UNDERSTANDING PHOTOGRAPHY - ScoobyNet.com - Subaru Enthusiast Forum



Photography Hints, Tips and general help in the art of photography

UNDERSTANDING PHOTOGRAPHY

Closed Thread
 
 
 
Thread Tools Search this Thread
Old 13 September 2006, 11:18 PM   #1
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default UNDERSTANDING PHOTOGRAPHY

UNDERSTANDING PHOTOGRAPHY
ScoobyNet Beginner’s Guide to using Digital SLRs



Introduction

SLRs - what’s the big deal?

Setting the exposure

What is correct exposure?

Photo jargon explained

Lens Focal Length

Focal Length and Image Format

Lens numbers – what do they mean?

Depth-of-Field

Lenses and Perspective

Filters and lens attachments

Understanding the shutter

Creative use of shutter speeds

Camera-shake – avoid it!

Automatic exposure modes

Flash in daylight

Flash at night

Pixels – how many do you need?

JPEG and Raw images

Composition

Getting started is easy

Answers to the exposure question in post #3


Introduction
Many people new to the art and craft of photography find some of the terminology confusing and the theory complicated. But actually, once you understand the ground rules – and they’re mainly quite logical – it is pretty straightforward. And then, when you’re in control and start getting creative, you will see what all the fuss and the fun is about. Here is a guide to knowing the basics.

Scroll down to the post # for each chapter. Enjoy

Last edited by Scoobychick; 04 January 2007 at 08:43 AM.
Hoppy is offline  
Old 13 September 2006, 11:21 PM   #2
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

SLRs – what’s the big deal?
SLR stands for Single-Lens Reflex and this is the design of camera used by all keen photographers for decades. Most other types of camera have two lenses - one viewfinder lens for framing the picture, and a second lens that actually records the image. These two lenses show much the same thing most of the time, but with an SLR, you view and frame the picture by looking through the same lens that records the image. In this way you can see exactly what the camera will record. That’s what the single-lens bit means.

If you want to take more than casual snaps, an SLR gives you total control and it also allows you to fit different lenses like long telephoto zooms, ultra-wide-angle lenses, or macro lenses for super-quality close-ups. And in all these situations, what you see in the viewfinder is what you’ll get in the final image. This is not possible with most other kinds of camera, and there are dozens of different lenses made to fit SLRs, not to mention powerful flash guns, and lots more clever bits and bobs that give unrivalled versatility.

In order to make this single-lens bit work, SLRs have a reflecting mirror and prism arrangement for viewing (that’s the Reflex part) and the instant before you take the picture the mirror flips up out of the way, allowing light to pass directly on to the digital sensor. This happens very quickly, in about 1/10sec or less, so is almost instantaneous. When the mirror is up, it blacks out the viewfinder while the shutter opens and the picture is recorded, then the mirror flips back down again for continued viewing. The whole operation happens in the blink of an eye.

All SLR cameras work like this, whether using film or digital, and the relatively new breed of ‘SLR-style’ digital cameras are different in that they do not have a reflex viewfinder, and some do not have a secondary optical viewfinder either – just an LCD screen. They are also known as ‘bridge’ cameras because they bridge the gap between more basic compact cameras and true SLRs. The LCD screen is fed an image by the same sensor that records the final picture. It is a clever adaptation of digital technology but unlikely to lure enthusiasts away from their beloved true SLRs where their tremendous versatility and superb image quality remain unparalleled.
Hoppy is offline  
Old 13 September 2006, 11:23 PM   #3
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Setting the Exposure
All digital sensors, like film, need a certain amount of light to give their best – not too much, and not too little. This is controlled by two key functions: the camera shutter which controls the duration of time the sensor is exposed to light, and by the lens aperture which controls the brightness of that light.

For a given amount of light, if you double the duration of the shutter time, but also halve the size of the lens aperture, the effective amount of light reaching the sensor - the exposure - remains the same. Correct exposure is obtained by juggling these two settings, which can be used in a great variety of combinations. Both settings also have a big influence on how other aspects of the final picture will look, so understanding exposure gives you creative control at the same time.

Shutter speeds (durations of time) are marked in fractions of a second, eg 125 means 1/125sec, 250 is 1/250sec, 500 is 1/500sec and so on. It’s easy to see here that each speed is either half, or double, the one either side. Cameras can also give intermediate speeds, but the halving/doubling principle is fundamental to all aspects of exposure control. Remember that.

These shutter speeds might seem incredibly brief, but this is what you need to control movement. To freeze fast action, you will often need 1/500sec or less, but you can also create intentional blurring by using a much slower speed, like 1/15sec or longer. You also have to balance this with your ability to hold the camera steady and avoid accidental blurring caused by camera-shake, or maybe use a tripod for steady support.

The lens aperture is an adjustable hole that controls the brightness of the light, by varying the size of the hole through which light passes to the sensor. The larger the hole, the brighter the light becomes, and this is set by adjusting the lens aperture, marked with an ‘f’ number. Now this bit can get confusing, but stick with it and the mists will be lifted. Large, bright, lens apertures are given a low f/number, like f/2.8 or f/4, and small, dark, lens apertures have a higher f/number, like f/11 or f/16.

Have you remembered that all exposure controls are linked by that halving/doubling factor? Well, believe it or not, so are these f/numbers. If you multiply, or divide, each f/number by the square root of 2, which is 1.4, you’ll discover that f/2.8 is twice as large as f/4 (2.8 x 1.4 = 4) which is twice as large as f/5.6, and so on. The full range of f/numbers you’re likely to see is f/1.4, f/2, f/2.8, f/4, f/5.6, f/8, f/11, f/16 and f/22 and each setting has either half or double the light passing area of the one either side. All lenses can close down their apertures to the smaller and darker settings of f/16 or f/22, but very few can open as big and bright as f/1.4 or f/2. These lenses are both highly prized and highly priced. Your camera will also let you set intermediate lens apertures, like f/4.5 or f/5, which dovetail with intermediate shutter speeds to give you the best result, so it’s a neat mix and match.

A given f/number is always constant and directly comparable in the amount of light it allows to pass, regardless of lens, camera, or sensor size. This applies to everything from a camera-phone, to zoom lenses, or to those massive super-long professional telephoto lenses. An f/number is already adjusted for these variables because it’s a ratio between the focal length of the lens and diameter of the aperture. So f/numbers are always equal in terms of exposure control, but depending on the focal length of the lens you’re using, different f/numbers have a major role to play in deciding which parts of the picture you want to be sharp, and which parts blurred – called the depth-of-field - which we’ll come to later.

Decisions, decisions! Are you following this? Making different choices is at the heart of good photography. This is crucial stuff, so there’s a little test to follow.

