Tuesday, September 26, 2017

Friday Night Lights (with Crappy Lights) - Part 1

I know, I know, everyone thinks that the lights on their high school athletic fields are the worst but after a little over a year of shooting Friday night HS football games I can safely say that Algonquin high school has the worst lights around. I am an engineer at heart so I am trying to learn the technical aspects of how light impacts exposure and how the available light can be quantified and normalized for comparison. I understand how to utilize the exposure triangle to properly set up my camera and shoot in manual mode almost exclusively but since light is such an important aspect of photography I want to learn as much about it as possible. My primary goal of this blog post is to demonstrate the impact of lighting on picture quality, using high school football as an example, so that people understand why some photos look better than others. My secondary motivation is to justify to myself, and more importantly to my wife, the benefit my new gear provides. This post turned out to be longer than I expected so am going to break it into two parts. In the first part, which you are currently reading, I will present a very basic primer on exposure and try to explain the relationship between shutter speed, ISO speed and aperture and how they need to be balanced to achieve the best exposure settings for the subject matter you are shooting. Knowing the subject matter is critical to setting the exposure because, even in the same lighting conditions, the optimal settings for sports, landscape and portrait work can be very different. In the first part I will provide an objective comparison of available light by normalizing the exposure settings I use on different fields. In part two of the post I will attempt to demonstrate the same principles with a subjective comparison of photos taken with different exposure settings on different fields with very different levels of available light with a couple different cameras. I think that the second part should be far more intuitive and demonstrate how critical adequate lighting is to achieving photos that pop but I wanted to present the theory first.

Exposure Theory

Every photographer will eventually move of the automatic modes to one of the manual modes (for Canon these are M, Av and Tv) in the camera and have to learn how to properly set up their camera. The term exposure refers to exposing the sensor (or film back in the old days) to the correct amount of light for just the right amount of time with the sensor set to the proper sensitivity so that the photograph is not too dark or too bright. In order to quantify the parameters that affect exposure a system called the exposure triangle was developed which consists of three components. Shutter speed, ISO speed and aperture are the three knobs you have to control exposure and adjustments to each parameter have secondary effects that will impact the way a picture looks. We will take a look at how to adjust each of these parameters below and what these secondary effects are and how they impact the final product.

Shutter Speed

Shutter speed is pretty simple to understand - it determines how long the shutter is open and determines how long the sensor will be exposed to the light. The required shutter speed for proper exposure is proportional to the amount of available light so doubling the shutter speed (i.e. going from 1/500th second to 1/1000th second) will reduce the exposure by 2x, or 1 EV, assuming a fixed ISO, aperture and light intensity. Maintaining the same exposure with the higher shutter speed  requires twice the available light. Put another way, you will need to shoot at slower shutter speeds as it gets darker if ISO and aperture are fixed. The secondary effect of shutter speed is how well the camera freezes motion - the faster the object you are photographing moves, the higher the required shutter speed. This is demonstrated on the graphic below where shutter speed is listed in one stop increments and the resulting image blur is depicted in the stick figure.



ISO impacts the sensor's light sensitivity and the required setting for accurate exposure is inversely proportional to the amount of available light -higher ISO speeds correspond to higher sensitivity. For example, for a given aperture and shutter speed, the darker it is the higher you need to set your ISO speed and for a given shutter speed and aperture, doubling the ISO will raise the metered light by 2x or 1 EV. The downside of increased ISO speed is that above some baseline level, which varies from camera to camera, the noise or grain in the picture also increases. This effect is demonstrated in the figure below where the amount of background noise increases with increased ISO speed. Noise can be managed and reduced in post-processing but is generally bad and one of the main reasons that professional sports photographers spend the big bucks on "fast glass", which refers to the aperture of a lens, enabling them to shoot at lower ISO speeds. Aperture is the next subject that will be covered.


