Shooting streetball; the next Olympic sport

www.paqphoto.com

In 2020, 3X3 basketball will be played for the first time at the summer Olympics, held in Tokyo. I’ve been shooting the sport since 2013, when I happened upon a game in Rotterdam. The game is fast and action-packed, even compared to full court basketball. It is also considerably more aggressive, in keeping with its origins as “streetball”, corner pick-up games on urban courts.

Unlike full court basketball, 3X3 basketball is usually played outdoors, and the court is much smaller. What this means to the photographer is that it is possible to get very close to the action and the lighting is almost always superior. I have been close enough to the action at times that I have to be very careful to avoid being stepped on when players go out of bounds. More than once, my camera or I have been hit with the ball. On other occasions, I’ve caught the ball when it goes out of bounds and tossed it back into the game—not always easy with a Phase One camera in one hand.

When I first saw 3X3 basketball in Rotterdam, I was in the city to test out my new Zeiss 15mm lens on a still fairly new Nikon D800. The organizers liked the photo, and invited me to shoot the next game in Den Haag. Encouraged by their positive reaction, I picked up a Nikon 35mm 1.4G to shoot the game with. Since then, I have upgraded my cameras and lenses many times. Currently, I usually shoot the games with a Phase One XF-100 MP camera, and Schneider-Kreuznach lenses. I have shot with the 28mm, 80mm, and 150mm. Of the three, I think I prefer the 28mm, but shooting with it is difficult because I have to be very close to the court to use it. A few weeks ago, in Amsterdam, a player grazed the barrel of the lens as he passed, and then in Den Haag, the ball thankfully hit me in the head rather than the camera. One of these days, I will get a 55mm to shoot the game from a safer distance, without sacrificing the wide-angle compositions I like.

This year, I decided to perform some tests to see what was the best way to shoot this game. The tests were:

·         Shooting speed vs. image quality (DSLR vs. MF)

·         High ISO vs. low ISO

·         Intentional motion blur

·         Off-camera lights vs. natural light

·         Auto-focus vs. manual focus

These specific tests were chosen because, after every game, I inevitably started second-guessing some of the quick decisions made during these fast-paced games. Shooting speed was an issue because of the high number of shots discarded during the editing process, usually nine out of every ten, with an average of 2,000 images shot per game. Even after discarding 90% of the images, I usually only had about 10 shots that I liked well enough to think of them as potential portfolio material. Most of the shots were discarded because they were out of focus, or the critical moment was missed. I wanted to see if using a DSLR instead of the Phase One would improve the number of keepers after each game. This was because my Nikon D800 and Sony A7r can both shoot more images in a second than the Phase One. With the Phase One, I had to wait until I thought the key moment would happen, then press the shutter release and hope I got it because I wouldn’t be able to take another shot until the players were doing something else. The Nikon and Sony can both fire multi-picture bursts, making it less likely I would miss key moments.

Sometimes, the games are played indoors due to rain. Sometimes this isn’t so bad, because the venue will have huge windows (as in Tilburg). Other times, the venue may not have any windows at all, and one location in Utrecht has two rooms where all of the walls and ceiling are painted black—in addition to not having any natural light. Shooting in that location forced me to start increasing the ISO from ISO 50, my favorite setting, to as high as ISO 12,800. To test the breaking point, I did a shoot in Utrecht where I varied the ISO considerably, to see how far it could be pushed.

Getting motion blur on the players was both easier and harder than I expected. Easier because it isn’t that difficult to shoot at a slow speed, harder because it wasn’t easy to decide what was the right slow speed, nor easy to figure out what part of the player was most likely going to be still enough to get a good action shot. I tried this indoors and outdoors.

For the last couple of years, I’ve been bringing a ProFoto B1 flash to games, and usually an assistant as well, to aim the light at the ball during the game. This sometimes yields very dramatic lighting, and sometimes artificial lighting. I hadn’t shot using natural light in more than a year, so this year I decided to leave the light at home for a couple of games and see what happened.

Tracking high speed action from between one and four meters away is very hard to do. All of my cameras have had a hard time with this at these games, and manual focus doesn’t always give the desired result either. After shooting most of the games in auto focus, to test the Nikon/Sony/Phase One AF systems, I shot the last game in manual focus to see what happened. The goal, like the others, is to reduce the number of discarded photos due to blur.

Last, I did a studio portrait shoot of one of the teams, the Rotterdam Concrete Lions, who were nice enough to come into the studio for a day to pose.

DSLR vs Medium Format

                I did my first DSLR/MF test at a game in Breda using the following cameras: Phase One XF-100, Nikon D800, Sony A7r. I could not compare the cameras with similar focal length lenses because I didn’t have enough similar lenses to go around. Instead, I put a range of lenses on the cameras so that I could shoot wide, normal, and close-up. The Phase One had an SK 28mm 4.5, the Nikon had an 85mm 1.4G, and the Sony was mounted with the Zeiss Alpha 135mm 1.8. This test was done to determine which camera did the best job of capturing an in-focus shot using auto focus (figure 1).

                While I was shooting, it seemed like the Sony was by far the best at getting in-focus shots. This was because, after each shot, the image would flash up in the electronic viewfinder, and they almost always looked like they were in perfect focus. However, when I got home and put the images in Capture One for processing, the Sony images didn’t hold up as well. Overall, AF accuracy seemed close for all three cameras, with a slight edge to the Nikon. That said, superior resolution produced by the Phase One also had the effect of making its in-focus shots sharper than anything produced by the Nikon.