Finally, with digital SLRs, you can also control the sensor’s sensitivity to light by adjusting the ISO number (International Standards Organisation). Here we thankfully return to logical numbers, and the halving/doubling principle is obvious: ISO 100 needs twice as much light as ISO 200, which needs twice as much as ISO 400 and so on. But the really good news is that with digital you can increase the ISO setting at any time, and while there is some loss of detail in the darker parts of the image (called noise) it is usually hardly noticeable and picture quality remains very good. With film, not only are you stuck with the ISO rating the particular film has, but image quality suffers badly with higher ISO films. With digital, the opportunity to change ISO at will is a photographer’s dream come true. Make the most of it.

Okay, see if you’ve got it with this test. You’re shooting motor sport with your new digital SLR and the action is hot. Your camera says 1/125sec at f/8, set at ISO 100, but you want to freeze the action with a shutter speed of 1/500sec. What adjustments can you make to maintain correct exposure, but using a shutter speed of 1/500sec? The answers are at the end in post #22.
Hoppy is offline  
Old 13 September 2006, 11:25 PM   #4
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

What is correct exposure?
Whole books have been written about this! But times have moved on and modern technology is very clever and your camera can measure and interpret complex situations extremely quickly and accurately. So the simple answer to correct exposure is, if it looks right to you, then it is right. Check the screen of your D-SLR. Looks okay? Take a glance at the histogram graph of light levels (see your camera handbook) and that will tell you if all important parts of the image are within the dynamic range (the range from brightest white to darkest black) of the sensor. If that’s okay too, and histograms are very easy to understand, then no worries.

However, if the image does look too light or too dark, then you need to adjust the exposure with the compensation control. If it’s too light, then that’s over- exposure and you need to reduce the amount of light reaching the sensor. You must dial in some minus compensation and take another test shot. If the image is too dark, that’s under-exposure and you need to increase the light with plus compensation. With a D-SLR, taking a test snap is quick, easy and decisive, so use this facility. You new guys don’t know how lucky you are.
Hoppy is offline  
Old 13 September 2006, 11:26 PM   #5
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Photo jargon explained
What is a ‘stop’? What is a ‘fast’ lens, what is a ‘fast’ or a ‘high’ shutter speed, or indeed a ‘slow’ shutter speed? What is a sensor’s ISO ‘speed’? This is all photographer’s slang and sometimes the meaning is clear and logical, and sometimes it’s, well, strictly speaking wrong but kind of makes sense.

A lens stop is its f/number and goes way back to the dawn of photography when the light entering a lens was controlled by a piece of wood with a hole in it. A smaller hole would ‘stop’ more light, and each hole was half or double the area of the next size, as with our f/numbers now. That’s where the term originated but today a stop has become shorthand for a halving or doubling of anything exposure-wise and is also often applied to shutter speeds as in ‘go a stop faster’ which more accurately translates to ‘halve the shutter duration’ but you get the drift.

A fast lens is one with a large, bright maximum aperture (low f/number) such as f/2 or f/2.8. It’s fast because it allows you to use a fast shutter speed, which is slang for a short-duration shutter speed like 1/500sec or 1/1000sec that will help to freeze fast action. Likewise, fast film or fast ISO settings refer to high sensitivity to light, which also allows you to use shorter duration shutter speeds. And in all these situations, the term slow may also be heard – it means (pleasant surprise) the opposite of fast.

From these basic explanations, you will appreciate that photographers are generally obsessed with how bright the light is – always have been, and always will be. The fundamental issue is that the contest between what you see with your eye and brain, and what your camera can record, is massively one-sided and the camera loses every time. Almost always, you can’t have too much light. On the rare occasions when this might be a minor inconvenience, it is easy to control. But if you haven’t got enough light, which is very common, it is more difficult. Man might have invented the flashgun, but it’s a poor substitute for that big bright thing in the sky.

Here are a few more terms you’ll come across, in no particular order. Many others are explained elsewhere.

Burnt-out or blown-highlights means that part of the image is over-exposed, often the sky. You can control this by adjusting the exposure, and maybe use a polarising filter, or a graduated filter – see later.

Infinity is the furthest focusing distance of a lens, the point at which everything at that distance and beyond is perfectly sharp.

Bokeh is a Japanese word that roughly translates to ‘fuzzy’. In photography, bokeh refers to the out-of-focus areas of the image, and whether these areas are attractive and enhance the main subject, or are unattractive and distracting.

Hot-shoe is the slot on top of an SLR’s viewfinder where you can fit accessories such as a powerful flash gun. It is ‘hot’ because of the electronic contacts that maintain communication with the camera.
Hoppy is offline  
Old 13 September 2006, 11:27 PM   #6
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Lens Focal Length
The focal length of a lens is the distance from the optical centre (usually not the physical centre) to the image sensor. This distance also affects magnification of the image it will produce. Long focal length lenses like 200mm or 400mm are called telephotos and will make distant subjects appear larger, or closer. Good for sport, nature, and other subjects where you can’t get near enough.

At the other extreme, short focal length lenses like 15mm or 20mm are called wide-angles because they take in a wide view, which is good for landscapes, or cramped situations where you need to include everything in the picture.

In-between these focal lengths are standard lenses, which will often come already fitted to the camera as part of a kit, and they are good general- purpose lenses. You can have a go at most things with these, and they will handle popular subjects like portraits and close ups quite well.

Lenses that have just one fixed focal length are called ‘prime’ lenses, but of course, most popular lenses today are zooms. A zoom lens varies its focal length and most fall into the broad categories mentioned above, such as tele-zooms, wide-angle zooms or standard zooms. They will typically have a zoom range of around 3x, which means the longest focal length is about three times the short focal length, eg 70-210mm (70 x 3 = 210). Generally, the lower the zoom range of a lens, the better the image quality will be, and the maximum aperture might also be larger and brighter, but that’s not always the case as cost plays a big part. Some lenses cost much more than the camera.

There are also zooms with a focal length range that goes all the way from wide-angle to telephoto in one neat package, sometimes called super-zooms with a zoom ratio of maybe 10x, which is extreme. They are very handy of course, but some corners have to be cut in their optical design so image quality is often lower than less ambitious lenses. They will still beat any lens fitted to a non-SLR, but compared to the best SLR lenses their image sharpness will be slightly lower especially at the edges, there will be some vignetting which means corners of the picture will be slightly darker, and there will be some distortion where straight lines near the edges of the picture are slightly curved, such as the walls of a building. You may also notice something called chromatic aberration (CA) particularly at the wide-angle end. This shows up as colour fringing, often pinkish or green, and most obviously towards the edges of the picture. Finally, the maximum aperture of these lenses is always relatively small and dark, particularly at the telephoto end, which is the opposite of what you ideally want.