I think that aperture is the most difficult concept in photography for people to understand because its relationship to exposure is not linear like it is for ISO and shutter speed. Also, aperture is expressed in a funny notation, like f/1.4 or f/2.8, and does not seem to make sense. The notation simply refers to the ratio of the focal length of the lens (f is for focal length) to the aperture diameter (the number on the bottom, or denominator) and can be used to find the aperture surface area. The aperture surface area determines how much light reaches the sensor and the larger the aperture, the more light gets through. However, unlike ISO and shutter speed, the relationship is exponential because it is based on area (powers of 2). For example, if you have a 50mm f/2.8 lens, the aperture diameter is 50mm/2.8 = 17.9mm. Remembering back to geometry, the surface area of a circle (the aperture) is pi x (diameter/2)^2 = 250.4mm^2. If, on the other hand, you have a 50mm f/1.4 lens, which is a 'faster' lens, then the aperture diameter is 50mm/1.4 = 35.7mm and the aperture area is 1001.8mm^2 or 4 times the surface area of the 50mm f/2.8 lens. Therefore, smaller aperture numbers result in larger aperture opening and more available light. This means that the 50mm f/1.4 lens will allow 4 times as much light, or 2 stop (2 EV), to hit the sensor than the 50mm f/2.8 lens for a given ISO and shutter speed and it will cost at least twice as much. Therefore, aperture scales with the square root of 2 (~1.41) for each stop of light. An f/1.4 lens is 2 stops "faster" than an f/2.8 lens and 1 stop "faster" than an f/2.0 lens. As the available light decreases you want to move to a larger aperture (lower f-number) for a fixed shutter speed and ISO. One of the secondary effects of aperture is that the depth of field, or how much of the subject is in focus, front to back, changes. The wider the aperture, or the lower the f-number, the narrow the depth of field. This can be used as an artistic element or it can present challenges to sports shooters if the depth of field is too narrow.  The other secondary effect of aperture is cost - the price of lenses seem to scale exponentially with aperture size as it is much more expensive to manufacture large pieces of flourite glass which is required for sharp, fast lenses. The graphic below shows the trade-offs between aperture, depth of field and cost.

Exposure Example

Finally, to pull it all together, the following illustrations demonstrate how to change either aperture, shutter speed or ISO speed to maintain constant exposure. Each figure shows shutter speed, aperture and ISO in one stop (1 EV) steps and arrows indicate the impact to exposure for adjustments in each direction. All three parameters have been plotted so that shifts to the left reduce exposure and shifts to the right increase exposure. If any one parameter is changed then some combination of the other two parameter need to be adjusted an equal number of steps in the opposite direction to maintain constant exposure.

Consider the following example. Assume that we are shooting a night football game and initially set the shutter speed to 1/500 second, the aperture to f/2.8 and ISO to 6400 as shown in the first plot. After a few shots we find that we are seeing too much motion blur and want to increase the shutter speed.  What are the options?

One possible solution is shown below - if we increase the shutter speed to 1/1000 second to freeze the motion and maintain the same aperture (f/2.8) then the ISO must increase to 12,800. Therefore, the shutter speed moved 1 stop to the left and ISO moved one stop to the right. The sliders are analogous to a balance scale - any change in one direction needs to be balanced by an equal number of steps in the opposite direction.

Another possible solution would be to maintain the same ISO of 6400 and move the aperture to f/2.0 if we are lucky enough to have such a fast lens. Again, for the one stop to the left in shutter speed we moved ISO one stop to the right.

If the shutter speed moves one more stop to the left, to 1/2000 second, then the three most likely options would be:
  1. Move the aperture one stop to f/2.0 and the ISO one stop to 12,800
  2. Move the aperture two stops to f/1.4 and leave ISO at 6,400
  3. Leave the aperture at f/2.8 and move the ISO two stop to 25,600
All three choices produce the same exposure but they all have slightly different side effects. Choice 2 will (probably) produce the best quality photo but is the most expensive as an f/1.4 lens will be quite costly. Option 3 is the least expensive but will produce the noisiest photo since the ISO is so high. Choice 2 is the best compromise and should produce an acceptable photo. 

