The material here was prepared by Keith Renicle of South Africa. I have included the text of his e-mail to me on this page and then attached the photographs of the maneuvers.

An Analysis of F2B Flying at the World Champs

by Keith Renicle

I've been getting a little better at digital video editing lately. In my ongoing saga of trying to analyze what on earth we are all trying to do when we attempt the stunt pattern, I have found a method to manually trace the pattern from video footage. I can run an avi video file in the background in my Blender software, and then step through it frame-by-frame. This allows me to trace the path of the model. I used the footage from Richard's DVD's from the last world champs in Muncie. I had to find footage where the camera is stationary for the duration of the maneuver, otherwise this method does not work. Richard's footage was invaluable here.

Anyway, these pics show some interesting things that I have mentioned before. The tops of the squares are flown as straight lines (the shortest distance between the corners), instead of the rule requirement of parallel to the ground. The big problem is the huge corner radius that we fly, and this does not leave much room to follow the required path. I even tried this with my diesel stunter with a slow engine run. It's simply too quick. Billy Werwage is very accurate on his intersection in his square 8. I've only traced the 2nd half of his maneuver, because the camera shifted slightly. I've actually checked many squares and square 8's, and all of the tops are flown this way.

Now to the clover..........Goran referred to the FAI rulebook clover as the "Foul-Leaf Clover." If you compare these shapes to Pete's plot of how it should look, then you can see just how far the world's best fliers are from the ideal shape. I've included one from Ted Fancher and the other from Gilbert Beringer flying his twin Typhon. Just by the way, all of these photo's are from round 3 in the finals. The main problem is the horizontal joining lines that are all over the place. The vertical intersections and bottoms are rather good however, and that may account for the 7.5 and 8 scores that they received. Ted's clover shows him flying through the first loop in the vertical path to loop 3. This is a little exaggerated, because the camera shifted slightly, but he still flew through loop 1 anyway. In Gilbert's clover, you can see that I've added a black "halo" at the point of entry to show a path "parallel to the ground" for the horizontal links. Once again, the bottoms and vertical intersection are pretty good. I have also experimented with superimposing some 3-D grids, and lining it up with the view of where the camera is situated, takes a bit of time. When I get it to work well, I'll send these photo's as well.

One interesting point is that when we develop an electronic tracking system, it is going to be quite a challenge to write the software to judge the shapes. When I attempt to line up a projected 3-D shape on the video footage, you can easily see some of the problems. For one thing, the pilots refuse to stand in one spot! The other thing is that the scores are going to come down drastically. Having said that, I suppose that the pilots will soon learn to fly more accurately anyway. Who said that fly around in circles was boring??

I've attached a triangle flown by Serge Delabarde (nice guy and good flier) and you can see the size of the corner radii. In "Serge D triangle 2" I've added a green line that gives us some idea of the angle of the line elevation. The camera is offset to the left, so we can a bit more angle than "height." If we dead upwind, we would only see this as a vertical line. I've tried adding a 3-D grid, and everyway that I look at this, I reckon that the angle is over 55 degrees, or maybe even 60. I noticed this continually while I was present at this event. The high scoring stunts were way over the 45 degree mark. Picture 3 shows the scores that we given. Rather generous, I would say!

I've attached two hourglass pics also flown by Serge, and just like the triangle, we see a very round shape. In the second hourglass pic, I've added a wire-frame hemisphere. I've tried to line it up as best as I can, and I reckon that I'm pretty close. Some of the grid lines have been removed so that you can see the rear section a little clearer. The grid spacings are 22.5 degrees. I used 16 longitudinal lines, and 8 horizontal (latitude) lines on the hemisphere. You can see immediately that the bottom section of the hourglass is way over the specified "slightly more than 45º" in the FAI rules. Most likely, it is pretty close to 70º wide.