Even the best lenses suffer at least some of these minor problems, but many image quality issues can be largely eliminated by post-processing with your PC using image editing software, if you have a little time and skill.
Some people love the convenience of super-zooms and use almost nothing else, while others hate them and will have a heavy bag full of separate prime lenses, each one tailored to a particular purpose. It depends how picky you are, not to mention the cost, but photographers are always searching for better lenses in an endless quest for perfect pictures. Most will have at least two or three lenses, and lust for more. It is easy to fall in love with these beautiful toys, and lose sight of what you want them for in the first place. A useful catchphrase to is ‘zoom with your feet’ which means that very often just getting yourself into the best shooting position will produce much better pictures than any costly lenses.

Before moving on, there is one more type of lens to mention – the macro. Macro lenses should not to be confused with the ‘macro’ setting on many zooms, which is really just marketing-speak for close-up. A true macro allows you to take extreme close-ups of very high quality. With these lenses the minimum focusing distance is not the best way to compare them so their macro capability is quoted as a magnification ratio, such as 1:4 (quarter life-size), 1:2 (half life-size) or 1:1 (life-size). A life-size ratio of 1:1 means that if you are shooting a bumblebee maybe 20mm long, the image on the sensor will also be 20mm long. Imagine that when enlarged to a 10in print – the bee is now the size of a pigeon and the detail is just amazing. Macro lenses can also be used for general picture taking and are popular for (human) portraits.

Choose your lenses carefully and spend as much as you can afford. You will be using your favourite lenses for a long time – long after your new camera has been superseded and replaced.
Hoppy is offline  
Old 13 September 2006, 11:28 PM   #7
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Focal Length and Image Format
While the focal length of a lens affects image size, this only becomes really meaningful when it is coupled directly to an image format (sensor size). This is when you can measure the angle-of-view (sometimes called field-of-view). That is how much, or how little, you’re getting in the picture and it is only then that you will begin to see what a lens is capable of. With the introduction of digital cameras, using a variety of different sized sensors, this can get a bit complicated, but it needs to be understood before you can make sound choices.

Pre-digital when we all used film, popular cameras had a 35mm format. This measure refers to the overall width of the film, including the sprocket holes and it has nothing at all to do with focal length, despite the confusingly similar terminology. The actual size of the film image frame is 24 x 36mm, and this is referred to as Full-Frame 35mm (FF-35mm). Most D-SLR cameras use an image sensor that is less than half this size, often called APS or APS-C, which is another confusing reference to an old film format of similar size (the improbably named Advanced Photo System - Classic). With their smaller sensors, D-SLRs need shorter focal length lenses to produce the same angle-of-view. To help you understand what you’re getting in the picture, you will see the term ‘35mm equivalent’ used as a reference to focal length. For example, a 17-85mm lens for a D-SLR might also be described as a ‘28-135mm equivalent’. This means that the lens has the same angle-of-view as a 28-135mm lens on a FF-35mm camera.

Manufacturers make it easy to do this conversion by quoting a crop-factor for their D-SLRs, which is related to how much their image sensors have been cropped (reduced in size) compared to FF-35mm, and it allows you to easily convert focal lengths into their 35mm equivalents and make comparisons. One popular range of consumer D-SLRs has an image sensor that is 22.2 x 14.8mm and you need to multiply the actual focal length of a lens by the crop factor of 1.6x to calculate the FF-35mm equivalent focal length. Most camera manufacturers use a similarly sized sensor so their crop-factors are also similar.

There are also a few FF-35mm D-SLRs around, which of course do not require any adjustment for focal length. The large image sensor also delivers fantastic picture quality even in the biggest enlargements, but they are also very expensive and intended for professionals or well-heeled amateurs. The new smaller sensors of D-SLRs aimed at the consumer market are much cheaper, they are here to stay and are getting better all the time. Manufacturers are also producing new lenses specifically for these cameras, which are optimised for the smaller image sensor, and this is a sure sign of commitment to the future.
Hoppy is offline  
Old 13 September 2006, 11:30 PM   #8
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Lens numbers – what do they mean?
As an example, take the 17-85mm lens mentioned previously. Around the front of the lens it reads 17-85mm 1:4.5-5.6. The focal length range is 17-85mm and the maximum aperture is f/4.5 at 17mm, and this reduces slightly to f/5.6 when the lens is zoomed out to 85mm. Only the lens’s maximum aperture (the largest/brightest) is ever quoted and this is always preceded by the number 1. This is another ratio, but since every lens always quotes the same value of 1, you can ignore it.

You will also see other initials such as USM, IS, VR, EF-S and many others. These all refer to manufacturers’ brand specific features for different focusing mechanisms, anti-shake control, camera format and so on.
Hoppy is offline  
Old 13 September 2006, 11:31 PM   #9
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Depth-of-field
Depth-of-field is one of the key aspects of an image that can be controlled to great creative effect. It is also one of those photographic techniques surrounded in mystery.

Depth-of-field is the zone of sharpness within a picture – the area that is in clear focus from near to the camera, towards the distance. This zone of sharpness extends a certain distance in front of the point you focus on, and a certain distance behind it. In a wide-angle landscape picture, it can extend right from your feet to the horizon, or in a close-up of a butterfly it can be just an a few millimetres of exquisite detail with everything else a complete blur. You choose.

Imagine a good portrait, with head and shoulders turned slightly towards the camera. Sharp focus is on the eyes, but closer to the camera the near shoulder will be slightly out of focus, and behind the subject, the background will be completely out of focus. The photographer has deliberately chosen to use shallow depth-of-field, creating a zone of sharpness only about six inches deep – that’s the measure of the depth-of-field in this case - with the main point of focus on the eyes to concentrate your attention on the face, and everything else nicely blurred to reduce distractions in the foreground and background. (The term depth-of-field is often confused with depth-of-focus which is actually quite different, but don’t worry about it.)

There are just two factors that control depth-of-field - the size of the lens aperture (f/number), and the size of, or magnification of, the image. Let’s look at each factor, starting with the easy one.

Large, bright, lens apertures (low f/numbers) like f/2 or f/2.8, give shallow depth-of-field, and small, dark, lens apertures (high f/numbers) like f/11 or f/16, give much deeper depth-of-field. Selecting the f/number is the main control used to determine depth-of-field when setting the exposure and composing the picture. So, lenses with large apertures not only give you more options when it comes to setting the exposure, they can also be used to reduce depth-of-field. As in the example above, large aperture lenses are good for portraiture, and because they can simultaneously allow a fast shutter speed, they are also good for action photography, or for taking pictures in low light. Not only that, but you get a brighter viewfinder image for easier focusing and framing of the picture. No wonder that fast lenses are so sought after by photographers - what’s the catch? They are usually big and heavy, and can be very expensive.

Image size is the other factor that determines depth-of-field, and the bigger the subject appears in the picture, so the depth-of-field reduces. But there is a slight complication in that there are two ways of varying the image size. The most obvious one is that if you move closer to the subject, the image gets bigger. Alternatively, if you fit a longer focal length lens such as a tele-zoom, this will magnify the image making it bigger and so this also reduces depth-of- field.

