Читать книгу The DSLR Filmmaker's Handbook - Andersson Barry - Страница 9

Until Nikon released the D90 in 2008, buyers had to choose between a digital still camera and a digital video camera. Filmmakers were using film or traditional video cameras for production. When the D90 and, quickly afterward, the Canon 5D Mark II were released, you finally had the ability to shoot digital stills and HD video on the same device. Since then, every major camera manufacturer has added DSLR cameras that can shoot video.

At the time, HD video on a still camera was controversial. A lot of photographers worried that improvements to the still camera would be limited because it seemed all the attention was being placed on the video side of the camera. Independent filmmakers took one look at the early footage and realized the vast potential of this new technology. HD video has been around since the 1990s but was practically available only on traditional video cameras. The design and function of traditional video cameras prevented a lot of the cinematic qualities that traditional film cameras provided.

DSLR cameras allowed filmmakers to easily and inexpensively use interchangeable lenses to craft the look of their film more like traditional filmmaking. These factors, along with an available shallow depth of field and low-light capabilities, were not available on most traditional video cameras. These issues, coupled with the price and quality of the video image, helped supersize the growth of the DSLR market.

As stated previously, since the launch of the Nikon D90 and the Canon 5D Mark II, manufacturers have released an endless string of DSLR cameras that shoot video. The still/HD video hybrid has become the norm for capturing video. You can now just compare models and find the right functions and price point for your project and start shooting.

Sensor Size

If you are not a photographer and not accustomed to dealing with sensor sizes, let’s put it in motion-picture film terms. Sensor size is a bit like choosing whether to shoot on 8 mm, 16 mm, Super 16 mm, 35 mm, Super 35 mm, or 70 mm film. Just as you would with motion-picture film stock, you choose your sensor size based on your budget, the depth of field, and the aesthetic look for your film. In general, the bigger the sensor, the more expensive the camera (just like 35 mm or 70 mm film); the smaller the sensor, the cheaper the camera. This is a generalization, because some higher-end cameras have smaller than full-frame sensors.

A full-frame sensor is approximately the same size as a single frame of 35 mm film from a traditional still film camera (Figure 1-1).

Any non-full-frame sensor is referred to as a crop sensor (Figure 1-2). These sensors vary in size but are smaller than a single frame of 35 mm film from a traditional still film camera.

Figure 1-1: A full-frame sensor and 35 mm still film are the same size; the sensor area is 36×24 mm, or 864 mm².

Figure 1-2: A crop sensor is smaller than 35 mm film. The Canon APS-C sensor area is 22.2×14.8 mm, or 329 mm².

The sensor size affects the “grain” in your image, the light sensitivity, and the depth of field aesthetic for that camera. At the time of this writing, there are two dominant sensor sizes: full-frame sensors and APS-C crop sensors.

Figure 1-3: Nikon APS-C (left) vs. Canon APS-C sensor (right). The Nikon sensor is also used by Pentax and Sony. Notice that the Canon APS-C sensor is slightly smaller than the Nikon APS-C sensor.

APS-C is currently in all non-full-frame Nikon cameras and the Canon 7D Mark II, EOS 70D, and Rebel T5i. To make things slightly more confusing, there is a slight difference between the Canon APS-C and Nikon APS-C sensors (Figure 1-3): specifically, the Nikon APS-C sensor (22.2×14.8 mm, or 329 mm²) is slightly larger than the Canon version (~23.6×15.7 mm, or about 370 mm²).

Figure 1-4: Field of view comparison between full-frame sensor (blue) and crop sensor (red)

If you are using a crop sensor, be aware of how this affects your lenses. When you’re shooting with lenses from traditional 35 mm film cameras, the field of view will not match up with the given focal length on the lens. This is due to the fact that the sensor is smaller than the area the lens would normally be filling when shooting with 35mm film or a full-frame sensor camera.

