Friday, December 24, 2004
[Seeing as everyone else will be talking about all things Christmas, such as turkeys which cost a pound (Sainsbury's misread their £7 reduced sticker), and how awful it all is, or how marvellous it is, I thought I would be truly cunning, and publish a post which I started ages ago, and which then grew. Anyway, here's just a little something I flung together on the subject of photography. It is only a rough draft at this stage, and I hope to be able to add, correct and improve at some as yet uncertain time in the future. Anyway, hope you enjoy it or find it useful and informative, or just don't hate for it. It's not terribly good, but I have spent far too long on it already].
So this'll be the photography post then?
Maybe, if I stayed focussed enough. Speaking of which...someone stuck another pin in the map in the sidebar. This time it's from Portugal, and links to his blog. One slight problem. I don't speak Portuguese. I'm not even sure I know how to say that in any language other than English. So about all I understand are the pictures, and they are giving me a slightly unusual impression of the guy. Anyone care to enlighten me?
Carrying on the theme of virtual globe-trotting, does anyone have any more information on why the Open University's Hitmap project is suspended? Because I think it's quite funky [and pretty, see GfB for an example], and each time I try to add it, I find the message about it being temporarily off line.
[It now appears not to be. Expect one to pop up fairly soon on here].
And back to photography [and if it feels like I'm lecturing you, I'm not, I'm just trying to remind myself].
The basics [as I consider them to be. Most of the photography sites start way above, and most beginners somewhere below this level].
On my camera the following are variables:
- Film speed
- Shutter speed or Exposure length.
- Focus
- Aperture
- Zoom
- Exposure compensation
- Time delay
- Single or continuous shooting
- Whether the thing beeps or not.
These are roughly in order of importance. Film speed is obviously something you need to get right, as even the most basic cameras allow one to adjust this (or nowadays probably have it automatically adjusted to suit the film). Mine sets on the DX setting, which basically means it does what the film tells it too. Most of the time I use 100 or 200 colour 35 mm film.
Why? Because it's what I've always used. It's what most people mean when they say film. 35 mm [pronounced, round here at least, as "thirty five mill"] is the common film that's been around a fair while. Why colour? Because it's more colourful, and if I was using black and white film, I'd feel guilty for not developing the film myself. Developing BW film and making prints isn't that hard, you just need some equipment, a suitable room, a few chemicals, a bit of preparation and some time. However I don't have most of that, and I can't really remember enough of what I ought to do. Whereas I know colour film is much more complicated, and a lot more has to be done in the dark. So I leave it to people who might get it better than I would.
Why 100 or 200? Because I take pictures of buildings, hills, and people standing round. I don't usually take pictures of Formula 1 cars going at full pelt. I don't need a faster reacting film [such as 400]. Also, the general rule is that the faster the film, the less depth the resultant image carries. I don't know who decried this, and I haven't rigorously tested it. By depth, I think they mean the level of colour saturation [ignore for the time being the suggestion that saturation might suggest a maximum, as it is saturated, and thus can take no more. In this context, saturation can be thought of as more like a percentage of the maximum level].
So, I've done film, right? Basically I use the standard stuff, because I don't know better. I tend to use refer Kodak over own-brand stuff, because it is very forgiving [the Tanzania shots were taken with Kodak Gold, and a Halina 160 (Made in Hong Kong). Have you ever heard of the Halina brand? Me neither. It's a basic point and click rangefinder (separate viewfinder lens) camera from Boots. I don't think they make them that simple anymore], and has given me so gorgeous results. Which given my photography skills, says a lot for the film. Yes, it is much too expensive, but I haven't figured out a way round that.
Why is shutter speed adjustment more important than focus adjustment? [shutter speed meaning the period the shutter is open for, and not necessarily the speed of the shutter movement] It is important to be able to focus properly on the subject, but in certain conditions this will not matter. And those conditions are whenever the picture is liable to be over- or under-exposed, or blurred. It doesn't matter if the photographer can count the subject's eyelashes, if the image they produce is a mass of blacks, in which the eyes cannot be seen.
What does adjusting the exposure length allow one to do? Avoid [hopefully] under-exposing or over-exposing an image. Avoid problems caused by movements [either by the subject, or the photographer]. For example, today might be gloriously sunny. You might be trying to take a picture of vanilla ice cream melting in the sun. Take that picture. Done? Good. But now, this being England, a thunderstorm rolls in. It just got a lot darker. But you're in no hurry, so you'll wait it out in the doorway outside a bridal shop. There's a gown in window, and you know a friend is getting married soon, and having a nightmare finding the right dress. It looks like what she said she wanted. Take a picture, so you can show her. Golly, the rain is taking a long time to pass. By the time has cleared, the sun has set, and the sky is looking very dramatic. But you are forever taking pictures of clouds, and they come out right. Hmm, but that light is stunning. As you walk home, you notice the reflections in the wet tarmac. If you can get that to come out it will be a stunning shot. You try. As you continue home, the colour is draining from the sky. The temperature drops sharply, and suddenly mist drifts in. The streetlights in the mist should also make a good image, especially with that silhouetted tree. Another frame gone. You get home, having left the mist in the valley. The moonlight is picking out the lighter colours, against a very black night. You only had a few shots left on the film when you started, so you may as well use the last frame to see if you can get the eerie way the daisies appear to be floating in nothing.
You go in...and I fast forward through your life...you've just picked up the pictures you had developed from the film you were using on that awful day. Before you open it, tell me how many of the shots do you think came out. Ah, but you've noticed, I haven't said what camera you were using. Suppose, for an instant, that it was that old point and click one, the one with the broken flash which you can't replace. So what could you adjust on that? Well, there's the film speed, and the flash, and the latter is permanently off now. You used that same film all day, so why change the film speed? So, that leaves how many variables? Yes, that's right just the stuff in front of the camera, which is quite frankly a bit hard to control [unless you happen to be a god, and even then it's too much like hard work].