It is often said that telephoto lenses always reduce depth-of-field but that’s not strictly true - they only do this by increasing the image size. Let’s take our portrait again as an example. If you frame your subject for a head and shoulders picture with a standard focal length lens, at a given f/number you will have a certain amount of depth-of-field. If you then fit a telephoto lens, but also move back so that once again your subject appears the same size in the viewfinder, then at that same f/number the depth-of-field will also be the same.

So it is untrue to say that telephoto lenses always produce less depth-of-field, for the reason given above, but there is a little more to consider here in order to get a full understanding. In some circumstances, telephoto lenses can appear to give less depth-of-field. This is because telephoto lenses have a narrow angle-of-view, and therefore when you move back and shoot from a more distant position, there is less background appearing in the picture behind the main subject. So the angle-of-view is reduced and this usually gives the impression that depth-of-field is also reduced and although it is actually an illusion, it is one that can be used to great pictorial effect.

By the same token, the reverse of this phenomenon applies to wide-angle lenses that are often said to give greater depth-of-field. They don’t, but from a fixed camera position they will make the image smaller, and that does increase depth-of-field. Likewise, the wide angle-of-view will include more background and while this doesn’t actually increase depth-of-field, it often looks that way.

Finally, with the introduction of digital cameras, there is now another aspect of image magnification to consider – the size of the sensor. We’ll not dwell on this as there are a lot of aspects to it, but one is very important, and that is the smaller the sensor, the greater the depth-of-field will be. Basically, a D-SLR with a sensor crop-factor of 1.5x or 1.6x will produce images with significantly greater depth-of-field than FF-35mm SLRs. The difference is about 1.25 stops, so that f/5.6 on a 1.6x crop SLR will deliver approximately the same depth-of-field as f/9 on FF-35mm. This can be an advantage or a disadvantage, depending on the type of photography you do. And of course, the effective focal length also changes by the crop-factor so you have to be careful with comparisons.

That’s a lot to take in, so here’s a quick reminder. Large, bright, lens apertures (low f/number) and large image size (move closer, use a telephoto) reduce depth-of-field. And cameras with an image sensor smaller than FF-35mm have greater depth-of-field. If you want to know exactly what your depth-of-field is going to be, most SLRs have a button that closes the lens down to the picture taking f/number you’ve selected so that you can see for yourself. Unfortunately, this often makes the image too dark to see easily so you can buy a simple calculator that will tell you the exact distances, or even more simply, download one of the many free DIY versions from the internet – just type ‘depth of field calculator’ into any search engine.
Hoppy is offline  
Old 13 September 2006, 11:32 PM   #10
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Lenses and Perspective
Another creative application of different focal length lenses is in the control of perspective – that is the relationship in size between objects close to the camera and those further away.

With a wide-angle lens, you can move in close while still including everything in the picture, and the perspective will be exaggerated. Objects close to the camera can be made to appear very large, while those just a little further away are dramatically smaller. These images can have a powerfully dynamic 3-dimensional effect that leaps out at you.

At the other end of the scale, long focal length telephoto lenses allow you to shoot from a greater distance, yet still fill the frame with your subject. Due to the distant viewpoint, all objects are at a relatively similar distance from the camera, so they appear at relatively the same size. The perspective is flattened. A common example is at the start of a Formula One race, with the camera looking towards the cars from the far end of the straight. The cars look as if they are almost on top of each other because they all look about the same size, and when there’s a pile-up you can’t see what has happened without a side view or aerial shot. That’s the compressed perspective effect of shooting from a long distance with a telephoto lens.

And in between these two extremes are standard focal length lenses, with an angle of view between 40-50 degrees, that usually (but not always) produce normal-looking perspective. Photos tend to look natural, generally as they appear to the naked eye (but comparisons between eyes and lenses are difficult because we see things in a completely different way to the camera).

In all these situations and with different focal length lenses, it is important to understand that the lens is only partly responsible for the pictorial effect. The change in perspective is directly the result of the camera-to-subject distance, not the lens. It is just that a wide-angle lens will allow you move in close and yet still get everything in the picture, while a telephoto lens lets you shoot from a distance but still fill the frame.

It is easy to prove this for yourself by taking two pictures from the same position, one with a wide-angle lens and another with a telephoto. Print the telephoto image, and then enlarge the centre of the wide-angle picture to show the same view as the tele lens. Okay, the wide-angle print will probably be rather poor quality, but if you check the perspective it will be identical in both pictures.
Hoppy is offline  
Old 13 September 2006, 11:33 PM   #11
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Filters and lens attachments
When you buy a camera, the first thing the salesperson will ask is would you like an ultra-violet filter (UV)? This screws on to the front of the lens, and is one of many different attachments you can fit. Ultra-violet light is not a problem at all these days and the filter is only to protect the lens, so some people say yes please, and others say no.

Lens protection is worthwhile in certain situations and if the filter gets damaged, you throw it away and not your expensive lens. Near the sea, salt spray tends to get everywhere, and if it is windy or dusty, physical protection is obviously valuable. Take care when photographing young children with sticky fingers or pets with inquisitive noses. If you’re into rallying, a face full of gravel could be very costly, as well as painful.

But if you want the very finest image quality, the answer is not so easy because a filter can sometimes impair it. The problem is flare or glare, as there is always some reflection of light from any air-glass surface. You will be aware of coloured spots that are clearly visible when shooting into the sun, which is an extreme example. But there is always some non-image-forming light bouncing around inside the lens that you can’t see, but has the effect of reducing contrast in your pictures.

A filter can make this worse, but if you use only the best multi-coated filters, you are unlikely to get problems. Buy a cheap filter, and sooner or later, you probably will. With or without a filter, keeping your lenses absolutely clean is vital. The odd spec of dirt will do no harm and is very hard to avoid, but check your lenses regularly and remove specs with a soft brush or a blower. Fingerprints are bad as they are oily and can smear, and that will cause flare for sure. Use a micro-fibre lens cleaning cloth (from photo dealers or opticians).

Whether to fit a lens hood is a much easier decision – get one and use it all the time. Hoods shade your lens, just as you shade your eyes with your hand in bright light, reducing the possibility of flare washing out the image. Lens hoods also give good physical protection. No down-sides there, and low-cost insurance.

A polarising filter is also a good idea, especially for landscapes where they darken blue sky and reduce reflections, which increases contrast and brightens colours. Sometimes a polarising filter makes a dramatic difference, and sometimes hardly any difference at all – it depends on the amount of polarised light around. If it is dull and overcast, polarised light from the sun is diffused and the filter won’t do much, but on a bright day with a clear sky, a polariser can make the sky turn a wonderful tone of blue. You can see the effect through the viewfinder of course, and you must rotate the filter to get the best effect. If your lens has a front section that rotates as you focus, using a polariser can be a frustrating business. You must also use a ‘circular’ polarising filter with D-SLRs and this refers to the optical construction of the special glass used and has nothing to do with the physical shape of the filter itself.