Comparing Available Light based On Exposure

I wish that I owned I light meter so that I could quantify the light level at each of the football fields I shoot at but I will have to settle for an online EV calculator I found for this comparison that estimates the available light from exposure settings (shutter speed, aperture and ISO). The calculator allows you to convert an in-camera exposure setting at a specific ISO to any other ISO for comparison. The calculator also computes the light intensity, expressed in terms of EV (Exposure Value), and normalizes this to ISO 100. Most cameras meter available light (usually reflected light for photography) and express it in term of +/- EV on the light meter. EV is in powers of 2 so each increment of one or decrement of one is a doubling or halving of light, respectively. In photography each 1 EV step is referred to as a stop. ISO, shutter speed and aperture will be adjusted based on the subject matter and the available light.

After half time, when the sun has completely set, I was setting exposure based on only the available reflected light from the field lights and I was shooting with a shutter speed of 1/1000 second to stop action as much as possible, aperture of f/2.8 and ISO 20,000-25,600. According to the EV calculator this is equivalent to 5 EV, or approximately 80 LUX, at ISO 100 with the same shutter speed and aperture. That is dark! During the most recent game that I shot at Wachusett I was also shooting at 1/1000 second and using an f/2.8 lens but my ISO speed was between 3200 and 6400 for most of the game. That means the field was at 7-8 EV and lights at Wachusett are between 4 and 8 times brighter than the lights at Algonquin. From what I could find on line most college and NFL stadiums are 8-9 EV, which is 3-4 stops more available light (that is 8-16 times more light) allowing me to drop from ISO 25,600 to at least 3200 if not 1600 in a good stadium. One can only dream.

I will post part two in the near future and show examples of photos taken under each of the lighting conditions described.

Gear used in this comparison:

Wednesday, September 20, 2017

Photographing Cross Country

As much as I enjoy shooting football my favorite sport is probably track and cross country because that is the sport I did. I suppose that there is not as much action but I ran cross country all through high school and college and can feel the struggle all of the kids are going through on the course. I had a minor lull in blogging due to some technical issues with my PC (let's just say, I will not be using liquid cooling in my PC anymore) but I am back on track now.

Both the boys and girls cross country teams traveled to Shrewsbury High School on September 12th for a dual meet and I brought my 5D Mark IV, 1D X Mark II, Canon EF 70-200mm f/2.8L IS II USM lens and Canon 300mm f/2.8 IS II USM lens for the shoot. This was a very difficult race to shoot because I had no idea where the race started or finished, there did not seem to be any good places to shoot from and the sun was very bright casting harsh shadows on all of the athletes. I found the start but did not get too many good shots from there and then I had to hustle, carrying all of my gear, to the finishing area, which was about half a mile away, in time to catch the runners. On top of the confusion about the course it was almost 85 degrees and I do not enjoy hot, humid weather.

It was really cumbersome running around with the 300mm f/2.8 lens and I was a bit discouraged on the way home and did not think that I would have many good shots. However, I think that the photos came out all right in the end.

Slide show - I still can not get the embeded SmugMug slideshows to work.

The next race was home at Algonquin High School against Leominster High School. The conditions were very different as we are getting the leading edge of Hurricane Jose. Based on how awkward it was carrying around the 300mm lens I decided to go with the Canon EF 70-200mm f/2.8L IS II USM lens with the 1.4x teleconverter figuring that I would have to stop down to at least f/4.0 to get a large enough depth of field when the runners were grouped together. In the end I kind of wish that I had the 300mm lens because it provides incredible sharpness, isolation and compression. Plus, you really look like a pro carting that thing around.

The pictures for this meet came out really well except for the white balance. I have gotten into the habit of shooting RAW for night games for the extra flexibility but shooting JPG for day games to speed up the post processing. For some reason I had to make minor adjustments to all of the photos in these galleries to get decent skin tones and there were a few photos that I just could not save. I may revert back to using the 300mm lens, even for the cross country, and shooting in RAW so that I can apply a blanket WB setting across all images in a gallery.