So, knowing you changed nothing on the camera, how many of the images do you think will be an accurate representation of what you saw? There were 5 pictures you took. If you're lucky you might get 3 which don't inspire someone to ask "what was that?". If you're me, you'll get one at the level of just about passable, the rest are just various shades of grey.
So why didn't some of them work? Because we used the same film throughout. So why not just change the film to one that suits a dark subject in post-thunderstorm light? Because then you'd also need an ice cream in sunlight film, a wedding dress in a shady shop window film, a streetlights in the mist film, and daisies in pitch black lawn film. Which is quite a lot of films. Then if you bear in mind how fiddly changing films can be, and the fact that once you've wound a film into the case, you can't pull it back out again. So that's several types of films, all of which would have to be single shot to avoid wasting the other 23 or 35, and you would have to spend more time changing film than taking pictures. Not really viable is it?
So what else can you do? Get your camera to compensate. Suppose you have a film ideal for use in poor [typical English] light. Only now, despite it being the beginning of December, the sun has come out. And it's really, really bright. If you expose the film to this light using normal settings, the poor thing will go white with the shock [well, ok black, because it's the negative, but that ruins the analogy]. But you want a picture of that church, which isn't helping matters by being built out of cream Portland stone. So what can you do?
Ah-ha! (not the "take on me" type). There just so happens to be a cunning dial on your camera, which allows you to alter the length of time the shutter remains open for. Mine runs all the way, halving as it goes, from 2000 to B, but the last number is 1 [it also has a couple of other settings, but that's just needless complication]. Now what they don't print on the dial is what these numbers mean. So is 2000 big or little, fast or slow? The numbers correspond to the amount by which one second is divided, giving the exposure period. For example, with 2000, it means the shutter remains open for 1/2000th of a second. Which isn't very long. As it isn't very long, not much light will come through. At the other end of the scale 1 means 1/1 seconds, which means the shutter stays open for a whole second, and so lets in a lot of light.
Right so getting back to the church in the sunshine. You have miserable day film, but a nice day shot. How can you turn nice into miserable? By only showing the film a little bit of it. So, if you used a exposure of 1/125th sec, on a grotty day, then use a much shorter exposure on a sunny day. Each notch upwards [towards the smaller numbers] doubles the period of exposure, so should allow twice as much light in. So if you want to lessen the light coming in, use a larger number [faster speed] on the dial. As for how much you should do this by, use a light metre to work it out (handily my camera has an inbuilt one).
Right so that's shutter speed, and the impact on the exposure covered.
Focus
You know how eyes work right, and you've played with lenses? Please say yes, as I really don't want to have to explain basic optics. Just as we can use focusing to create a clear image on our retina, we can use focusing to create a clear image on the film. If you have an SLR [single lens reflex - the viewfinder shows the view through the same set of lenses as are used when the film is exposed], you can see the effects of the focusing. Ensuring the camera is on manual mode [for those whose cameras run to such luxuries], point it at something fairly close, move the focusing ring to its fullest extent in one direction. Look at the image that provides. Now move it back to the other extreme. Which is less blurry? Or is somewhere in-between clearer? Find the clearest point.
Now aim the camera at the horizon, or as near as you can get. Not so clear is it? Go back to whatever it was you were focusing on beforehand. Is the main feature still in focus in most of the view [ignore the section in the middle]? Make it so. Now, [assuming you have a focusing aid like mine, which you might not] looking at the three sections in the middle, there is a ring with a grid superimposed on it, and inside this ring two semicircles. Looking at the central part, which is split done the middle, what do you see? Does each half match the other? Do features crossing the line meet in middle? Do the tones match? If they do, well done, and now nudge the focus, and try to make them line up again. If it wasn't lined up in first place, guess what I'm going to say. Yep, line it up. It's a fiddle I know, but getting things in focus is kind of worth it. If you get it right, both images should meet, and the dark and light differences should disappear [but the tonal stuff is secondary].
Moving on to the weird scattering ring. It supposed to be used for "quick focusing" and on surfaces lacking in linear features. However it is easiest to demonstrate where there is a defined difference, such as printing on a flat surface. Now try focusing the image within the grid on the ring. One does this by adjusting the focusing until the scattering effect becomes minimal. I find this method of focusing the hardest, and it usually makes me feel slightly sick. This is why I suggested using printing, as it is easier to marry up the true image, and the faceted versions, than with some other subjects [for example trying to work out where each piece of green goes in the leaves on a tree].
So that is 3 sorts of focusing: general field, microprism ring, and split-image, which are in ascending order of their accuracy. Theoretically that is the mirror image of their speed of use, but I have problems with the prism version.
Aperture.
The aperture is a ring of plates which can be progressively opened or closed, forming a central hole of variable size. This controls the amount of light coming to the camera, and effects the focusing. But we have already had options that allow us to manipulate both of those, so why have it? Because it gives us greater control.
The aperture size is usually given as the f-stop or f-number. My camera has f3.5, 5.6, 8, 11, 16, and 22. The numbers might appear to be haphazardly chosen, but there is a pattern to it. Unfortunately I can't remember the pattern, other than it has the square root of 2 in it somewhere, and was fairly complicated. I think, that despite appearances, each change in the f-stop adjusts the level of incoming light by a set proportion. So just as one can adjust the exposure length by doubling it for each increment down the scale [towards 1/2000th], so too can one double [roughly] the area of the aperture [and so incoming light] by each increment. In this case the smaller the f-stop figure, the larger the aperture, or hole through which lights comes. Got that? You'll need it. So f3.4 is as wide open as my camera will go.