Color was a serious issue. Though it wasn’t something I intended to test at first, the 16-bit color depth of the Phase One images stood out prominently in comparison to the DSLR images. While editing the RAW files, images produced by the Nikon and Sony had little room for adjustment. The Phase One files, on the other hand, had tremendous dynamic range that could be manipulated in processing. The color was also much cleaner in the Phase One files. To put it bluntly, the color from the Nikon and Sony looked terrible compared to the Phase One—as would be expected with the higher bit depth. On this basis, I decided that the shooting speed and AF accuracy of the Nikon and Sony were not enough to make up for the superior color and resolution of the Phase One.

Figure 1 image quality comparison, Phase One XF-100, Nikon D800, Sony A7r

Low light ISO test

                Shooting in the dark isn’t much fun, especially when everything has to be shot at f/2.8 to maximize the light. This makes for a thinner than ideal focal plane, and very difficult focus. As it happens, I didn’t get any shots in focus at ISO 6400, and most at 3200 and above are soft. The grain was acceptable to me at 3200, but the falloff in colour quality after ISO 1600 was quite serious. For that reason, I converted the tiny number of usable shots made at ISO 3200 and up to black and white. The full resolution crops represent only a tiny portion of the surface area of these images, which tends to exaggerate the grain (figures 2 and 3).

Figure 2 ISO test, from ISO 800 to ISO 12,800

Figure 3 f/4, 1/1000s, ISO 3200, SK 80mm LS

Motion blur

                I didn’t do very many motion blur tests because I was worried about missing shots, but in the end, some of my favourite photos this year have strong motion blur (figures 4-8).

Figure 4 f/7.1, 1/125s, ISO 200, Nikon 85mm

Figure 5 f/5.6, 1/125s, ISO 800, SK 80mm LS

Figure 6 f/2.8, 1/250s, ISO 800, SK 80mm LS

Figure 7 f/2.8, 1/250s, ISO 800, SK 80mm LS

Figure 8 Extreme blur f/32, 1/3s, ISO 50, SK 28mm LS

Flash and natural light

                The B1 flash was essential to create dramatic lighting in the indoor venues (figures 8-10). For these shots, there was a very high skylight to illuminate the background with natural light. The B1 with a ProFoto octabox was used as a fill light. It is nice that flash is allowed at these games, because without it, it would have been much harder to shoot such fast-moving action in such a dark room.

Figure 9 f/2.8, 1/1500s, ISO 800, SK 80mm LS

Figure 10 f/3.4, 1/250s, ISO 800, SK 80mm LS

Figure 11 f/2.8, 1/1500s, ISO 800, SK 80mm LS

                Outdoors, flash was extremely useful to illuminate players with the sun behind them, or to create a rim light to clarify the edges of tangled limbs near the basket (figures 12-15).

Figure 12 f/4.5, 1/1500s, ISO 100, SK 28mm LS

Figure 13 f/4.8, 1/1500s, ISO 50, SK 80mm LS

Figure 14 f/5.6, 1/500s, ISO 50, SK 80mm LS

Figure 15 f/6.8, 1/700s, ISO 50, SK 28mm LS

                Natural light worked well given the high dynamic range of the XF-100, but in some situations, it would have benefitted from a flash (figures 16-19).

Figure 16 f/5.6, 1/2000s, ISO 800, SK 28mm LS

Figure 17 f/8, 1/2000s, ISO 800, SK 28mm LS

Figure 18 f/4.5, 1/2000s, ISO 400, SK 28mm LS

Figure 19 f/4.5, 1/1500s, ISO 400, SK 28mm LS

Manual focus

                The manual focus test was quite successful. Not only did I have many more in-focus shots than before, but when they weren’t in focus, many were acceptable, regardless. This is because the range of focus prevented the kind of problems experienced with autofocus, where trees in the far background were as likely to be in focus as the players. For the day I shot manual focus, I also didn’t have a light, so I was reliant on natural light. Also, I alternated between shooting with the Phase One with a 28mm lens and the Sony with a 35mm Leica Summilux. This was so that I could easily give the sponsors JPEG shots for their website while also getting 100MP shots for prints (figures 20-25).

Figure 20 f/4.5, 1/1000s, ISO 400, SK 28mm LS

Figure 21 f/4.5, 1/1500s, ISO 400, SK 28mm LS

Figure 22 f/4.5, 1/1500s, ISO 400, SK 28mm LS

Figure 23 f/4.5, 1/1500s, ISO 100, SK 28mm LS

Figure 24 f/4.5, 1/1500s, ISO 100, Leica 35mm summilux

Figure 25 f/1.4, 1/2500s, ISO 100, Leica 35mm summilux

Studio shoot

                I’ve been wanting to get some basketball players into a studio for a shoot for some time. Finally, I was able to arrange it, and rented a studio up near Amsterdam for the purpose (figures 26-31). The players, Christiaan Grives, Manuel Tumba, Miquel Pocorni, and Argil Randon-le Couvrier, are from the Concrete Lions 3X3 basketball team.

Figure 26 f/4, 1/30s, ISO 50, SK 120mm LS

Figure 27 f/8, 1/350s, ISO 50, SK 120mm LS

Figure 28 f/4, 1/30s, ISO 50, SK 120mm LS

Figure 29 f/11, 1/250s, ISO 50, SK 120mm LS

Figure 30 f/9.5, 1/250s, ISO 50, SK 120mm LS

Figure 31 f/9.5, 1/250s, ISO 50, SK 120mm LS