Slide show of boy's Leominster race
Slide show of girl's Leominster race

Gallery links:
Shrewsbury race
Girl's Leominster race
Boy's Leominster race

Sunday, September 10, 2017

Algonquin vs Worcester North - 2017.09.09

The Algonquin THawk football team kicked off the season with an away game against Worcester North at the iconic Foley Stadium. Saturday was a beautiful day for playing, watching and photographing a high school football game. It was sunny and 70 degrees with some scattered clouds which are great for diffusing the sun and eliminating harsh shadows. This also marked my first regular season game with my new gear which should, hopefully, help me generate higher quality pictures. So far I think that my pictures are better than last year in the two test runs that I did at the team's scrimmages (you can view the pictures from the Clinton and Franklin scrimmages on www.CrainPhotos.com). The pictures from the Clinton scrimmage were decent because it was a day game with similar conditions to today's game. The Franklin scrimmage, on the other hand, was a night game at the very dark Algonquin stadium. I am working on a future blog post about this but I think that the pictures with the Canon 1Dx Mark II and 5D Mark IV were much better than the pictures I took last year at night games with the 5D Mark III and 7D Mark II.

Slideshow: I am working on getting the SmugMug embedded slideshow to work but it seems like Blogger has an issue with the HTML5 code that SmugMug generates. Stay tuned.

I shot this game with two basic setups but I added a twist in the second half to mix things up a bit. My main camera for the day was the Canon 1D X Mark II with the Canon 300mm f/2.8 IS II USM lens and my second camera was the 5D Mark IV with the EF 70-200mm f/2.8L IS II USM lens for when things got close. In the second half of the day I added the Canon EF 1.4x Mark III teleconverter to the 1D X rig for an effective focal length of 420mm and a maximum aperture of f/4.0. The reduction in aperture was not a big deal because it was so sunny and I was still able to shoot with a shutter speed well over 1/1000th second. I wanted to see how the pictures would turn out with the extra reach because it forced me to shoot from much further away from the line of scrimmage (usually from the back of the end zone) and really changes the perspective of the pictures. I almost exclusively shoot in manual mode with auto ISO for football but I decided to shoot in aperture priority mode (AV) today with the aperture set to f/2.8-f/3.2, or f/4.0 with the teleconverter, and ISO 100-320, depending on the cloud cover. The shutter speed stayed above 1/1000th second for all shots and reached as high as 1/5000th second with direct sunlight. All photos where in RAW format - I have gotten into the practice of shooting RAW instead of JPEG because of the flexibility it gives me in post for the night games to fix white balance, exposure and noise issues. I liked using the teleconverter and that setup worked particularly well for this game because the Worcester North team ran 90 percent of the time and I could zero in on the runner and the line men. I did all of my editing in Lightroom CC and pretty much just tweaked the white balance and exposure and bumped shadows and blacks.

The next game will be on September 15th vs Marlborough at Algonquin so it will be another dark night of shooting. This whole week will be busy with the JV football game, the varsity Cross Country meets and Applefest next weekend. You can see my full schedule here.

Tuesday, September 5, 2017

Shooting the Moon

I have never tried photographing the moon and decided to take a shot the other night. The moon is  in the Waxing Gibbous phase (97% illumination) and will be full on 9/6 at about 3:02AM EDT so I figured it was a good time to try. After doing a little research online and waiting for it to get dark enough I headed out into the driveway and started shooting.