Some people say that the focal length will be changed when used on a crop-sensor camera, but that is not accurate. Standard still lenses were designed so that the field of view would cover the full frame of the 35 mm film (Figure 1-4). A crop sensor is smaller than a standard 35 mm film frame, and when a standard lens is used, the field of view is greater than what is captured on the sensor (Figure 1-5). This creates a magnification effect. For example, your 50 mm lens will have a narrower field of view. This does not in any way change the actual focal length of the lens, just how much of the area of view is captured (Figure 1-6).

Figure 1-5: Full-frame sensor captured with an 85 mm Zeiss lens

Figure 1-6: APS-C sensor in the same camera position with the same lens. Notice the magnification effect as compared to the full-frame image in the previous figure.

The various sizes of crop sensors have their own multiplication factor specific to that size of sensor; these are referred to as either the crop factor or the focal length multiplier. Specific crop factors range between 1.3 and 2 depending on the size of the sensor. The way the crop factor is determined is a simple division of the size of the sensor by a full frame. For example, a full-frame sensor is 36×24 mm, and a Canon APS-C sensor is 22.3×14.9 mm. Dividing 36 by 22.3, we get 1.614, which we round to 1.6. If you are using a standard 24 mm wide-angle lens on a 1.6 crop sensor, your field of view is more like what you get with a 38 mm lens than with a 24 mm lens. This can hurt you if you are shooting in a really tight location, because you may not be able to achieve a wide enough angle.

The APS-C crop sensor is almost identical in size to the standard 35 mm film that Hollywood uses. So, don’t get worried if you have a crop-sensor camera. Before you decide which camera you should buy, look at some footage from the cameras you are looking to shoot with and choose the one that best aesthetically matches the movie you want to make. Decide the speed of film (ISO on your camera) and the grain tolerance (sensor size), and choose as you would between standard film stock, Kodak Vision stock, and so on.

Full-frame sensors are, for a variety of reasons, the most desirable, and the Canon 5D Mark III and the Sony A7s are the two leading cameras in the DSLR space with full-frame sensors. The great part of the full-frame sensor is that traditional 35 mm film lenses retain their true focal length. If you have your trusty 35 mm or 50 mm lens (or any lens, for that matter), then there is no learning curve for what image you will get. It will look the same as when shooting still images.

A good thing to note when comparing a full-frame digital sensor to 35 mm still or motion-picture film is that a full-frame digital sensor is in fact larger than 35 mm film. In reality, a full-frame sensor is almost equivalent to a VistaVision frame (Figure 1-7).

Figure 1-7: VistaVision film frame (left) vs. 35 mm film frame (right)

VistaVision

VistaVision was created in 1954 at Paramount Pictures; 35 mm motion film stock is 24×36 mm, whereas the full-frame digital sensor is 36×24 mm.

In VistaVision, instead of recording an image horizontally from edge to edge of 35 mm motion-picture film, the image is recorded vertically, allowing a much larger area of the film stock to be used for each frame. The main benefit is a much higher-resolution image and the possibility of a much greater depth of field.

Because of the lack of speed of the film stock circa 1954, usually productions blasted the scenes with light to create a large depth of field and usually didn’t take advantage of the ability to have a narrow or shallow depth of field. Thus, if you watch VistaVision movies like Alfred Hitchcock’s North by Northwest, you won’t see a shallow depth of field because they lit everything with mega Hollywood lights. Because DSLR cameras are so sensitive, now for the first time filmmakers are able to shoot at narrow depths of field previously not seen on a mass scale.

Table 1-1 lists the dimensions of the most common DSLR sensors; Figure 1-8 compares those dimensions visually. Figure 1-9 compares various physical film sizes.

Table 1-1: Sensor dimensions

Figure 1-8: Comparisons of sensor size

The major benefits of using a camera with a full-frame sensor are that it is more light-sensitive, creates less noise in your image, and offers the ability for a narrow depth of field.

More Light Sensitivity The reason that a camera with a full-frame sensor has more light sensitivity is simple – there’s more space for light to hit the sensor and bigger pixels collect more light (photons). The full-frame sensor has more than double the area of the APS-C crop sensor. The bigger (fatter) pixels catch more of the light than the smaller sensors.