But we also control light by using the shutter speed, so why both again? What was I saying about greater control? Yep. The astute amongst you might have noticed that if you have two methods for controlling the same thing, then one can also use those methods for controlling something else. So imagine we are back in that sunny place, outside the wedding dress shop. This time there's something about the way the veil is pleated which catches your eye. So you want to take a picture. You don't know what settings to use, and so opt for a middle f-stop, about f11. You're lucky and the inbuilt light-sensor recommends that you use an exposure length of 1/60th of a second [it must be shadier than you thought inside the shop]. You take the picture.
Suddenly you hear the squealing of tires, and a cacophony of sirens. It's coming from somewhere over…there they are. They've turned onto the road you are on. Neglecting to realise the danger you are in, and ignoring the instinct to flee, the photographer in you takes over. You drop down into a crouch, in order to make a more dramatic angle. The cars are coming fast. You've got to take the picture. You will not get another chance so this has to be right. But they're moving so fast. What do you do? Fast cars means fast shutter, unless you want a series of indistinct streaks across the frame. You wang the speed up to 1/2000th. The LEDs are flashing. It's too dark. You remember the aperture can allow more light in. But which way was it? Big is more, no, less, no, yes, right? You spin that round to the smallest number and the largest hole. Damn, it's still too dark. Help! You drop the shutter speed down to 1/1000 s. The LEDs are happy, and there's no more flashing lights, over than those on the police cars. They're nearly here. A bit of rough focusing and ker-click. You have the shot. And I'll just F6 this little version of GTA2 [by the way, I feel I should warn you, as there is a very tempting chain of Elvises on the pavement directly behind you, that you might want to jump soon].
So what happened there? You started on f11 with an exposure length of 1/60 s, and ended on f3.4 and 1/1000 s. How's that work? (Other than it being jolly handy that the buildings across the road cause exactly the same amount of shading of the white police cars as the shop window did of the veil). At a set light level there is a shutter speed that works best with a certain aperture, for example 1/60 s with f11. But if you want to adjust the shutter speed, to avoid blurring for instance, then you are able to adjust the aperture to compensate. If you halve the exposure time, and the amount of light hitting the film, by going to 1/125th s, then at that light level, you will underexpose the image. To compensate you must somehow double the light coming in, back up to the original level. Hence you use the aperture to do that. Fortunately, it has been designed so that each notch it turns towards the largest opening doubles the area of the hole at the previous notch. So to counterbalance the change from 1/60 s to 1/125 s, you need to move the aperture setting from f11 to f8. As one gets physically bigger the other must get physically smaller [although remember that the aperture settings are the inverse of the size of the hole, and so a big f-stop is a small hole].
So how many options do we have at this fixed example light level? My camera will do the following pairings: f22 1/15, f16 1/30, f11 1/60, f8 1/125, f5.6 1/250, f3.5 1/500. Why on earth do we need 6 options to take one set of conditions? Well, we have already had to cope with high-speed subjects. I also mentioned before the belief that a shorter exposure reduces the apparent saturation of the resulting print. So we have one trade-off: sumptuous versus quick.
However there is also another effect of changing the f-stop. It changes the depth of field, which is the range of the focused area. Manipulation of the depth of field is a very common technique used in photography. I'm sure you can think of examples in which one item in an image is in focus, and another blurred. It is used to throw the attention onto a specific subject. Imagine you're a private detective [or deranged stalker if you prefer]. You are watching your target through a window. She is standing in the kitchen, with a cluttered work surface behind her. A man comes in, throws something onto the work surface, and they kiss. You ought to be taking a picture right now.
But, you've just noticed that this is your last shot on that film [happens a lot doesn’t it?], so you'll only get one go. You want to capture the action clearly, so you want a fast shutter speed. So you pick f3.5 and whatever the camera suggests, which is 1/250 s this time. You try to set up the focus, and manage to get the window frame clearly. Well, it will be useful for showing which house this is, but it is not what you really want. Change the focus, and you've got the couple, mid-embrace. Great. But what was that packet he threw on the counter? Focusing beyond the couple you can see it is a brown envelope with what looks like money in it, and an address on it, which you can't make out through the viewfinder. So, do you show the location, the action, or the goods? The camera is only letting you show one of these, as the depth of field is so shallow that the rest are out of focus.
But surely there must a way round this. And there is, and you've probably guessed it by now. Adjust the aperture. Due to very complicated reasons, the aperture changes the depth of field, which means it changes how much of a scene one can have in focus at once. The narrower the aperture, the greater the depth of field. So if you want to have that window frame, the couple, and the envelope all in focus, you need a narrow aperture, and so a large f-number, such as f22. But this means that you will have to adjust the shutter speed to cancel out the change in incoming light. So how many notches will you have to move it? You moved the aperture 5, so… yep, 5. In which direction? You have less light coming in, so you need a slower speed. So, from 1/250 s, you should end up on 1/8 s. Which is slow.
So we have another trade-off - depth of field versus speed, which in this light level is not a good choice. So what should you do? An eighth of a second is a long time in photography, and so you will need a tripod (or a very steady hand). But it is either that or using several different shots to show each item, and you have only got the one frame left on the film. It is up to you. You could of course feel the need to ensure the liaison occurs earlier in the day, that that cloud is not there, or perhaps that they will not notice that the bush outside the window has grown 3 metallic legs, as well as to two human feet.
Coincidentally, they have been kissing for a very long time, have they not?
Moving on, or possibly zooming on. Yes, you have got it; it is now time for the audible warnings switch. Or maybe the zoom instead.