Waxing Gibbous Moon shot with Canon 5D Mark IV (f/11, 1/200 sec, ISO 200)
I have a couple cameras at my disposal but figured that a full-frame camera would be better suited to the task than a camera with a crop sensor. To test my theory I used both the 5D Mark IV and 7D Mark II - I was not sure if the higher resolution sensor with better dynamic range would be better or if the lower resolution camera with the 1.6x crop factor would win out. I used the 300mm F/2.8 IS II USM lens and the 1.4x Mark II teleconverter to get as much reach as possible. On the 5D the effective focal length was 420mm and on the 7D the total focal length was 672mm. I used basically the same settings for both cameras:
  • 5D Mark IV: f/11, 1/200 sec, ISO 200 w/ auto focus with center point
  • 7D Mark II: f/11, 1/100 sec, ISO 100 w/ auto focus with center point
One other item that I almost forgot to mention that played a crucial role in getting clean pictures is that I used a tripod and a remote or delayed shutter trigger on both cameras. I tried hand-holding the camera for the first few shots and this failed miserably. The camera and lens weigh about 10 pounds together and even the smallest amount of shake made it hard to even keep the moon in the viewfinder since it is so far away. I quickly broke out my tripod and got back to work. I also tried manually pushing the shutter button but I have a fairly crappy tripod and the camera shook for quite a while after I pushed the shutter button. I tried the remote trigger on the 5D IV through the EOS app, connecting through WiFi, but did not have much success with that either. I think that was purely user error though because the EOS app itself seemed pretty flawless - it was the first time that I used it and I could not get good focusing results. I settled on the 10 second delayed shutter option because this worked for both cameras and was the most straightforward solution. The only challenge was that there were fast moving clouds in the sky and I had to time pressing the shutter button, accounting for the 10 second delay, so that the moon was visible through breaks in the clouds.

In post I adjusted the RAW images in Lightroom CC by converting them to black and white, adjusting the exposure, increasing the highlights some and adding a bit of clarity and sharpness. The first picture above is from the 5D Mark IV and I think that this one is better overall. The bottom photo is from the 7D and while the craters on the bottom left portion of the moon are more detailed there is less detail overall in the middle of the picture. I guess that even though the total focal length on the 7D was effectively 60% higher (1.6x crop factor) the higher resolution and better sensor in the 5D made up for the difference after crop.

Tonight is a full moon so, if it is not too cloudy, I will take another shot and see if I can get better results.

Waxing Gibbous Moon shot with Canon 7D Mark II (f/11, 1/100 sec, ISO 100)

Sunday, August 27, 2017

New School Year, New Toys - 2017.08.27

A new school year is approaching which means fall sports will begin soon and the bulk of all the shooting I did last year was high school sports for Algonquin Regional High School. One of my biggest struggles during the varsity football games (Friday Night Lights baby) was getting a decent balance between shutter speed and ISO speed that provided reasonably low noise and sharp pictures. I bought the Canon 5D Mark III and 7D Mark II at the beginning of the season and found both cameras work great during day games but they both struggled (especially the 7D) shooting night games or indoor sports (cheer). I had a lot of out-of-focus shots due to the mediocre auto-focus, the colors seemed a bit muddled and the shots were just not sharp. I am willing to admit that some of the issue was probably user error (I did not know about the focus micro-adjustments for quite some time) but based on review that I have read, both of these cameras struggle a bit in dynamic range and high  ISO noise levels. I came to the conclusion that my cameras were limiting the quality of my pictures and the only way to overcome this was to upgrade my gear.


This is where the new toys come in. I am keeping most of the gear that I bought last year as the EF 24-70mm f/2.8L II and EF 70-200mm f/2.8L IS II USM were my workhorse lenses. However, I have updated to the Canon 5D Mark IV and I added the Canon 1D X Mark II to the line-up. The real improvement of the 5DIV is not in the 30.4MP sensor (the 5DIII has a 22.3MP sensor). Rather, the real benefit is in dynamic range and relative ISO invariance (ability to adjust exposure over a large ISO speed range with little impact to noise performance). This camera should allow me to get sharper and cleaner pictures at high ISO during night games. I am probably the most excited to work with the 1D X Mark II - this is a professional level camera and pretty much the top of the line DSLR currently offered by Canon. The 1D X II only has a 20.2MP sensor but it can shoot at 14fps in bursts up to 170 RAW images before buffering. I am not one to follow the spray and pray mentality but having the higher frame rate provides more flexibility in post production and increases the odds of getting the exact moment in action that you want. For comparison, the 5D and 7D families shoot 5-10fps (allegedly) and buffer with fewer than 30 RAW images. The auto-focus system is also second to none (Nikon D5 and Sony A9 owners will dispute this).


