before changing the scene (doesn’t happen with newer cards unless both the programmer and end-user set things up to allow it).

    I remember seeing something almost identical to this when our particle generator wasn’t frame-locked but our main renderer was, if you took a screenshot at exactly the right time you saw the bottom of an explosion emanating from an object that appeared intact.


  30. My guess would be that the camera’s sensor doesn’t capture the image all at once. It very quickly scans the image from the top of the sensor to the bottom maybe it takes 1000 ths of a second, so when it got to the bottom of the balloon it hadn’t burst yet, but when it got to the shadow it had

    If the shadow was lower it may have captured the balloon whole, but the shadow may have shown the balloon burst more than the picture provided

  31. It’s a artefact of the way the image is read off the sensor which is clearly bottom-to-top. “Rolling Shutter” artefacts are well known in videography.

  32. This is the “redshift” phenomenon familiar to astronomers. Only it works in reverse, because the balloon is green, which is the opposite of red, which is why the *shadow* shows up first, not the balloon…

  33. I think it’s 2 balloons.

    The shadow of the water looks like is already beginning to fall down, not the symmetrical burst you’d expect if it was caught at the exact instant of contact. A split second later the water is beginning to fall and the hidden balloon is already shrinking so it is well behind the first balloon.

    That being said, it looks like there’s a dark line below the first balloon which could almost be the bottom of the hidden one.

  34. I’m almost convinced by the vertical scan explanation. But I’d expect the shadow of the exploding water to be horizontally narrower at the top than the bottom which it isn’t.
    Also the baloon shadow doesn’t look quite right to me, like it’s being cast from a different angle to that implied by the reflected light on the balloon.
    Dunno if you can do a double exposure with that sort of camera, I’m thinking perhaps a double exposure of
    1 a flash photo of the balloon against a dark background
    2 the balloon infront of a back projection of the shadow from behind the white screen.

  35. I don’t think it’s as technical as suggested. It’s a bright sunny day. The photograph was made outdoors as grass is evident at the bottom of the wall. The brightness of the sun implies that it is almost overhead and yes that would produce the reflections seen on the balloon. The balloons shadow would fall on the ground. I now angle the wall backwards so that I can cast a balloon like shadow on the wall with a cut opaque elliptical shape held outside of the frame above the wall. Note the shadow does not match the balloon shape. Close but not perfect. Also there is no hint of color in the shadow that might occur from a green translucent balloon. Take first shot. Then pull up some grass and add it to the out of frame false form so that it looks like water. Take the second shot. Note the water shadow is opaque, not transparent, and occupies a clearly defined area. Looks like grass. Also, the skin of the balloon does not appear to be ruptured.

  36. Pingback: Elliptical Machine
  37. My dream retirement is living on a 45ft center cockpit Hunter sailboat with my man and dog sailing the inter coastal waterways, the Caribbean, and any port I choose.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s