Another popular filter for landscapes is a graduated filter, often called a ‘grad’. These are dark at the top, fading to clear at the bottom, and are good for darkening and colouring dull skies. A neutral grey is very useful, and blue or orange-brown colours are also popular. The strength of the effect can be varied by moving the filter up or down, or adjusted with the lens focal length and aperture settings. With wide-angle lenses and small lens apertures (high f/number) the effect can be dramatic but again, if the front of the lens rotates while focusing, it can be a bit of a fiddle.

Close-up filters are another kind of useful attachment, although technically they are not filters at all, but accessory lenses. They are very simple lenses (basically similar to your gran’s reading glasses) and available in different strengths at very low cost. They allow your normal camera lens to focus much closer and are good for subjects like flowers. If you don’t expect stunning image quality you may be pleasantly surprised at how good, cheap and easy to use they are.

There are many more different types of filter – soft-focus, soft-spot, vignettes, colour-correction, starburst, multi-image prisms and many more. There are also commercial filter systems available, that allow you to use multiple filters in endless permutations. However, with digital and the potential that post-processing software offers, these special effects are often better applied with a computer where, if you don’t like it, you can modify or remove the effect.

While filters are fitted to the front of the lens, you can also fit other accessories between the lens and the camera body, notably extension tubes, and tele-converters.

Extension tubes allow you to focus much closer, by increasing the distance between the lens and the camera. This is what happens inside the lens when you focus normally on closer subjects – the tubes just take it a big step further. While you can get very big close-ups by using extension tubes, they are not a true substitute for a macro lens because they are working with a lens that has been optimised to work at longer focusing distances. Unlike close-up lenses, extension tubes need electrical contacts that maintain communication with the camera, so they are more expensive.

Tele-converters, also called extenders, are another optical accessory that fits between the camera and the lens. They ‘extend’ the focal length of the lens, increasing it by 50% (1.4x) or 100% (2x), for a magnified telephoto effect. Your 200mm lens suddenly becomes a 400mm lens with a 2x tele-converter. Wow! There must be a catch? Yes, but sometimes it’s worth it.

Tele-converters magnify the centre of the lens’ image, by either 1.4x or 2x. In doing so, they magnify any lack of sharpness in the lens by a corresponding amount and they also reduce the maximum aperture by either one stop (1.4x) or two stops (2x). For example, when fitted to a 200mm f/4 lens, this is changed to a 300mm f/5.6 (1.4x teleconverter) or 400mm f/8 (2x). This is just the laws of physics at work, not a fault.

If you fit a high-quality tele-converter with a high-quality lens that has been designed from the start to be used with a tele-converter, then the trade-off is worth it. Get it right, and tele-converters are great - get it wrong and they’re a waste of money. Broadly speaking, 1.4x tele-converters are better than 2x, and they work best with long focal length lenses, not wide-angles. There’s no free lunch here, but with the right combination, it can be a tasty one.
Hoppy is offline  
Old 13 September 2006, 11:34 PM   #12
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Understanding the shutter
SLRs use a focal-plane shutter to control the duration of the exposure, so-called because it sits behind the mirror and immediately in front of the image sensor’s focal-plane. It is one of the few hard working mechanical components of a modern camera, and is capable of making tens of thousands of exposures, as briefly as 1/8000sec, and sometimes at a rate of over 5fps (frames, or photos, per second).

Yet the secret of this amazing piece of machinery lies in its ingenious simplicity, rather than great technical complexity. It actually operates at only one, fixed speed, resulting in a finely tuned and uncompromised mechanism that is both accurate and reliable. The fixed speed is usually around 1/125sec (it will be the speed marked for flash synchronisation – x-sync - on the camera) and it is only at this speed, or longer, that the entire surface of the image sensor will be uncovered at any one moment, and the much shorter exposure times that it can produce are something of a trick.

Here’s how it works. There are two metal blinds that cover the image sensor. They used to be made of cloth, and are still referred to as curtains. When the shutter release is pressed, powerful springs fire the first curtain and it travels downwards to uncover the surface of the sensor, exposing it to light. After a brief delay determined by the shutter speed selected, the second curtain does the same thing, covering over the sensor again, and terminating the exposure.

The way it produces variable exposure times lies in the delay before the second curtain is released. As mentioned above, at a shutter speed of around 1/125sec, the second curtain will not begin to move until the first curtain has reached the bottom and uncovered the whole of the sensor. But to create shorter exposure times, the second curtain will move before the first curtain has reached the end of its travel. In this example using 1/125sec, if the second curtain starts to travel when the first curtain is only half way down, the effective exposure time will be halved, ie 1/250sec. As the shutter speed is increased, so the delay between the first and second curtains is progressively reduced, and when running at the highest speed, the gap between the first and second curtains will be only 1-2mm as they travel down in tandem over the sensor.

It’s a neat trick, but there is one drawback and that is that flash can only be used at slower shutter speeds, up to the maximum x-sync speed. This is because the duration of the flash is very short, 1/1000sec or less, and if the whole of the image is to be evenly exposed by flash, it must be fired when the entire image sensor is uncovered by the shutter. Clearly, if the flash was to fire at higher speeds, it would only expose a band across the image – just the gap between the shutter curtains as they travel down close together. (There are some flash guns that use a very rapid strobe feature to sync with any shutter speed, but power output and range is greatly reduced.)
Hoppy is offline  
Old 13 September 2006, 11:35 PM   #13
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Creative use of shutter speeds
We’ve read that shutter speeds are used to control the exposure and also how they allow different lens f/numbers to be chosen to vary depth-of-field. The shutter’s other function is to control movement.

The most obvious situation is when you use a fast shutter speed to freeze action. That is straightforward enough, although you may be surprised how fast you need to go – anything longer than 1/500sec is not really that fast, and you’ll need 1/2000sec to see something like droplets of water frozen like ice in a fountain.

Or you can look at it another way and choose a longer shutter speed. Shoot that same fountain at 1/20sec and the droplets will turn into a slightly blurred stream of water, much as it appears to the naked eye. The human eye cannot detect movement any faster than 1/20sec or so, which is why TV images show smooth movement, even though they are composed of a sequence of still images, but when presented at the rate of 25fps or faster the eye cannot see the gaps in-between. With the fountain, go further and at 1/2sec the stream of water will begin to look quite surreal and misty, more like clouds than water. You will need a tripod to use long shutter speeds like this.

You can also deliberately introduce blur to emphasise the impression of speed and movement with a popular technique known as panning. Panning is when you move the camera to follow the subject, often using a mid-range shutter speed like 1/125sec or 1/60sec. The subject should remain static in your viewfinder as you track it, but the background will be rushing past in a blur emphasised by the slightly longer shutter speed. Not only is the impression of speed enhanced, but the blurring of the background makes the main subject stand out more. Panning is a great technique, but needs practise. The secret is smooth and accurate tracking of the subject, but even the experts will take two or three shots if possible.