Lenses come in all manner of, well, everything. There are 50 mm such-and-such's, 200 mm whatever's, et cetera. But what does it mean? The number expressed in millimetres reflects the focal length, between the final internal lens and the film surface. So in a 50 mm lens, the final internal lens is five centimetres away from the film. Some lenses have more than one focal length. This allows one to zoom in and out, by adjusting the magnification of the image. My camera ranges between 28 mm and 70 mm [it also has a Macro setting, but that can left alone for now]. This is not a huge zoom, but I cannot afford a huge zoom lens.
What effect does the zoom have? Let's get back to that kitchen window. At 28 mm, the view through the window occupies the middle third of the frame. At 70 mm, it fills two thirds of the frame. If we move closer until it fills the frame and then zoom back out, the hole fills just under half the frame. So if we decide that we want the frame of the window to match the frame of the image, we move closer, and use the zoom. But hang on, moving closer to that menacing man, who is in the process of dropping off that dubious package, whilst we are clutching a camera with which to record to evil deeds…not today thank you. Shall we just stick where we are, in this nice comfortable holly bush, and let the camera do the work?
Whilst I'm doing zoom, I may as well mention the dreaded parallax. Parallax is the variation in apparent position which occurs when our eyes move relative to two objects. So for example, you are driving your car. You think you are doing 70 mph, as that is where the needle appears to be over on the face of the speedometer. You lean your head to the right, so it would be out of the window if you are English, Indian, Australian or Kenyan [or anywhere else we invaded], and over towards the passenger seat if you are American. Now look at the speedometer. The needle appears to be somewhere over 60 mph. Oh dear, driving below the speed limit - we can't be having that. So you accelerate until the needle appears over 70 mph once again. Ok, so our American friends have scavenged whatever they want out of the glovebox, while those of us who drive on the right [as in correct, not as in not left] side of the road decide that inhaling exhaust fumes at 70 mph is not all that fun. Reverting back to our normal driving positions, and we look once more at the speedometer. Oh no! Now we are doing nearly 80 miles per hour! My goodness, if we go any faster, we will find ourselves time-travelling [Whilst I would never break the law (much), I have discovered that Hollywood lies. 88 mph is not a magical number. I did not find myself 50 years hence. Although I wasn't driving a Delorean, so might be why]. Getting back to the point, if we move in comparison to two objects [the needle and the printed dial], they apparently move compared to each other.
Take two equal sized objects, such as two pencils. Hold both upright. Hold your right arm straight ahead of you, with the pencil sticking up at the end. Place your left hand on the crook of your elbow, holding the other pencil up. Shut your left eye. Keeping you arms still, move your head until they line up. How big is the apparent distance between them? How large are they relative to each other? Um, well you should be able to see the nearer one, so it was a bit cruel to ask.
Now, open your left eye, and shut your right. And how did you manage to read this, and look at the pencils at the same time? Wow, the power that is multitasking. Once again how big do they seem comparatively, and how far apart do they appear? Well, you can see both for a start. Now flip back to using your right eye. And back to the left. Yes, you are playing peek-a-boo with a pencil. But you are also discovering your own internal parallax. Which actually provides the foundation for 3D vision, but I am not going to get into that now.
So what where the differences you saw? When you could see it, the nearer pencil appeared larger. Why? Perspective, and once again I am not going to cover this now. If you do not know what that is, go and watch the Father Ted episode with the toy cows, and those which are merely far away. Suffice to say, the nearer one is to something the bigger it appears. So in camera terms, the more of the frame it will fill.
And what about the distance between them. Well, we know that your right arm is unlikely to have shrunk and grown during the experiment, so the actual distance between the pencils remained the same. But your view of the pencils did change. In one they were so near as to be indistinguishable, and the other they appeared a far distance apart. But what I forgot to do was to get you to look at them as you moved your head gradually. So we will do that now.
Lay both pencils so they overhang the edge of a table, about 10 cm apart. If possible get them perpendicular [at right angles or 90o] to the edge, and parallel to each other. Shut one eye, and stick with using only that eye. Now lay your head on the edge of the table, a fixed distance away from the nearest pencil, using a ruler or your hand-span to help you (just don't poke your eye out). Can you see both pencils? Probably not, or only just. Move your head so it is halfway between surface of the table and a point directly over the first pencil, check the distance away from the first pencil. Now what do you see? The separation of the pencils has increased. Now move until you are directly above the first pencil, still keeping the same distance away. The separation should appear to be about double what it was at the midway [45o] point. Now move back round the curve you have just come up, rotating around the nearer pencil. Watch how the separation wanes. It appears to change slowly at first, then speeds up in the middle, and then slowing again. It follows a sinusoidal pattern, but as that is a bit too complicated for now, I will leave it.
Anyway, despite rotating evenly around the pencils, the change is not uniform [and I have made it a bit less uniform by using a pencil as the axis, not the midpoint between them, but I thought an invisible point might be a bit hard to stick to, especially when one of the items it is derived from cannot be seen].
So by now you probably understand that as one's point of view shifts, the view one sees shifts. If one moves around two items, the view changes. If one moves towards two items, the view changes. Exactly the same effects work on the camera. Stand down the lane from two gateposts flanking a path off the road, and they will appear together, as a slight change in the rhythm of the fence which runs through your shot. Walk up the lane, and they diverge. Stand opposite the gateway, and they are even more widely separated, showing the track down to the farm beyond. Walk through the gate, and one post passes to one side of you, and the other to the other side. You would be hard pushed to get both in the same shot.
