View Full Version : GoPro on F1 car at Bathurst - no WP case Awesome


Ian Newland
April 1st, 2011, 05:43 PM
GoPro was mounted on the front wing on the RHS tip and was naked - no waterproof case.
Shot in R5 Mode 1080p, almost no jello. This will show you what the GoPro is capable of.
The original footage is clean and almost artifact free considering the car was doing 300kmh.
These F1 cars run very smooth, hence the lack of jello in 1080p, 720p @60fps would have been better again.

The 720p video can be downloaded from here. It's only 2000kbps but the detail at 300kph is amazing.
Full Onboard Camera Lap of Bathurst with Jenson Button!.mp4 (http://www.mediafire.com/?ifvigespdsqswtc)

jenson button bathurst on Vimeo

Jim Cowan
April 1st, 2011, 05:51 PM
Hi Ian,
Thanks, that was pretty cool.

thanks
jim cowan

Ian Newland
April 1st, 2011, 06:06 PM
This was the mounting Position. You can see it on the front wing in front of the tyre.

http://img716.imageshack.us/img716/5384/65306947.jpg

Screen capture at 300kph

Wayne Reimer
April 1st, 2011, 07:24 PM
Ian,

Very cool footage..
Those cars are so incredibly fast, and the low shooting height really emphasizes the speed. I put a mount on the front axle stub on my Harley last year and shot some footage....no where even CLOSE to 300kkmh, but it looks like I was going far faster than I really was.

Finding that point of minimal/no vibration is a bit tricky but when you hit the sweet spot, those little cameras really perform!

Ian Newland
April 1st, 2011, 08:03 PM
The trick is to not use the Gopro mounts or case. It induces too much vibration. The case mounting prongs flex due to the mass of the camera and the case. Without the case, the 1080p mode (110 degree FOV) can be used at 25p vs using 720p mode ( 170 degrees) at 50fps. If the horizon is center frame, there is not much evidence of the fisheye/distorted look. However 50fps is smoother.

If you do use the housing, attach the camera with a 4 sided clamp arrangement, rather than the single bottom provided prongs. CMOS Jello comes from tilt rotation, if you eliminate or limit this jello is minimized.

To explain this better, if you raise the camera 2ft vertically but keep it on the same horizontal plane there is little change relative to the horizon. Think (crane or vertical slider). Even if you move the camera up and down 2ft 20times per second, there will be no jello. Tilt the camera on it's horizontal plane though and the horizon moves from the center to the top or bottom of the frame, this is what causes the extreme jello as the sensor scan reads top to bottom.

R

Ian Newland
April 2nd, 2011, 01:15 AM
Here's the results of some tests I've been doing with mounts that don't utilize the GoPro housing prongs. The mount location and mount is chosen to minimize Jello and picture distortion.

There's 720p HD or SD version or you can download the uploaded file if you are a vimeo member.

And yes we drive on the wrong side of the road Down Under.

Car on Vimeo

Ian Newland
April 2nd, 2011, 03:44 AM
Now this is cool. It's an OFX plugin for Sony Vegas that adds a very configurable motion blur to pixels when they hit a preset motion threshold.
What this does is it replicates the video as if it was shot at a shutter of 1/50 (which is the correct 180 degree shutter setting for 25fps) when in fact the video was shot at somewhere near 1/750 - 1/1000

The visual effect is that it removes the frame judder/strobing/hyper-real effect that is hard on the eyes. Notice the trees/cars blur instead of strobe, notice when i walk in the shop i have a nice motion blur but everything else is sharp. Also the far way objects are still sharp in the center of the driving clip.

This sample taken from the video above could be reduced a little, but i wanted to make it easy to see the effect.

motion blur3 on Vimeo