Less Noise By having the larger pixels to catch the light, the camera doesn’t have to amplify them in order to match the same ISO from a smaller sensor. Think of it as blowing up your image. The larger the image you begin with, the less noise in your final print. The larger the sensor you start with, the less noise in your final footage.

Depth of Field Most filmmakers were never happy with the look of video. When HD came into existence, it was touted for its clear and sharp image. Many filmmakers didn’t like the look because it didn’t look cinematic. That all changed with the release of the first DSLR cameras. The ability to have a shallow depth of field and the more natural color rendering of flesh tones made HD video desirable to many filmmakers who previously disliked the look of HD video.

Figure 1-9: Various motion-picture film sizes

Depth of field is what we unconsciously think of when we want something to look cinematic. Look at your favorite movies to see how much you see of the background in any given shot. You will see that many scenes have a shallow depth of field where the background is more or less out of focus. With a traditional home video camera, you always have a deep depth of field, and when you view your footage, you will find that most things are in focus almost as far back as you can see. This is because sensors in home camcorders are small and the lenses are not fast so they have a bigger aperture than a DSLR camera.

The larger the sensor, the more shallow the possible depth of field; the smaller the sensor, the deeper the depth of field will be. Also, shooting at lower f-stops will cause a shallow depth of field vs. a higher f-stop on any sensor size.

With DSLR cameras, you can now create films just like Hollywood does. As a matter of fact, DSLR cameras have now been used to shoot major Hollywood film theatrical releases, many leading TV shows, and even many images you see when watching sports. You can choose a lens and paint with light just as filmmakers have since the dawn of the movie industry. And that is why shooting on a DSLR camera is revolutionary. We are making movies that look just those we have been watching for as long as the medium has existed.

Frame Rates

Current DSLR cameras offer a range of frame rates depending on which camera you buy or rent. Let’s talk for a moment about the standard frame rates in both film and video production. We call these frames per second (fps). These are the most common, or standard, frame rates:

24 fps is the standard rate at which motion-picture film gets run through the camera. So, any movie that is shot on film that you see in the movie theater was shot at 24 fps and is the holy grail of the “film feel” of your footage.

25 fps is the standard in most of the world (outside the United States and Japan) for video broadcast. This is close to the “film look” and was widely sought after in video cameras in the United States for filmmakers looking to get away from the 30 fps look of U.S. video cameras.

29.97 fps is the standard for broadcast in the United States. Most people refer to this as 30 fps, but there is a huge difference between 30 fps and 29.97 fps when it comes to broadcasting or viewing your footage in traditional formats (that is, TV, DVD, VHS, and so on).

30 fps is the standard more or less for web video. On the Web, there are no rules for frame rate. The Canon 5D Mark II originally was able to shoot only 30 fps and was limited to web-only video or complicated transcoding that doesn’t always work without problems.

Slow motion would be any frame rate greater than 30 fps. The two most common frame rates on DSLR cameras are 50 fps and 60 fps. This means you are recording double the number of frames as you would at 25 fps or 30 fps, and you can (in post) play these shots back at half speed smoothly, giving you slow motion.

fps and p vs. i

Many times fps is not listed next to your frame rate. Instead, you will see the frame rate of 24, 30, and so on, and either the letter i or the letter p will follow it. For example, if you want to shoot at 24 fps, you can select 24p, which stands for “24 frames per second progressive.”

When you see 60p vs. 60i, the p stands for “progressive,” and i stands for “interlaced.” Interlaced video records every other line, whereas progressive records a full-frame image. For example, 60p means you are recording 60 full-frame images each second, whereas 60i means you record half an image each second, so you end with a total of 30 full frames.