So what has all this stuff about kitchens, cars and gates shown us? That if you move the camera, the composition within the frame is effected, and so the image is not the same. But what if you want the angled gate to fill the frame, but you happen to be standing down the lane. You could try walking closer, but then you would be looking down on the gate. You could then kneel, but then the brambles just this side of the gate become too prominent, and the nearer gatepost is much larger than the further one. You cannot take quite the same picture. You cannot move the camera and expect the same image. You cannot use your feet to zoom. You can get closer, but that is not the same. Zooming allows one to crop an image and enlarge it in one go, but without the usual faults enlarge often causes.
But on the other hand, zooming, because it is operating on items far away, invariably has a long depth of field. And so causes flattening of the image, with the background and much of the foreground being equally in focus. So it makes it harder to use the plane of focus to pick out a certain detail, whilst discarding the rest within a blur.
But I think I might have done zooming to death, so I will start on the next item. The exposure compensation control. Which I have never quite got round to using, despite there being instances when I should have done so. This setting allows one to adjust to exposure to under or over-expose the entire image. Why on earth should anyone wish to do this? The example where I could have used this feature was when I was at a friend's birthday party on boat going up and down a river. There was seating facing inboard along the gunwales of the boat. I tried taking pictures of my friends. Most of them are dark ovals with halos of hair in the middle of a stunning picture of the estuary behind them. Oh dear. What I should have done was use the dial to adjust the shutter speed by a notch or two. Which would have left me leave my friends looking normal, but with an overexposed and washed out looking view through the window behind them.
The opposite scenario would be if one were taking a picture of someone in a spotlight. Let's say that it is your eldest child's first school play, and they are playing the lead role [unlikely I know, but not everyone can be Sheep Number 3]. The school are really piling the pressure on, and your child is giving a monologue, as a solitary spotlit figure against a dimly lit set. It's beautiful, and you want to treasure the moment forever [so you have some embarrassing blackmail material for use on your offspring's potential partners]. You point the camera at you child. He, she or it is a slender figure, who only reaches two thirds of the way up the image. The camera's inbuilt light sensor decides it is quite dark, and so adjusts the suggested shutter speed up to a reasonable level, to allow the average light levels to match the requirements of the film.
Note that word average. Three little piggies each own one house. The first house has a value of £3,000 pounds, as it only built of straw. The second, being built of wood, is worth £6,000. The third, a stone house, is worth £60,000 [it is in the back of beyond, hence the low asking price]. So there is a mean [average] value of £23,000. So when Hurricane Wolf huffs and puffs and blows all the houses down, how will the three pigs respond when the insurance company pays out the average cost of an house in the area? Two of them will be delighted, being much richer than before, and one will be a bit miffed, as he has just lost nearly forty thousand pounds. [And no pointing out that the insurance company would probably opt for the median value of £6,000 or claim the pigs were not covered for anything officially known as "the Big Bad…"].
So averages can be unfair. Some get too little, and some too much. Cameras cannot average out the light that hits film, as doing so would cause every image to be various shades of grey. But light metres do only use the average. So, the averaging light-metre would instruct the camera to use a shutter speed which lets too much light hit on one part of the film, and not enough on the rest. The greater the disparity, such as the smaller the child in the frame, or the stronger the contrast, the more the average is skewed from the desired exposure. Think of DOS. One small white chevron in a sea of black is not going to make much impact on the average, which will be very dark grey. And so to recreate this very dark grey, the automatic response of the camera is to adjust the shutter speed so most of the field of view would come out very dark grey. So the DOS screen would have an odd blob of blinding white in the upper left-hand corner. And the shot of your firstborn would allow you to see the ghost of you child, in an all concealing white haze. But you can pick out the screws on the scenery behind the child.
So to stop this happening, we need to drop the exposure down a bit, by either using the adjustor dial, or by ignoring the advice of camera, and selecting a shutter speed which is faster than that the camera suggests. We do this and the child appears to float above the darkened depths, like the angelic being that we known them to be.
There is also (according to the manual, which I cannot check, having lost it again), a button by the self-timer control, which allows one to use the exposure recommended for one view on another. For example, if one wanted to take a shot of the roof of a house against the sky, but did not want the roof to the silhouetted, then one could set the exposure whilst the building fills the frame, and then swing the camera upwards, so the composition is right. This would allow the roof to be properly exposed, whilst the sky would probably become over-exposed.
However, despite having used this button, it miraculously seems to have become part of the self-timer, and caused me to take an unintended shot of a pile of boxes. So I won't be pressing that again, until I find the instructions.
Which brings me on to the self-timer. It allows one to delay the point when the photograph is taken. Usually just long enough for the people in shot to begin to wonder if the thing is still working, and so they begin frowning and muttering amongst themselves. It also allows enough time for any faults in tripods to become apparent, as the camera starts sagging, or rolls off the convenient perch one had found for it. How it can take so long, and yet not be long enough for the last minute changes one invariably needs, I do not know, merely that it does. I have never successfully used it, and usually only manage to set it off accidentally.
Now onto something for which there might conceivably be slightly more use. Switching between single and continuous shooting. If one is taking shots of a scene which is changing rapidly, continuous shooting might prove useful. If there can only be one chance to take pictures, one may as well cram as many in as possible. So instead of only having one shot of a rally car crashing, during which a spectator fleeing blocks most of the view, one could have a series, thereby increasing the likelihood of getting a decent shot. Also a series of similar, but temporally displaced photographs can prove very effective visually [and for demonstrating how horses gallop]. Having said that, film is expensive, as is developing, which given I am not a freelance photographer capable of selling my pictures, means I having used this feature in earnest, but only by mistake.
And so onto the last adjustable feature: the audible warnings switch. Which sits on mute the whole time, as it flattens the batteries, scares off wildlife, and the beeps and whines irritate me beyond measure.