It is also much easier to pan when you are further away from the subject and using a longer lens, than when you are close. This is because the subject will tend to hold a more constant speed as it comes towards and past you, which is easier to track accurately. When panning at close quarters, the subject will appear to accelerate rapidly as it approaches you, which is much harder to follow. A monopod is a useful accessory to help give your panning skills a boost.

Last edited by Hoppy; 14 September 2006 at 12:20 AM.
Hoppy is offline  
Old 13 September 2006, 11:38 PM   #14
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Camera-shake – avoid it!
It is not only the subject that is likely to move in your pictures, so will the camera unless it is bolted down to a solid tripod. Camera-shake is the result of trying to hand-hold the camera when using a shutter speed that is too long. Camera-shake is very common, and while it is often obvious, more often it is not so obvious but just takes the edge off a nice sharp picture. You might blame the lens, or poor focusing, but it could just be your unsteady hand.

Using telephoto lenses emphasises camera movement, in the same way that it can sometimes be hard to hold a pair of binoculars absolutely steady. There is a popular rule of thumb that can help you estimate the slowest shutter speed that you can use with confidence, and that is the shutter speed should never be longer than the focal length of the lens, so if you use a 200mm lens, don’t go slower than 1/200sec. However, this old rule must be qualified, corrected, and taken with a large pinch of salt.

To start with, this guide only applies to FF-35mm cameras and because the effective focal length is increased by the crop-factor on most D-SLRs, the new 35mm equivalent focal length must be used. So on a D-SLR with a 1.6x crop-factor, 200mm becomes 320mm, and therefore your slowest shutter speed should be 1/320sec.

But even more importantly, you, the photographer, must also be figured into the equation. Nobody’s hands are the same, and how steadily you can hold the camera will also depend on the picture-taking situation. Are you standing upright, relaxed and leaning against a wall for extra support, while holding your elbows firmly - but not too tightly - against your chest? Or are you down on one knee, and stretching uncomfortably to get just the view you want? The difference between these two examples in terms of the shutter speed you need to guarantee a sharp result will be at least two stops (to use the photographers’ jargon) so that 1/320sec needs to be raised to 1/1250sec in the latter circumstances.

One of the best experiments you can ever do, that will only take a few minutes but will benefit every picture you will ever shoot from then on, is to take a series of test pictures with different focal length lenses and in different shooting positions, and compare the results. You will need a tripod in order to get a perfectly sharp image to compare your hand-held shots against. Choose a distant subject so that there’s no danger of inaccurate focus affecting the results (a road sign is good) and run through as many different shutter speeds as you can, with as many different lens focal lengths as you’ve got, and in various shooting positions. Take several pictures each time. With digital, it is so easy to do this and enlarge them on your PC monitor to see just how steady you actually are, compared to the tripod images, and draw up your own personal hand holding formula.

When you’ve done this, you’ll be pleasantly surprised that with wide-angle lenses, you can often hold the camera steadily enough to use quite long shutter speeds, because they reduce the effect of camera-shake, for the same reason that telephotos increase it. The benefits of this are obvious in terms of the exposure settings and the depth-of-field options it gives you. Wouldn’t it be wonderful if you could get away with using longer shutter speeds all the time? Well, now you can, and without using a tripod.

There is new technology available that has a dramatic effect on steadying your shaky hands. One manufacturer calls it Image Stabilization (IS) and another Vibration Reduction (VR) and most manufacturers offer something similar. They all use movement detectors coupled to an electronically driven device that cancels out camera-shake significantly. There’s only one caveat, and that is if you need 1/500sec to freeze action, then shooting at 1/60sec with an anti-shake-system is still going to give you a blurred subject. Nevertheless, this is a major and extremely practical technological breakthrough, and even the extra cost is reasonable.
Hoppy is offline  
Old 13 September 2006, 11:39 PM   #15
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Automatic exposure modes
All D-SLRs have a wide range of exposure control options that provide endless creative permutations to explore. They range from fully Manual operation (M) where every control setting is made by you, through semi-automatic where you make the main choices and the camera does the rest, and on to fully automatic Program (P) where all you have to do is point and shoot.

There will also be a selection of automatic ‘scenic’ exposure modes, which are marked with a symbol depicting popular subjects like action, or a portrait or landscape, and so on. These modes will set the exposure fully automatically, but will also bias the shutter speed and lens aperture selection to give the best pre-programmed result for the type of subject you’ve selected.

For example, if you select the action symbol, the camera will always try to use the fastest shutter speed it can, according to the brightness of the light available. For a portrait, it will assume you want shallow depth-of-field and will chose a large lens aperture, and for landscapes it will go for maximum depth-of-field, so setting a small lens aperture. Clever stuff? Well, yes and no.

For action photography, you will often want a fast shutter speed to freeze movement, but not if you’re panning and want a slower speed to help blur the background. For a close-up portrait, you would usually go for a wide lens aperture for shallow depth-of-field, but if you want to show some of the background to illustrate the subject’s environment, you would probably opt for a smaller lens aperture and zoom back for a wider view.

In situations like these, the semi-automatic modes can be very useful. There are two, marked Tv or S (Time value, or Shutter) and Av or A (Aperture value, or Aperture). With Tv, you set the shutter speed you want and the camera will automatically match it with the correct lens aperture. It will also vary the aperture if the light changes while you’re busy concentrating, so that’s one thing less to worry about. Setting Av does the same thing, but in reverse of course, after you have selected the lens aperture.
Hoppy is offline  
Old 13 September 2006, 11:41 PM   #16
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Flash in daylight
Flash doesn’t have to be used only at night – it can also produce wonderful pictures in daylight. The technique is called fill-in flash, when the flash is used to fill-in dark areas of shadow. It works particularly well for portraits.

If your subject is facing the sun, they will squint and look uncomfortable. So turn them around, so that the light is shining from the side or from behind them. With the sun in their hair, they’ll look great - girls, guys, kids, it doesn’t matter. But now the face will be in shadow and if you make a normal exposure, the camera will see lots of bright light that will lead to under exposure of the face. That will give you a silhouette effect.