You might hear a few other terms related to frame rates. For example, NTSC stands for “National Television System Committee” and is the analog television system used in North America, South America, South Korea, Taiwan, Japan, Burma, the Philippines, and some other Pacific islands. NTSC has been the standard for more than 50 years in broadcast media in the United States and represents a 4:3 aspect ratio (think of the standard TV image, that is, non-wide-screen models) and a frame rate of 30 fps (also 29.97 fps). Both 30 fps and 29.97 fps are referred to when talking about NTSC. Although 30 fps was the initial standard for NTSC in 1941, in 1953 with the introduction of color television, the committee required a slight reduction in frame rate down to 29.97 fps. This reduction was needed because of visible interference with the chrominance signal and the sound signals over the airwaves. (In June 2009, the United States transitioned from analog to digital transmissions, and the new standard is called ATSC, which includes the digital formats 16:9 and 1920×1200 resolutions.)

ISO Settings

What is ISO? ISO represents how sensitive the image sensor is to the amount of light available. The higher your ISO, the more sensitive the image sensor is, thus increasing your ability to get shots in low-light situations. By raising your ISO rating to shoot with less light, you will be adding more noise or video grain into your image. Just note that often when you change one setting you are not changing just one thing – likely you are affecting something else by your choice. This is why it is critical to know all the basics and how they all interact so you can make sure you understand and predict what the end result will look like in the video image.

ISO is mostly an issue when you are shooting in low-light situations. If you are outside on a bright, sunny day or if you light your scene, you will be able to stay with a lower ISO. Think of ISO as a tool to help you capture an image if you aren’t in ideal conditions or if you don’t have enough lights to illuminate the scene.

Which ISO settings are available will differ from camera to camera.

If you have a still-film background, you may be more familiar with this being referred to as the ASA rating. ISO is the digital-photo equivalent of a film stock ASA rating.

As the famous quote states, “You can’t have your cake and eat it too.” There is a trade-off to high ISO settings: noise. When you boost your sensor’s sensitivity by selecting a higher ISO, you are enabling the camera to record a fainter light signal. By enabling the camera to record a fainter light signal, you, at the same time, are allowing the camera to record the fainter noise signal. Noise is defined as any signal that is not attributed to the light from the subject you are shooting. Noise appears as colored pixels usually most visible in the shadows and dark areas of your footage.

The sensor in your camera is an analog device and as such will create some noise itself in capturing your footage. This, coupled with the increased ability of the sensor to capture the light signal and noise signal, creates the visible noise in your captured footage. If you ever shoot high ASA film stock, then you have grain instead of noise; in general, film grain is acceptable, whereas digital noise is considered bad.

Your sensor size and camera manufacturer determine the range of ISO settings available on any given camera without being affected by noise (at least noticeable noise). The signal-to-noise ratio (or the S/N ratio) is the amount of light (signal) captured in relation to the amount of noise captured. This is why, in general, the larger the sensor, the less noise present in your footage. The reason for this is the number of pixels and their density on the actual sensor. Look at various camera models, and you will see that the manufacturer has placed a rating of 8 megapixels (MP), 10 megapixels, 12.1 megapixels, and so on for the sensor of those cameras (a megapixel is 1 million pixels).

This can be a bit deceiving, though, because it’s possible for two cameras – one that has a crop sensor and the other a full-frame sensor – to have the same megapixel count. In this case, the manufacturer has crammed the same number of pixels on the smaller sensor as on the larger sensors. This causes the pixels to be much closer together and affects how much signal (light) can be captured through each pixel. For instance, consider an 18 MP Canon 7D (Canon APS-C sensor) and a 16 MP Canon 1D Mark IV APS-H. You might think that because the 7D has more megapixels, it would yield a better image. Actually, the 1D Mark IV, with only a 16 MP count, will yield a better (less noisy) image because the pixels are less densely packed on the larger sensor.

So, you cannot look just at the megapixel count; you also need to look at the sensor size and the density of the pixels in the camera you want to use. The best possible situation is a full-frame sensor with a high megapixel count. If you are looking at cameras that share the same sensor size but have different megapixel counts, you may want to rent both cameras and shoot some footage to test the noise signals and see which one gives you the better image for your project.