Speaking of scaring off wildlife, I need to get a flash, as well as learn how to use one. This because most of the times I want to take pictures, the shutter speeds end up too low, and I get blurred images.
That appears to be all the variables on my camera, but I have hunch that this is a bit too long winded for most people. So here is a recap of the important stuff.
- Cameras. Viewfinder: separate viewer from shutter lens. Rangefinder: separate lenses, use to find distance, then adjust camera lens. Single Lens Reflex: proper camera, shutter and viewing screen share lens.
- Film speed: Fast = faster subjects, but grainer and less rich. Use slowest practical.
- Shutter speed controls exposure length. Inverse exponential scale, as 1/X second. Bigger number = faster, for use in bright conditions, or action shots.
- Focus. Split image most accurate [align]. Microprism for soft shapes [reduce speckles]. General field for quick estimate [least blurred].
- Aperture. Smallest number, biggest hole. Each notch bigger doubles incoming light. Allows flexible shutter speed, and depth of field trade-off.
- Depth of field. Greatest with furthest focus, smaller [higher f-stop] aperture. Approximately1/3 before, 2/3 behind optimum focus.
- Zoom. Change dominance, enlarge subject, without perspective or parallax problems.
- Exposure compensation. Graduated, persistent override of automatic light sensing. Alternative to manual shutter speed based adjustment.
- Self-timer. Delay shutter release, e.g. for self.
- Single/Continuous shooting. Series can be effective, also scattergun approach.
And now for the quiz.
1. A magpie habitually lands on the roof of your house, just above a window. You get a stunning view of the splayed wings as it comes in to land. It is a bright, yet overcast day. For ASA 100, f8, my camera suggests using 1/60 s. Which of the following should you use?
A. ASA 100, f3.5, 1/500 s, 28mm lens, fd 7 m.
B. ASA 200, f8, 1/50 s, 70mm lens, fd 0.9 m.
C. ASA 200, f11, 1/1000 s, Macro lens, fd 1.2 m.
D. ASA 400, f11, 1/125 s, 28mm lens, fd ∞ m.
E. ASA 800, f5.6, 1/2000 s, 70mm lens, fd 0.9 m.
2. I am using 1/125 s, f8, with ASA 200 film. You have ASA 100 film, and your camera is jammed on f16. What shutter speed should to use?
A. 1/8 s.
B. 1/15 s.
C. 1/30 s.
D. 1/60 s.
E. 1/125 s.
3. Select the lens with the widest field of view.
A. 20 mm, f11.
B. 30 mm, f3.5.
C. 40 mm, f22.
D. 50 mm, f5.6.
E. Macro, f3.5.
4. You are using ASA 200, f5.6, 1/8 s. What are the conditions?
A. Strong direct sunshine.
B. Strong reflected light.
C. Overcast sky, but bright.
D. Unlit room.
E. Night.
5. Family portrait time. Your family are sitting at a table, with 3 down each side, with you standing at the head to take a photograph (you have them well trained). Select the best settings for ASA 100.
A. 1/2000 s, f3.5.
B. 1/1000 s, f5.6.
C. 1/250 s, f11.
D. 1/125 s, f16.
E. 1/60 s, f22.
Oh, do you want answers now? Really some people. Ok, highlight the section after this, as the text should have been set to match the background colour.
S
P
A
C
E
R
1: e. Fast film, fast shutter speed, zoomed in to fill the frame [the bird is very predictable], close focus. Aperture is being driven by shutter speed. Maybe not ideal, but the best of the bunch.
2: b. ASA 100 is twice as slow as 200. So that the shutter speed needs to halve, to 1/60 s. Then it is 2 stops from f8 up to f16, so the shutter speed needs be quadrupled [doubled twice]. So, 2 x 2 x 1/60 s = 4 x 1/60 s = 4/60 s = 1/15 s.
3: a. The shorter the focal length of the lens, the wider the field of vision. Aperture is irrelevant.
4: d. Slow shutter speed means a long exposure. Long exposures are generally bad, as things [including the camera] move, and therefore would be avoided if they could be. f5.6 suggests that the aperture is as open as possible whilst still retaining the depth of field. ASA 200 is fairly normal speed film, but no conclusions should be drawn from its use, as it could just be what is in the camera. So the camera is set up to make almost maximum use of light. Which suggests there is not much. But 1/8 s is not really suitable for use during the dead of night. So it is probably inside a building, either with poor daylight levels, or on a dull day.
5: e. The family start quite close to the lens, and finish a few metres away. Turning your uncle into a soft focus blur is not a good idea, so you need a long depth of field. The house is not big enough to have everyone in focus at near infinity, and you need to choose the aperture which allows the greatest depth of field. Which is the smallest hole, and highest f-stop, or f22.
Comments, queries, objections and corrections can all go in the comments section below. You could email me, but if there's a counter-argument then it's probably that other people know about it.
References
- Spillman R, 1971. ### Complete + links ###
Well that was fun.
Anyhoo,
So this'll be the photography post then?
Maybe, if I stayed focussed enough. Speaking of which...someone stuck another pin in the map in the sidebar. This time it's from Portugal, and links to his blog. One slight problem. I don't speak Portuguese. I'm not even sure I know how to say that in any language other than English. So about all I understand are the pictures, and they are giving me a slightly unusual impression of the guy. Anyone care to enlighten me?
Carrying on the theme of virtual globe-trotting, does anyone have any more information on why the Open University's Hitmap project is suspended? Because I think it's quite funky [and pretty, see GfB for an example], and each time I try to add it, I find the message about it being temporarily off line.
[It now appears not to be. Expect one to pop up fairly soon on here].