You now have two options to correct this: firstly, you could dial in some plus exposure compensation, which will lighten the face, but will bleach the bright background into pure white (blown highlights). Ideally, what you want is to balance the attractive bright light coming from behind with more light from the front. You could do this with a reflector, but they are big, cumbersome, and tend to get blown away (literally!) so your second easy option is to use fill-in flash. Fortunately, fill-in doesn’t need much power so the built-in flash will be fine, and the camera will do all the exposure calculations for you, to ensure a good balance between the daylight and flash. It will also prevent you from using a shutter speed faster than the x-sync speed, lest you forget. Easy, and it looks great with the flash also putting a nice highlight in the eyes.
Hoppy is offline  
Old 13 September 2006, 11:41 PM   #17
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Flash at night
When the sun goes down, a lot of easy picture taking opportunities go with it, but you are also presented with a new set of challenges that you can work around for great reward. You just need to try a little harder - crank up the ISO, open the lens to its largest aperture, use the anti-camera-shake feature if you have it, and hone your technique with careful focusing and a steady hand. Your reward will be photos full of natural atmosphere.

Artificial light, from either ordinary light bulbs or fluorescent tubes, is much less bright than sunlight and even though you can see perfectly well, your camera cannot do the same. Also, most kinds of artificial light are not the same colour as daylight, and it’s necessary to adjust the white-balance control to get natural-looking pictures. Just let the camera’s AWB (Auto White Balance) take care of it and make any fine adjustments at the printing stage.

Or you can use flash. While flash is no substitute for daylight, if you understand its strengths and weaknesses, you can adapt to its limitations and get some great pictures.

Red-eye is a common problem, so common that there are several tricks that camera makers use to get rid of it. None of them works very well, although many software programmes can detect and eliminate it quite effectively in post-processing. But it’s much better to avoid it in the first place. Red-eye is caused by light from the flash reflecting back off the retina of the eye, directly into the camera. It is worst when the distance between the flash head and camera lens is close, so moving the flash away from the lens is an instant cure. However, if the flash is built-in to the camera, this is obviously impossible and a good reason to buy a separate flashgun. This will have much more power and can be used in different positions and in various ways to greatly improve your flash photography.

Bounce-flash is one technique that you can use with a more powerful gun, and it will eliminate red-eye, give soft natural looking light, and get you around a tricky bit of physics called the inverse-square law. The inverse-square law simply means that if you double the distance from the flash to the subject, the light is reduced to one quarter. Try it yourself by pointing a torch at a plain wall and measure the circle of light. Move back to double the distance and when you measure it again, the circle of light will be double the diameter, which is four times the area and so also a quarter of the brightness.

Here’s a familiar example where the inverse-square law explains those horrible snaps at parties where someone’s face close to the camera is bleached ghostly white, while other people just a few feet away are almost too dark to recognise. Here’s why – if the nearest person was two feet from the flash, then people four feet away will be lit by only one quarter of the light, and eight feet away the light is reduced to one sixteenth. That is far too great a range of brightness (dynamic range) and the solution is firstly to move back so that the greater distance reduces the rate at which the light falls off, then try and arrange your picture so that all important parts are roughly the same distance from the flash. Finally, point the flashgun at a white ceiling and it will flood the room with light as the flash bounces off the ceiling and down again. It’s a really easy technique as all the complex exposure calculations are handled by the camera. The only problem may be a lack of flash power – bounce-flash just gobbles light, so try to buy the most powerful flashgun you can.
Hoppy is offline  
Old 13 September 2006, 11:43 PM   #18
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Pixels – How many do you need?
Pixels were once the be-all and end-all of any digital camera decision, but it has now reached a stage where the sharpness of digital pictures is often higher than you can actually see in the final print. The supremacy of film is now beaten and chasing ever-higher pixel counts needs to be considered in context with other very important aspects of image quality. In particular, D-SLR image sensors are physically much larger than those found in compact digi-cameras. Larger sensors have bigger pixels which inherently produce much less noise and have a greater dynamic range from bright highlights to dark shadows.

It is generally agreed that 300 pixels-per-inch (ppi) is the finest level of detail the eye can naturally detect, and you can still get very good looking prints at lower levels than that. Put another way, if you never make prints larger than 6x4in, then at 300ppi anything more than 2.2m pixels is wasted.

Don’t try to compare a camera’s ppi count against the dots-per-inch (dpi) of a printer as printers use many dots to reproduce one pixel depending on the colour, so direct comparisons can’t be made. However, let’s say that a printer capable of delivering 2,800dpi or more will produce very high quality prints. If they do not, check your printer settings before blaming the hardware.

By the same measure, moving up to 10x7in prints requires an absolute maximum of 6.3m pixels at 300ppi and in practise, you can get very high quality prints of this size from a camera with half this number of pixels. This is because larger prints are always viewed from a little further away, and bigger enlargements will also allow a D-SLR’s superior management of other image data, like noise and dynamic range, to come to the fore.

But having said that, it is always nice to have a little more resolving power in hand - a few more pixels - for big enlargements and for zooming-in at the printing stage to get the best composition, but without losing sharpness. Just bear in mind what has been said above and also that at larger print sizes, it is quite likely that the resolving power of the lens (its ability to show fine detail) will also be exceeded in some situations, or that your focusing accuracy will be slightly out, or that the image will be degraded by camera-shake. So, if you are seeking ultimate image quality, you must also ensure that your lenses are the best available, that you have a high quality photo printer using six or more inks (not just four) and that you always use the best printing paper. Last but certainly not least, your photographic and digital processing techniques have got to be absolutely impeccable.
Hoppy is offline  
Old 13 September 2006, 11:44 PM   #19
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

JPEG and Raw images
D-SLRs are mini-computers that contain sophisticated image processing software to help you get the best quality photos. The two most common types of image data file are JPEG (Joint Photographic Experts Group) and Raw, which just means unprocessed.

JPEG processing is a universal standard that both enhances and compresses image files, and it works very well. Unless you intend to spend time post-processing with specialist software, JPEG is the one to use. You can select different standards of JPEG compression, and the big bonus is substantially reduced file sizes – you can get hundreds of high quality pictures on to a 1Gb memory card. The camera can also handle these smaller files much more easily, so if you are shooting fast action at maybe 5fps, the camera will keep going longer before the buffer is filled and you are forced to pause briefly while data is transferred to the memory card.

You can also customise JPEG processing by adjusting sharpness, contrast, saturation and colour tone levels in the camera to get just the kind of photos you like straight out of the box. The camera’s default settings are usually a good starting point, but to see the effect these adjustments to levels make, do some test prints and compare. It’s similar to adjusting the picture settings on your TV, but what comes out of the camera might not look exactly the same on your PC monitor, and may be slightly changed again by your printer. This is not a fault and it is possible to get all your devices calibrated together, but you can usually fix things yourself with a few simple adjustments to your printer output.

Raw is for experienced users who want the absolute maximum quality possible and are prepared to put the work in to get it. Raw takes the whole image file from the sensor and records it on the memory card without any alteration. Unprocessed Raw images often don’t look their best, so you must post-process them with a PC and software, but then you will be rewarded with outstanding results.

Some cameras allow you to shoot both Raw files and JPEGs simultaneously, giving you the best of both worlds, but Raw files are also huge. Their size varies according to the total number of pixels but they are many times larger than JPEGs, and will fill up your memory card quickly.

