I enjoyed the super bowl commercials for the most part. You have to be careful not to have too high of expectations, or you will set yourself up for a big letdown. Here is a quick analysis of one of the commercials. This is a Bud Light commercial where at the end they throw a guy out of an office window because he suggests they save money by not buying beer. Are there any actual office meetings that serve beer? (Sorry if I spoiled the ending for you)
I am going to analyze the motion of the guy going out the window. Just because.
At some point in the future, I am going to make a short tutorial on video analysis. However, for now I will just say that I used KickYoutube to get the video from youtube. I then used Tracker Video to get t, x, y data from each frame. Enough of that.
A couple of notes. First, I had to totally estimate the dimensions of the movie. I guessed that the height of one of the office windows was 6 feet. Second, I am not sure that the motion was completely perpendicular to the camera. It looks close enough that I can get some data though. In the future (Budweiser, I am talking to you), make sure you keep the camera steady and perpendicular to the motion of the object. Include something that can be used to scale the video (like big markings on the side of the building indicating height). Thanks Bud.
Here is the vertical motion as a function of time. I fit a quadratic function to the data.
I fit the quadratic function only to the part of the graph that was possibly in free fall. There is a portion of the time that the guy is in the trees, and then he bounces. Clearly, that data isn't part of the fit. Surprisingly, this data shows an acceleration of around 9 m/s2. Maybe I am just a good estimator. I wouldn't put too much on this result, as I said before about the scaling and the video angle. Also, the movie was interlaced and that makes it a pain.
Some other things you can get from this data: Something happened in the trees. If he was just going along a free fall trajectory, he should have hit the ground about 0.1 seconds earlier. Also, you can notice (well, at least I notice) that it looks like the guy left the window with some vertical velocity. According to the function I fit, the velocity (in the y-direction) at t= 0 seconds was 3.7 m/s. This is pretty fast. If you jumped up with that initial velocity, you would go 0.7 meters high (way higher than I can jump), but there are many people that can jump that high. If you want to throw someone that fast, good luck.
Now, how about the x-motion? Here is x-position as a function of time.
Notice the constant velocity. This is a good sign (in my vast experience of looking at fake videos) of it NOT being fake. Well, I have two things I look for in a video to see if it is real. Real videos generally have -9.8 m/s2 acceleration in the y direction and 0 m/s2 acceleration in the horizontal direction.
Also, note the horizontal velocity: 7.7 m/s (17 mph). I guess that is possible for a person. Maybe he had help. Finally, notice the change in horizontal velocity from before to after he goes in the trees. After the trees it is 2 m/s. Clearly, some horizontal force must have been exerted on him (by the trees). I can calculate the average horizontal acceleration using:
The time for this tree-interaction is about 0.6 seconds. This gives an average horizontal acceleration of:
This is about 1.3 g's - which is well within the human g-force tolerance. So maybe this was a real stunt. Note: I did not do the same analysis of acceleration in the vertical direction because I just couldn't get good data for his speed after the trees.
Extra note: don't try this on your own. Jumping out of windows is probably dangerous. Just Saying.