To round out the kit, I also added the Canon 300mm f/2.8 IS II USM telephoto lens. Let me tell you now, this is a big lens! I was not really prepared for how large and heavy this thing is and it pretty much forces you to use a mono-pod. The other thing that I was not really expecting is that the lens is so long it forces you to shoot a different perspective than with the 70-200mm. It will take me some time to figure out the right balance between the two, which will most likely be my two primary lenses for a game.

I got a chance to shoot during a practice session for the Algonquin football team and again, the next day, at their first scrimmage. I have included a few pictures and a link to the scrimmage picture gallery to show how my first days with the new kit turned out. My first impression is that the 1DX Mark II with the 300mm lens is simply awesome. The pair is significantly better than the setup that I had last year and even the 5D Mark IV seems slow comparatively even though it shoots 7fps compared to the 5D Mark III's 5fps (I actually tested the cameras and the 5D IV does shoot a sustained 7fps). I have never believed the claim that the 7D Mark II shoots 10fps and I recently tested it and measure only 8fps with the Lexar 32GB 1066x CF card. The high frame rate can be seen in the pictures below - the series was shot with the 1DXii and 300mm lens and I got eight sharp and in-focus shots of the receiver catching the ball. I can select the best shot from the series for my gallery post or create a narrative with several of them to show the action.

I usually start a play by focusing on the quarterback and then either follow the runner in a run play or try to pan to the receiver in a pass play. I had good luck with that in the past because the lens was much smaller and I could swing the camera quickly but now the lens has quite a bit of mass and it is mounted on a mono-pod making it a bit more difficult to pan. I think that things work out in the end because of the faster auto-focus in the 1DX. In another series where the defender came out on top stopping a short slant I was able to capture four shots in the series demonstrating how fast and accurate the auto-focus is.

At the team scrimmage I was able to get another 2-3 hours of practice on the new setup and took about 900 pictures. I posted the full gallery of 94 pictures here but I included some of my favorite shots below to demonstrate how sharp the pair is (and how I am improving). I shot in RAW on the 1DX Mark II and used manual mode on all cameras (I also had the 5D Mark IV and 7D Mark II which I shot JPEG) with the aperture set between f/2.8 and f/4.0, I used auto ISO and set the shutter speed 1/2000th - 1/5000th to keep the ISO between 100 and 250. All three cameras were set up with AWB and no exposure compensation. I did mix-and-match the lenses and bodies and I also threw in a 1.4x Mark III teleconverter for fun. One experiment that I did was shooting from the back of one end zone with the 7D Mark II (which has a 1.6x crop factor) paired with the 300mm lens and a 1.4x TC for an effective focal length of 672mm. This was very difficult to keep steady, even with the mono-pod, and was almost too much zoom even when the teams were down at the other end zone. All photos were edited in Adobe Lightroom CC 2015.12 for exposure, WB and composition (crop).

Wednesday, April 26, 2017

Marlborough AYF Cheer-A-Palooza - 2017.04.22

All Photo Galleries

I took another shot at photographing cheer this weekend at the Marlborough, MA American Youth Football (AYF) Cheer-A-Palooza fundraising event at the Whitcomb Middle School. The event included several local vendors, a bouncy house for the kids and an exhibition by four Marlborough AYF squads and one Northborough/Southborough AYF squad. I have not been too pleased with how my previous attempts have gone and I spent quite a bit of time experimenting with different lenses, cameras and settings. I have a better idea of what settings to use on the camera and which lens/camera combinations work best. Ultimately, I would like to upgrade my gear to the Canon 1Dx II and do some more testing with fast lenses like the Canon EF 135mm f/2 and 85mm f/1.8 or the new Sigma 135mm f/1.8 Art. I hear that the Sigma is tack sharp and has a consistent focus but I don't know how fast the focus is.

