Editor’s Note:This article is the final installment of the 2015 Mtbr/RoadBikeReview POV Camera Shootout. We’ve also published individual reviews of the Drift Ghost-S, Garmin Virb Elite, GoPro HERO4 Silver, iON Air Pro 3 Wi-Fi, Shimano Sport Camera and Sony Action Cam Mini. Read all POV Camera Shootout articles HERE and you can download full resolution video files from all the tests HERE.
We did the first Mtbr/RoadBikeReview POV shootout in the spring of 2013 and it was a huge success. It was one of the first proper POV camera tests anywhere on the Internet where all the cameras were mounted together and recorded the exact same thing. Since then, a whole new generation of POV cameras has been introduced with improved features and performance. All the cameras in this year’s shootout record at 1920 x 1080 full-HD and 4 of them are able to record full-HD at 60 FPS for even better quality. All the test cameras also have built-in Wi-Fi with mobile apps, so you can start and stop recording, change settings, and see a live view display on your smartphone.
There are four parts to our 2015 POV camera shootout: the bike shop intro, the trail test, the tunnel test and a high-speed slow-motion demo. Each test was used to compare and evaluate different qualities of the cameras’ video performance. Along the way, we also got a good feel for how to use each camera. After all the testing was done, we put together the final video (below) and analyzed and compared all the footage to determine which cameras were the best. To learn more about the video and see our 2015 POV camera picks, keep reading. To download full resolution video files from all the tests, go HERE.
Bit Rate and Video Quality
Before we get into the POV video analysis, let’s talk a little about bit-rate and video quality. In the past, when you’d look at POV camera footage, the first reaction was, I must have done something wrong. No matter what the camera or settings, the video always looked super mushy — especially if there was a lot of motion. As a DSLR video shooter, that’s not at all what I’m used to.
Most POV cameras record at a relatively low bit rate and use a lot of compression. The result is blurry action video that lacks detail. The faster you get moving, the softer the video gets. However, GoPro’s Protune and Sony’s XAVC S format (Pro mode) record about twice the amount of data as the other cameras in this test (45 Mbps and 50 Mbps, respectively). The result is footage with much smoother tonal blends, finer detail, and better color rendition. It’s proper looking video.
In our last POV camera shootout all camera settings were kept as close to the same as possible, ignoring GoPro’s Protune mode. So even though the GoPro came out on top in that shootout, our test didn’t show its full potential. For the 2015 shootout, we tested the GoPro and Sony at their true maximum quality, using GoPro’s Protune and Sony’s XAVC S Pro mode. If you really look at the GoPro and Sony footage in this test, you’ll see they look a lot better than the other videos, especially in the fast bits of the trail video.
Bike Shop Test
The bike shop video was filmed in 1920 x 1080 full-HD at the fastest frame rate and highest quality offered by each camera. Although this shootout is mainly to show the video quality on the bike, most of users also end up using POV cameras to film off the bike as well. This part of the video, filmed at Over the Edge in Hurricane, Utah, is perfect for evaluating video quality indoors, with non-organic, manmade details and people. Since the light was pretty low inside the shop, it’s also a good place to evaluate low light performance.
One of the first things you’ll notice in the bike shop footage is the lens distortion. Every camera but the Ion makes the walls and ceiling look curved. The Shimano’s 180° lens has the most distortion, although it does show more of the space than the others. Ultimately, we thought the GoPro HERO4 Silver’s high bit-rate Protune footage looked best in the bike shop. Although the exposure was a tad dark, details in the HERO4 Silver’s footage were better than the other cameras, and the sound was very good. The Sony Action Cam Mini, which also features higher bit-rate footage, came in second. The Garmin was the worst performer in the bike shop, with very soft, blurry video quality.
The trail footage is the most important part of our test since that’s how most of us actually use our POV cameras. For the trail video, we set all the cameras to their fastest frame rate and best quality at 1920 x 1080 full-HD resolution. In response to the comments on our last POV shootout, we stepped up the trail video this time by following another rider (thanks, Clayton Coleman of Over the Edge, Hurricane!) and using a more entertaining trail. Riding anything really rowdy is out of the question because the test helmet weighs over 7 pounds and is completely unsafe. But this section of Jem Trail in Virgin, Utah, is relatively safe while still delivering some solid cheap thrills since it’s literally right on the edge of a cliff. The background scenery isn’t bad, either.
As mentioned earlier, one of the main problems with POV camera footage is detail loss when you really get moving. This is most obvious in faster bits of the trail video, except for the Sony and GoPro footage, which was recorded at a much higher bit rate than the other cameras are capable of. So of course the Sony and GoPro cameras come out on top. Which one looks better is a matter of personal taste. The GoPro footage has better detail than the Sony. However, the color and contrast in the GoPro footage are very flat and need some editing work to really look good. The Sony looks good right out of the camera, although the detail isn’t quite as good as the GoPro. It’s still far better than the four other cameras in our test, though. Sony’s built-in Steadyshot image stabilization also gives the Action Cam Mini the smoothest trail footage of all the cameras. You do sacrifice some scenery for the stabilization, though; turning on the Steadyshot image stabilization changes the angle-of-view from 170° to 120°. More experienced videographers may decide to leave the in-camera image stabilization turned off and correct camera shake in post, instead. That way you can fine-tune the anti-shake to your own taste, although that requires cropping the original full-HD footage, theoretically compromising quality.
Worst trail video is a tie between the Shimano and Garmin cameras. Part of that is due to both cameras maxing out at 30 FPS in full-HD; the other cameras record full-HD video at 60 FPS. The Garmin also suffers from over-saturated and surreal-looking color. With the Shimano, the problem is the lens. Between the crazy 180° angle-of-view and the plastic lens cover, sun flare is out of control, sometimes completely obscuring the rider in front of us. I believe the Shimano’s extreme wide angle is also a little motion sickness-inducing, something I didn’t experience with any of the other video footage.
The purpose of the tunnel footage is to test the responsiveness of the cameras’ exposure systems. You know how POV videos are often alternately too bright and too dark when a rider is going in and out of trees? Watch how the video reacts when we go in and out of the tunnel. Coming back out for the second time, you can clearly see the mountains resolve much sooner with the Drift, GoPro and Sony cameras. The outside details, including the sky, form the soonest with the Sony footage, followed by the GoPro, and then the Drift. The Ion fared the worst in this test, with sky detail not appearing until we’re well out of the tunnel and starting back up the hill.
Slow Motion Test
The high-speed portion of our shootout was filmed at Wasatch Indoor Bike Park in Salt Lake City, Utah, with each camera set to the fastest available frame rate at 720p. That’s 120 FPS for all of the cameras except the Garmin, which maxes out at 60 FPS. For those who don’t know, one of the main reasons to shoot at a frame rate faster than 30 frames per second is so you can slow the footage down and watch it in slow motion. The faster the frame rate, the more you can slow down your footage without losing quality. The video here was imported and edited at 120 FPS then slowed down to 30 FPS in post. Honestly, this is more of a slow motion demo than a real test, since it’s hard to discern any real quality differences in this test. However, if you look carefully, you can see the Garmin slow motion footage looks rougher and isn’t nearly as smooth as the clips from the other cameras. Faster is definitely better if you want good-looking slow-motion video.