How To Take Better Pictures — My Photography 101
Dr Will Bateman of CCell Ltd recently asked me to give a presentation to his scientist colleagues on how to improve their photography. CCell is a groundbreaking company that grows coral in the Ocean with an ingenious technique that utilises the Ocean’s waves. For more information, visit them here. The following is a transcript of my presentation:
OVERVIEW. Hello and welcome to my Photography 101. Thank you, Will and company, for asking me to do a presentation. I hope it’ll help you attain better pictures. You’re a bunch of engineers, so this should be a doddle for you.
One question I get asked: “How come pros always get better pictures?” The answer is simple, anyone can spend money on lenses that let in a lot more light than consumer cameras, and you can also spend money on cameras that have higher CMOS sensitivity. That’s why pro photography looks way better than conventional consumer snaps.
So, it’s just one thing: the signal-to-noise-ratio is a lot higher in pro-equipment than in consumer electronics, although the latter is improving in its own way. There’s also practice. The more pictures you take and review, the better you’ll get. Henri Cartier-Bresson famously said, your first 10,000 pictures are your worst. In DSLR photography, I’d say your first 50,000 pictures are your worst. The 10,000 milestone probably still applies to smartphones. There are, indeed, some amazingly skilled smartphone photographers around now.
What camera should you get? As a general rule, Single Lens Reflexes are the best photography cameras. You can also get good pictures with pro non-DSLRs.
Their application is stealth. People see a regular non-DSLR and underestimate the quality of the output, so they relax more. These cameras are expensive, and I wouldn’t get them unless you were a professional street or travel photographer.
I would argue, never buy a bridge camera, the kind where the lens cannot be replaced. I think bridge cameras are gimmicky and get in the way of photography.
LENS. If you buy a DSLR, buy it without the kit lens. A kit lens is too slow; meaning its aperture doesn’t open wide to let in enough light.
Manufacturers sell cameras with kit lenses in order to satisfy consumer demand for zoom.
Zoom lenses, however, are almost always worse in quality than prime lenses. Primes provide a much better quality of light than zooms. This is because primes have fewer elements or glass in the way.
What lens should you get? If you have a cropped sensor camera, a 35mm lens is a good first buy. If you have a non-cropped sensor, a 50mm lens is a good buy. It’s worth bearing in mind that some distance is needed between you and the camera because these are primes. A good wide-angle is the 24mm lens. I’ve included a chart of different focal lengths in an email.
When shooting, always turn off Auto. The best and fastest way to learn how to take a good picture is to turn it off and to switch on Manual mode all of the time. You’ll notice a drop in quality in your pictures, but this will be temporary. The more you experiment with Manual the faster you’ll learn. Always switch off Auto and switch on Manual. Auto is NEVER used by professional photographers.
You can get some interesting results with Aperture or Shutter priority in some circumstances, but Aperture priority and Shutter priority won’t teach you anything. So, it’s better to nail Manual first.
Now, let’s get down to business.
EXPOSURE. What is Exposure? Good exposure is the delivery of the right amount of light to the film or the CMOS sensor without overexposing or underexposing the medium — film or CMOS. This is achieved with three variables. The first is shutter speed, the second is aperture size and the third is film or CMOS sensitivity.
It’s down to you to set the value of each variable. If you change one variable far enough, the other two should be compensated accordingly. Both film and CMOS have cut-offs. Your job is to get the highlights, the mid-tones, the shadows and the blacks all within the sensitivity bracket of this medium.
This sensitivity bracket, or bandwidth, is known as dynamic range.
It’s a good idea to think of one of these variables as your anchor. Do you want a low ISO? Then set a slow shutter speed and open the aperture until the needle’s in the middle. Do you want a shallow depth of field? Then increase the shutter speed and lower the ISO until the needle’s in the middle. Do you want to freeze action? Then it’s the same story. Increase the shutter speed and increase the ISO until the needle is in the middle. Do you want a fine crisp image? Then, lower the ISO, open the aperture and increase the shutter speed as much as possible — until the needle’s in the middle.
All these settings depend on one resource. The amount of light available. It’s a negotiation.
These three settings work together in what we call the exposure triangle.
In digital photography, you can check the exposure by looking at the screen on the back of the camera. This is colloquially known as chimping. Whilst chimping, it’s a good idea to look at the histogram, to make sure the luminosity curves haven’t been cut off. If they have, adjust settings and take again.
Now I’ll go through the three variables in more detail.
ISO. The first is the ISO number. In the days before digital, exposure was determined by the size of the silver halide crystals in the film itself. The larger the crystals, the more light intensity and less open shutter duration is needed to capture a focused image; the smaller the crystals, the more both are needed.
Here, there’s a correlation to how grainy or fine an image becomes and crystal size. The smaller the crystals, the finer and crisper the images become; the larger the crystals, the grainier and noisier the images become.
How is crystal size measured? Answer: with an ISO number. The lower the ISO number, the smaller the crystals, and the more time is needed to expose. The higher the ISO number, the bigger the crystals are, and the faster to expose they become.