And back to photography [and if it feels like I'm lecturing you, I'm not, I'm just trying to remind myself].
The basics [as I consider them to be. Most of the photography sites start way above, and most beginners somewhere below this level].
On my camera the following are variables:
- Film speed
- Shutter speed or Exposure length.
- Focus
- Aperture
- Zoom
- Exposure compensation
- Time delay
- Single or continuous shooting
- Whether the thing beeps or not.
These are roughly in order of importance. Film speed is obviously something you need to get right, as even the most basic cameras allow one to adjust this (or nowadays probably have it automatically adjusted to suit the film). Mine sets on the DX setting, which basically means it does what the film tells it too. Most of the time I use 100 or 200 colour 35 mm film.
Why? Because it's what I've always used. It's what most people mean when they say film. 35 mm [pronounced, round here at least, as "thirty five mill"] is the common film that's been around a fair while. Why colour? Because it's more colourful, and if I was using black and white film, I'd feel guilty for not developing the film myself. Developing BW film and making prints isn't that hard, you just need some equipment, a suitable room, a few chemicals, a bit of preparation and some time. However I don't have most of that, and I can't really remember enough of what I ought to do. Whereas I know colour film is much more complicated, and a lot more has to be done in the dark. So I leave it to people who might get it better than I would.
Why 100 or 200? Because I take pictures of buildings, hills, and people standing round. I don't usually take pictures of Formula 1 cars going at full pelt. I don't need a faster reacting film [such as 400]. Also, the general rule is that the faster the film, the less depth the resultant image carries. I don't know who decried this, and I haven't rigorously tested it. By depth, I think they mean the level of colour saturation [ignore for the time being the suggestion that saturation might suggest a maximum, as it is saturated, and thus can take no more. In this context, saturation can be thought of as more like a percentage of the maximum level].
So, I've done film, right? Basically I use the standard stuff, because I don't know better. I tend to use refer Kodak over own-brand stuff, because it is very forgiving [the Tanzania shots were taken with Kodak Gold, and a Halina 160 (Made in Hong Kong). Have you ever heard of the Halina brand? Me neither. It's a basic point and click rangefinder (separate viewfinder lens) camera from Boots. I don't think they make them that simple anymore], and has given me so gorgeous results. Which given my photography skills, says a lot for the film. Yes, it is much too expensive, but I haven't figured out a way round that.
Why is shutter speed adjustment more important than focus adjustment? [shutter speed meaning the period the shutter is open for, and not necessarily the speed of the shutter movement] It is important to be able to focus properly on the subject, but in certain conditions this will not matter. And those conditions are whenever the picture is liable to be over- or under-exposed, or blurred. It doesn't matter if the photographer can count the subject's eyelashes, if the image they produce is a mass of blacks, in which the eyes cannot be seen.
What does adjusting the exposure length allow one to do? Avoid [hopefully] under-exposing or over-exposing an image. Avoid problems caused by movements [either by the subject, or the photographer]. For example, today might be gloriously sunny. You might be trying to take a picture of vanilla ice cream melting in the sun. Take that picture. Done? Good. But now, this being England, a thunderstorm rolls in. It just got a lot darker. But you're in no hurry, so you'll wait it out in the doorway outside a bridal shop. There's a gown in window, and you know a friend is getting married soon, and having a nightmare finding the right dress. It looks like what she said she wanted. Take a picture, so you can show her. Golly, the rain is taking a long time to pass. By the time has cleared, the sun has set, and the sky is looking very dramatic. But you are forever taking pictures of clouds, and they come out right. Hmm, but that light is stunning. As you walk home, you notice the reflections in the wet tarmac. If you can get that to come out it will be a stunning shot. You try. As you continue home, the colour is draining from the sky. The temperature drops sharply, and suddenly mist drifts in. The streetlights in the mist should also make a good image, especially with that silhouetted tree. Another frame gone. You get home, having left the mist in the valley. The moonlight is picking out the lighter colours, against a very black night. You only had a few shots left on the film when you started, so you may as well use the last frame to see if you can get the eerie way the daisies appear to be floating in nothing.
You go in...and I fast forward through your life...you've just picked up the pictures you had developed from the film you were using on that awful day. Before you open it, tell me how many of the shots do you think came out. Ah, but you've noticed, I haven't said what camera you were using. Suppose, for an instant, that it was that old point and click one, the one with the broken flash which you can't replace. So what could you adjust on that? Well, there's the film speed, and the flash, and the latter is permanently off now. You used that same film all day, so why change the film speed? So, that leaves how many variables? Yes, that's right just the stuff in front of the camera, which is quite frankly a bit hard to control [unless you happen to be a god, and even then it's too much like hard work].
So, knowing you changed nothing on the camera, how many of the images do you think will be an accurate representation of what you saw? There were 5 pictures you took. If you're lucky you might get 3 which don't inspire someone to ask "what was that?". If you're me, you'll get one at the level of just about passable, the rest are just various shades of grey.
So why didn't some of them work? Because we used the same film throughout. So why not just change the film to one that suits a dark subject in post-thunderstorm light? Because then you'd also need an ice cream in sunlight film, a wedding dress in a shady shop window film, a streetlights in the mist film, and daisies in pitch black lawn film. Which is quite a lot of films. Then if you bear in mind how fiddly changing films can be, and the fact that once you've wound a film into the case, you can't pull it back out again. So that's several types of films, all of which would have to be single shot to avoid wasting the other 23 or 35, and you would have to spend more time changing film than taking pictures. Not really viable is it?