On the subject of post-processing, there are many computer software packages available, many of them free. Even the most basic will allow you to make worthwhile improvements to your photos very easily, and with the most sophisticated programmes you can transform your pictures into absolutely anything imaginable (and unfortunately, some people do).
Hoppy is offline  
Old 13 September 2006, 11:45 PM   #20
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Composition
This is a controversial topic, where rules are often best when broken, so we’ll stick to some simple basics. Perhaps the best advice on composition is to stop, look and think. Now read that again, slowly. Stop. Look. Think. What is it that you are actually trying to photograph? Concentrate on that and try to eliminate anything that distracts your eye from the main subject, or adds nothing to the overall image. Check the background and either cut out anything unnecessary, or use a larger lens aperture to make it out of focus. Take your time, and choose your moment. Don’t just pick up the camera, zoom in and fire off a few frames in the hope that one will be good. Chances are that it won’t be.

If you can, walk around your subject, seek out interesting angles, and if you have an eye for a picture (and if you’re reading this, you probably do) very often the various elements in the image will fall into place, the composition will look balanced and the picture will work.

And when the picture does look right and well composed, you’ll usually find that key areas of interest correspond with the Rule of Thirds. It works like this. Imagine a rectangle, say 6x9 inches. Along the 9in side, measure 3in from the end and draw a line to the opposite side, move along another 3in and draw another line. Now move to the shorter side, measure 2in from the end and draw a line across, then move another 2in and do the same again.

Your basic rectangle will now be divided into nine smaller rectangles, and the lines will intersect at four points around the centre of the frame. These four points, each one a third-in from the outside edge, are the points referred to in the Rule of Thirds. This is where the eye tends to rest naturally when viewing an image. A simple test – imagine this template and overlay it mentally onto pictures that you like. You will be surprised how often keys areas of interest fall on one or more of the Thirds. Try it when watching TV. It’s almost spooky.

A simpler way of putting it is to avoid placing key elements of the picture right in the middle, but this is a little at odds with another good phrase to remember when composing your picture – Fill the Frame for Impact. For maximum punch, cut out all unnecessary detail, move in as close as you dare (and then a little more) and place your subject bang in the centre. It’s something of a shock tactic, but with the right subject, it is very powerful.

One more thing – leading lines. These are often the key to great landscape pictures where the foreground sometimes adds nothing to the image and can even distract from a great view beyond. So look for a leading line, often a road, river, a fence or wall, that attracts attention and then draws your eye into the heart of the picture towards the point of interest. This can be hard (unless you have a portable wall to hand) but when it works, it really makes the picture. Of course, leading lines don’t only apply to landscapes or to objects in the picture. They can work with any image, and might be just areas of light or dark, or of a contrasting colour, that helps to direct the eye.
Hoppy is offline  
Old 13 September 2006, 11:46 PM   #21
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Getting started is easy
That’s all for now… and plenty to be getting on with. Welcome to the world of creative photography. If you’re still struggling to take it all in, you really shouldn’t worry. Photography has now evolved to such an advanced technological state that the camera can do most things for you and you’ll always get a result – a result that you can learn from, develop and improve upon.

Take your time, use Program or the scenic exposure modes at first, use auto-focus, use auto-white-balance and shoot JPEGs. Just make sure you note what’s happening in the viewfinder readout, and when looking at your finished pictures, take time to check the ExIF (Exchange Image File) data tagged to the image file – it records all the technical details. You’ll soon begin to see what camera settings and photographic techniques bring out the best in different subjects.

You don’t need a computer either – just pop in to your local high street photo processor and slot your memory card into their in-store digital printer. Save the files to CD at the same. Or get yourself a direct-photo printer that has built-in slots to download your camera’s memory card, preferably one with a small LCD screen so you can preview your images. These printers even have simple software incorporated to eliminate red-eye, to zoom in, or to lighten and darken your prints. These printers are cheap, couldn’t be easier to use, and can turn out superb A4 enlargements to show-case your talent.

Good luck!

Produced by Richard ‘Hoppy’ Hopkins
Hoppy is offline  
Old 13 September 2006, 11:49 PM   #22
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Answers to the exposure setting question
The correct exposure indicated is 1/125sec at f/8 with ISO 100. In order to use a shutter speed of 1/500sec, you must either use a larger lens aperture, in this case f/4, or increase the ISO setting to 400. There is a third option of setting the lens to f/5.6 and sensor speed to ISO 200, which will also allow you to set the shutter speed at 1/500sec.

There are even more options available if you are prepared to push the ISO rating higher, to ISO 800 or 1600, but you run the risk of increasing noise that could degrade image quality. That will be okay for small prints, but will be more noticeable in larger prints. As always, the choice is yours.

In summary, the primary options as above are:
1/125sec at f/8, ISO 100
1/500sec at f/4, ISO 100
1/500sec at f/8, ISO 400
1/500sec at f/5.6, ISO 200

For an even faster shutter speed to freeze action, you could also set:
1/1000sec at f/8, ISO 800
1/2000sec at f/8, ISO 1600

All these settings will result in the same effective exposure, with the emphasis on a fast shutter speed. There are many more alternative combinations if you want to emphasise depth-of-field by setting the f/number, and then balancing the exposure with adjustments to shutter speed and sensor ISO.

All text Copyright Richard Hopkins, December 2005.
If you would like to reproduce any part of this article, please ask.
Hoppy is offline  
Old 13 September 2006, 11:51 PM   #23
Hoppy
Scooby Regular
Thread Starter
 
Join Date: Aug 2000
Location: Where age and treachery reins over youthful exuberance
Posts: 5,275
Default

Guys, please don't post anything on this thread. It will be made sticky and locked soon.

See here instead: https://www.scoobynet.com/understand...g-t540241.html

Many thanks,

Richard.

Last edited by Hoppy; 13 September 2006 at 11:54 PM.
Hoppy is offline  
 
 
Closed Thread

Related Topics
Thread Thread Starter Forum Replies Last Post
Phase 2 (v5) heads on phase 1.5 block. Anyone done this before??? bluebullet29 General Technical 9 05 October 2015 02:17 PM
MotorBike Races and Cars Littleted Non Scooby Related 6 02 October 2015 11:31 AM
Clean Boost Solenoid mistermexican General Technical 2 01 October 2015 04:30 PM
Where is the consistency in Scoobynet's moderators Blue by You Non Scooby Related 48 30 September 2015 01:27 PM
Please help me understand my dyno output StueyBII Engine Management and ECU Remapping 5 22 September 2015 02:13 PM


Tags
aperture, bbs, brand, camera, converting, equivalent, features, iso, lenses, photography, scooby, sensor, specific, speed

Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off

Forum Jump


All times are GMT +1. The time now is 07:59 AM.


 
  • Ask a Question
    Get answers from community experts
What's your question?
Send