There are two things to bear in mind when exposing a film or a CMOS sensor. They are the intensity of light and the duration of light.
By fast, I mean less light intensity and less duration needed to expose a film.
In the days of film photography, ISO 400 was the sweet spot for choosing a film that’s not too grainy or noisy and that’s fast enough in most conditions. ISO 100 was slow, ISO 800 was grainy and fast. ISO 100 is likely to be used in still-life photography and on bright sunny days; and ISO 800 is likely to be used in low-light environments where there’s a prohibition to flash photography or when shooting with a very long lens.
The same principles apply to digital photography, except the CMOS sensor is much more sensitive than film. On the other hand, film photography offers more contrast than digital, but that’s another story. Depending on camera, you can go up to 2000 or 3200 ISO, without immediately noticing any digital noise. My camera can go up to ISO 6400 and the graininess isn’t immediately obvious. Each camera will have an ISO noise profile you can test. It’s a good idea to become familiar with what looks good with your camera.
When you’re taking a picture, a low ISO is desirable, but you also have to strike a balance between a low ISO, an appropriate shutter speed and aperture size.
Make sure you have the correct camera setting for each environment. If the environment changes, change your camera settings. For example, if you go from outdoors to indoors, change accordingly — and preferably without anyone watching.
SHUTTER. Shutter speed. Shutter-speed-controls two things. The first is how frozen in time or blurry an image appears; and the second, the amount of time you’re exposing the medium — CMOS or film. The slowest shutter speed you can use without falling prey to camera shake is 1/60th of a second. Anything less than 1/60th, you will need to think about using a tripod, a monopod, a gorilla pod, or a brace.
1/50 of a second is magical because it’s the closest to how our eyesight interprets movement. Things look most natural with a 1/50 of a second blur.
In video, this corresponds to the 180° rule. Shutter speed in this context should always be double of the frame rate to achieve this effect.
The higher the denominator or bottom number, the faster the shutter; and the more your images will appear frozen in time. The more you push this, the more you’ll need to compensate by opening up the aperture and ramping up the ISO.
APERTURE. Consumer apertures fall at around f5.6 when wide open.
f5.6 is what comes with the kit lens, usually sold with DSLRs.
f5.6 is just too small of an opening, meaning it doesn’t let in enough intensity of light; so, accordingly, shutter speeds need to compensate with slowness and expensive ISO — resulting in a grainy image, or one that’s too dark.
In an interview, director Martin Scorsese was very clear about his biggest problem in the early days of his filmmaking. Slow lenses! He didn’t have access to fast cine lenses.
The lower the f stop, the more financially expensive the lens. Having said this, camera phones have a good low f stop number with f1.7 or f1.8.
Remember, the lower the f stop, the more light gets in.
As a sidebar, in my opinion, the pixel science in camera phones is too compressed and there’s no dynamic range — which is important to capture highlights, mid-tones, shadows and blacks, all at the same time.
In DSLR photography, you can get a 50mm Canon f1.8 with all of the benefits of intense short bursts of light for a bargain, but it has a noisy motor that sounds a little crass and may even frighten the horses.
An f1.4 isn’t noisy but it’s dearer. The disadvantage here is that its blades intrude on its circles of confusion, even when wide open.
An f1.2, on the other hand, doubles the amount of light that the f1.4 allows. It has an ultrasonic motor and the blades retract enough to avoid cropping its own circles of confusion when fully open, creating a beautiful aesthetic effect.
By having a fast lens like the f1.2 up to f2.8, you can have a good low ISO number and a sexily fast shutter speed. These wide-open primes are a vast improvement from the supplied kit zoom lens at f5.6
COLOUR BALANCE. Almost always, set the colour balance manually. For indoors, it’s either tungsten or white fluorescent or flash. For outdoors, it’s daylight setting unless it’s night time. Colour balance or white balance is measured with the Kelvin scale. 1,000K is blue or cool, 10,000K is red or hot. Daylight or white light is 5000K.
If you have time, you can stick a clear white piece of paper in front of the lens and press “Set”.
Incidentally, from time to time, you should also reset the Black Balance of your camera if it has that setting, but make sure you screw on the body cap tightly when you do that. I say do it in a pitch-black room to be extra safe.
You can use a ColorChecker chart to get the colours to be scientifically accurate in order to then instruct your photo editor in post with the pipette tool.
It’s a good idea not to fully trust your screen because different screens display differently. What might look good on your screen won’t look good on another screen. So, always use the luminosity histogram and colour scopes in your photo editor to grade; and display media on different monitors, if possible, to test. This is part of the colour mastering skill you would need to learn as a content creator. The same applies to video.
You can buy scientifically accurate monitors but they’re typically very expensive.
JPEG and RAW. What’s the difference between a JPEG and a RAW file?