When I arrived at the event I was surprised at how good the lighting was, especially for a middle school gymnasium. The reason that it was so bright is that there were open skylights along one whole side of the roof letting in natural light, despite the cloudy conditions outside. Unfortunately though, the event organizers decided to arrange the mats to have a more "intimate" event. This translates into poor conditions for a photographer because the bleacher seating came right up to the edge of the mats and the skylights in the roof were behind the cheerleaders causing a backlit situation. I was able to convince the organizers to move the seating back 5-6 feet but there still was not enough room to shoot head-on and I had to shoot at an angle at the cheerleaders left side. My bad luck continued into the routines as every one of the teams looked to their right when they were in their pyramids and I was shooting the back of their heads. Is that a thing or do I just have bad luck?

Since I did not have much room to work with and I wanted to get as many of the girls in the shots I decided to use the 7D Mark II with the EF 24-70mm f/2.8L IS II USM lens. I wanted to use the 5D Mark III but I was also trying to do video and the widest lens I had left was the Sigma 50mm f/1.4 Art and even this was too wide on the cropped 7D. In the end the video did not work out, mainly because the 50mm lens was not wide enough and people walked in front of the camera, so I probably should have just used the 5D for the photos.

I shot in manual mode, wide open (f/2.8), the shutter speed set to 1/800th of a second, auto ISO limited to 12,800 and auto WB. With the lighting as good as it was the average ISO was 2500-3200 and the maximum was only 6400. Because the 7D has the flicker compensation I only had to do some minor exposure and tone adjustments, slight noise reduction and cropping in Lightroom. Links for the final galleries for each of the five performances are at the top of the page along with a gallery of general shots that I took of the event before the routines began.

Thursday, March 23, 2017

T-Hawk Cheer Practice Shoot - 2017.03.23

I have been trying to improve my skills for shooting cheering competitions and arranged to do two pre-meet shoots with the Northborough/Southborough AYF T-Hawk squad. Cheer presents a unique challenge because most events happen in dimly lit gymnasiums, generally flash is not allowed and the girls move pretty quickly requiring a fast shutter speed. My standard equipment includes the same gear that I used for the entire football season: a 7D Mark II and 5D Mark III along with the EF 24-70mm f/2.8L II and the EF 70-200mm f/2.8L IS II USM lenses. I find that the 70-200mm lens is far too long for the facilities that I have shot in and the 24-70mm, although it is an f/2.8, is not quite fast enough to keep the ISO below 10000. For this shoot I took advantage of Adorama's generous return policy and tried out a new lens and pulled a rarely used lens from the bottom of my camera bag.

The first lens that I tried is the Canon EF 135mm f/2. Although this is a fairly old design, dating back to 1996, it gets solid reviews for sharpness and focus speed/accuracy so I figured it would be a decent candidate for cheer buying me a full stop and allowing me to cut the ISO in half. The second lens that I put through the paces is the Canon EF 85mm f/1.8 USM that generally sits at the bottom of my camera bag. I don't really have anything against this lens but I just have not found too many uses for it. This should be very similar to the 135mm lens and gives an extra 1/3 stop of light at the cost of reduced reach.

Below, on the left, is a sample shot with the 135mm lens at f/2, 1/800sec and ISO 12,800 with the 5D Mark III. The lens does focus quickly and accurately and I only had a few missed focus shots. The lighting was very poor so even with the full stop advantage from the aperture I still had to shoot at ISO 12,800. The noise from the 5D was noticeable in the full screen shot but after some minor noise reduction the final shot is decent. On the right is a similar shot with the 85mm lens at f/1.8, 1/800sec and also ISO 12,800. I think that the two lenses are very similar in sharpness and focus accuracy but I do like the look of the photo from the 135mm lens slightly better. This might have to do with the exposure and I could probably make the two shots to look a lot more similar.

Left: Canon EF 135mm f/2, Right: Canon EF 85mm f/1.8
In the end I decided to return the 135mm lens for two reasons. The first is that it was not significantly different from the 85mm lens and I could not justify spending $1,000 for it. The second reason is that Sigma announced their new 135mm f/1.8 lens on the same day that I purchased this lens and Canon followed suit announcing a $200 price drop on their lens. I could not have had worse timing. Based on the initial reviews of the Sigma lens, if I am going to have a 135mm lens in my bag that is likely the one I want. I also took several shots with the EF 70-200mm f/2.8 lens but with the lighting as bad as it was the ISO was 20,000 and the shots were not really usable. Links to the full galleries can be found at the top of the page.