RAW doesn’t set and bake in colour information in the files created on-the-fly. RAW files are larger than JPEGs, and more difficult to open, requiring more computer power. They’re worth it, though, as they allow more adjustments in post. I always shoot in RAW, but sometimes I set the camera to shoot in RAW and JPEG to save time. Check your camera settings for this option.
Having a fast card and with USB 3.1 or even faster, if you can afford it, really helps with transfer speeds.
A JPEG image is smaller in size but has the same resolution as its RAW counterpart. It’s easier to open, but there’s a lot less you can do with a JPEG in post than with a RAW in post.
In a RAW image, you can adjust luminance and colour. Sometimes, your camera will come with software that enables this post-processing.
To be sure, you would have to set values using your judgment, but with reference to the luminosity histograms and colour scopes.
In smartphone photography, there are apps that help you do this, often ones that come with your phone.
LIGHTING. Whilst colour is measured from blue to red with the Kelvin scale, light intensity is measured from low to high in lumens, lux, flux, or nits, depending on context.
Every environment will have an illuminance lux profile. When you’re on set, readings can be precisely determined with a luma-lux conversion calculator. The brighter the lumens, the higher the lux readings. With a handheld meter, you can also take readings which give you a set of appropriate f stop, ISO number and shutter speed combinations. Most light meters, though, are inside the camera. If you’re looking through a DSLR there’s a needle just underneath the viewfinder that needs to be in the middle for correct exposure. This is often emulated in smartphones but on the screen. It’s better, by the way, to slightly underexpose than to overexpose. If you overexpose an image, you can’t bring back the detail. In video, there are tools that help you avoid this — the most notable is the zebra. 🦓.
Bright light is a good thing in photography. It’s the reason why low f stops are sought after — they enable short bursts of light, which is ideal for the medium.
As a rule, the more light — the better. Having said this, you also need to know and consider the difference between hard light and soft light.
Hard light is created naturally by the midday Sun. This time of day is often avoided by professional photographers unless it’s cloudy out there. What hard light does is it leaves hard shadows where detail is lost inside those shadows. Hard shadow is often used in film noir as expressionist shadow to create villains or anti-heroes, but in general, is best avoided in most images.
To avoid hard light, avoid using small sources of light. A classic example of this is the flash. The source of light is small and therefore it creates a harsh light followed by hard shadows. This harsh light also creates specular reflections — in this instance, because the light is harsh, so are these specular reflections. If you’re using flash, always bounce the light off a large surface at an angle to diffuse; but make sure the surface you bounce off is balanced white. If you bounce off colour, then you’ll get a cast, changing the overall hue of the image. This is considered undesirable.
In-built flash is notorious for creating hard light, but it’s also known to flatten an image if it’s used as headlight. By headlight, I mean placing the flash on top of the camera. If you place the flash on-camera and take a picture, it’ll eliminate the sense of depth because of the way the shadows fall.
It’s much better to take the flashgun off the camera and to shoot it wirelessly from a different location than the camera body and at an angle to the subject. Whilst you’re at it, why not use a white umbrella or a lightbox to diffuse?
The same principle applies to continuous light.
The technique with the best effect in the business of diffusion is the bookend setup. By shining a bright light away from the subject and towards a reflective surface, you can then bounce the flux beam through a thin, white cloth suspended on a frame.
The light will reflect endlessly between surface and cloth and the result is a soft light with soft shadows, and interestingly, a wraparound effect. The more lumens a light source has, the better and softer the results.
A compact version of this is found in lightboxes. Lightboxes have two layers of fabric diffusion. Together with a reflective baffle inside the box, they soften light with wraparound over the face — making you look amazing.
Incidentally, this happens naturally every day in the Golden Hour of photography.
When buying a lightbox, don’t buy one with cheap fluorescent bulbs for three reasons. The first, the bulbs aren’t very strong, so you won’t get much benefit. The second, these lights emit a sickly green cast, which is highly unflattering; and third, if you’re shooting video, the flicker will sometimes interfere with the progressive scan of your camera. Cheap, weak LEDs have the same problem with sickly green.
You can correct the green cast with RGB curves in post, but not so much if you’re shooting with a compressed format.
In professional lighting setups, there’s the main light, known as the key light, there’s the fill light, and there’s the backlight. Ideally, you should have a bit of all three for a successful image. If you have the setup available, use dimmers to finely tune.
You can also use blue gels over hot tungstens to cool down the Kelvin to daylight.
DYNAMIC RANGE. What is dynamic range? Dynamic range gives you the ability to shoot bright light and shadows with gradations without crushing blacks or whitewashing highlights — all within the same image.
Film has a high dynamic range and good contrast. Compressed formats don’t have dynamic range and that’s why, often in consumables and in smartphones, if midtones are properly exposed, highlights are washed out and shadows are crushed into blacks.
The square inverse law becomes important in this context if an object in front of a bright light and close to the camera overexposes when the rest of the shot is properly exposed. It’s the interplay between lumens, lux, flux versus dynamic range. That’s when the luma-lux conversion calculator and live monitor become handy.
That’s it for now. Thanks for watching, now time for Q&A.