Monday, November 14, 2016

Regional Cheer Championships - 2016.11.13

Photos to Download

This weekend brought a new challenge that I have been hoping to attempt - photographing a cheer leading competition. I have read a bunch on photography blogs about the unique challenges of photographing fast moving action in poorly lit gymnasiums and wanted to try my hand at it. I do not know one thing about cheer so I brought both 5D Mark III fitted with the EF 70-200mm f/2.8L IS II USM lens and the 7D Mark II fitted with the EF 24-70mm f/2.8L II lens. I found that the 70-200mm lens was far too long for the setting and I used the 24-70mm lens on the 7D for all of the shots. I was a little amazed at how fast things move and how many photos I took in a short period of time. I typically take 800-1000 photos during a 2 hour football game but took about 550 photos during a 150 second cheer performance. It's the first time I have had the camera slow down due to buffering (they were RAW after all). In the end, I would give myself a solid B, but I am a bit disappointed in the photos and it all comes down to user error. The first mistake that I made was not getting to the event early enough. I only saw one routine before the Algonquin team performed and I found myself shooting from behind the team. I guess I should have noticed the judges table at the other end of the gym but, hey, it was my first time at a cheer event. My second mistake was not getting the correct white balance set up ahead of time but I was able to fix this fairly easily because I shot in RAW. My biggest mistake was how I set up the exposure - I opted to shoot in manual mode wide open (f/2.8), the shutter speed set to 1/800th of a second and auto ISO. I forgot to limit the upper ISO so most of the shots were at ISO 10,000 and have a lot of noise. I should have dropped the shutter speed to around 1/250th of a second, timed my shots for the pauses and probably used my EF 85mm f/1.8 USM lens on the 5D. This lens might have been a bit too long but the wider aperture, which is about a full stop faster than the f/2.8, would have allowed a much lower ISO setting and the photos would have cleaned up much better. So, there is my little secret - the reason that there are no close ups is that the photos are too noisy and they would have looked like crap. I tried to clean up the photos in Lightroom by pushing the luminance and detail settings pretty hard but but had to back off in the end because you get a lot of visual artifacts in the dark regions when you go too far.

 If I get the chance to shoot a cheer event again I will do a much better job based on what I learned today. It would be interesting to hear feedback on what the most preferred shots are. I like action shots where you can see facial expression but I think that it is important to see all of the competitors in most of the shots. I think that it really takes two photographers to shoot a cheer event.

Algonquin vs Shrewsbury, Central Mass Div 2 Playoffs - 11.12.2016

Photos to Download

I changed the look of my blog posts due to a mishap with Lightroom this past week which I will detail in a blog post later. Suffice it say, I have a much better understanding of the file structure in Lightroom now and know that it is impossible to recover deleted files from an SSD hard drive. Sigh ... It was unusual to shoot one of the varsity games during the day so I used the 7D Mark II with the EF 70-200mm f/2.8L IS II USM lens to cover most of the game. I felt a little out of place on the sidelines this week because there were quite a number of professional photographers with some really nice gear shooting the game. Unfortunately, the game did not go the way we wanted it to but the Algonquin players never backed down until the very end and I was able to capture some nice action shots because of it. Posting the photos was also pretty quick because I did not have to do a lot of post production work.

Tuesday, November 8, 2016

Algonquin JV vs Wachusett, Playoffs - 2016.11.05

Photos to Download

Once again, the weather at today's game was nearly ideal for photography - it was not too sunny like the XC meets, it was not dark like the varsity games, it was just right. All of these photos were shot with the 7D Mark II fitted with the EF 70-200mm f/2.8L IS II USM in manual mode with the aperture set to f/2.8, a shutter speed of 1/1250 sec to 1/1600 sec and auto ISO. The ISO values were typically between 250 and 400. I did the typical post-production work in Lightroom but I did not really have to do too much other than crop and make small adjustments to the white balance, exposure and highlights/shadows.