So what else can you do? Get your camera to compensate. Suppose you have a film ideal for use in poor [typical English] light. Only now, despite it being the beginning of December, the sun has come out. And it's really, really bright. If you expose the film to this light using normal settings, the poor thing will go white with the shock [well, ok black, because it's the negative, but that ruins the analogy]. But you want a picture of that church, which isn't helping matters by being built out of cream Portland stone. So what can you do?
Ah-ha! (not the "take on me" type). There just so happens to be a cunning dial on your camera, which allows you to alter the length of time the shutter remains open for. Mine runs all the way, halving as it goes, from 2000 to B, but the last number is 1 [it also has a couple of other settings, but that's just needless complication]. Now what they don't print on the dial is what these numbers mean. So is 2000 big or little, fast or slow? The numbers correspond to the amount by which one second is divided, giving the exposure period. For example, with 2000, it means the shutter remains open for 1/2000th of a second. Which isn't very long. As it isn't very long, not much light will come through. At the other end of the scale 1 means 1/1 seconds, which means the shutter stays open for a whole second, and so lets in a lot of light.
Right so getting back to the church in the sunshine. You have miserable day film, but a nice day shot. How can you turn nice into miserable? By only showing the film a little bit of it. So, if you used a exposure of 1/125th sec, on a grotty day, then use a much shorter exposure on a sunny day. Each notch upwards [towards the smaller numbers] doubles the period of exposure, so should allow twice as much light in. So if you want to lessen the light coming in, use a larger number [faster speed] on the dial. As for how much you should do this by, use a light metre to work it out (handily my camera has an inbuilt one).
Right so that's shutter speed, and the impact on the exposure covered.
Focus
You know how eyes work right, and you've played with lenses? Please say yes, as I really don't want to have to explain basic optics. Just as we can use focusing to create a clear image on our retina, we can use focusing to create a clear image on the film. If you have an SLR [single lens reflex - the viewfinder shows the view through the same set of lenses as are used when the film is exposed], you can see the effects of the focusing. Ensuring the camera is on manual mode [for those whose cameras run to such luxuries], point it at something fairly close, move the focusing ring to its fullest extent in one direction. Look at the image that provides. Now move it back to the other extreme. Which is less blurry? Or is somewhere in-between clearer? Find the clearest point.
Now aim the camera at the horizon, or as near as you can get. Not so clear is it? Go back to whatever it was you were focusing on beforehand. Is the main feature still in focus in most of the view [ignore the section in the middle]? Make it so. Now, [assuming you have a focusing aid like mine, which you might not] looking at the three sections in the middle, there is a ring with a grid superimposed on it, and inside this ring two semicircles. Looking at the central part, which is split done the middle, what do you see? Does each half match the other? Do features crossing the line meet in middle? Do the tones match? If they do, well done, and now nudge the focus, and try to make them line up again. If it wasn't lined up in first place, guess what I'm going to say. Yep, line it up. It's a fiddle I know, but getting things in focus is kind of worth it. If you get it right, both images should meet, and the dark and light differences should disappear [but the tonal stuff is secondary].
Moving on to the weird scattering ring. It supposed to be used for "quick focusing" and on surfaces lacking in linear features. However it is easiest to demonstrate where there is a defined difference, such as printing on a flat surface. Now try focusing the image within the grid on the ring. One does this by adjusting the focusing until the scattering effect becomes minimal. I find this method of focusing the hardest, and it usually makes me feel slightly sick. This is why I suggested using printing, as it is easier to marry up the true image, and the faceted versions, than with some other subjects [for example trying to work out where each piece of green goes in the leaves on a tree].
So that is 3 sorts of focusing: general field, microprism ring, and split-image, which are in ascending order of their accuracy. Theoretically that is the mirror image of their speed of use, but I have problems with the prism version.
Aperture.
The aperture is a ring of plates which can be progressively opened or closed, forming a central hole of variable size. This controls the amount of light coming to the camera, and effects the focusing. But we have already had options that allow us to manipulate both of those, so why have it? Because it gives us greater control.
The aperture size is usually given as the f-stop or f-number. My camera has f3.5, 5.6, 8, 11, 16, and 22. The numbers might appear to be haphazardly chosen, but there is a pattern to it. Unfortunately I can't remember the pattern, other than it has the square root of 2 in it somewhere, and was fairly complicated. I think, that despite appearances, each change in the f-stop adjusts the level of incoming light by a set proportion. So just as one can adjust the exposure length by doubling it for each increment down the scale [towards 1/2000th], so too can one double [roughly] the area of the aperture [and so incoming light] by each increment. In this case the smaller the f-stop figure, the larger the aperture, or hole through which lights comes. Got that? You'll need it. So f3.4 is as wide open as my camera will go.
But we also control light by using the shutter speed, so why both again? What was I saying about greater control? Yep. The astute amongst you might have noticed that if you have two methods for controlling the same thing, then one can also use those methods for controlling something else. So imagine we are back in that sunny place, outside the wedding dress shop. This time there's something about the way the veil is pleated which catches your eye. So you want to take a picture. You don't know what settings to use, and so opt for a middle f-stop, about f11. You're lucky and the inbuilt light-sensor recommends that you use an exposure length of 1/60th of a second [it must be shadier than you thought inside the shop]. You take the picture.
Suddenly you hear the squealing of tires, and a cacophony of sirens. It's coming from somewhere over…there they are. They've turned onto the road you are on. Neglecting to realise the danger you are in, and ignoring the instinct to flee, the photographer in you takes over. You drop down into a crouch, in order to make a more dramatic angle. The cars are coming fast. You've got to take the picture. You will not get another chance so this has to be right. But they're moving so fast. What do you do? Fast cars means fast shutter, unless you want a series of indistinct streaks across the frame. You wang the speed up to 1/2000th. The LEDs are flashing. It's too dark. You remember the aperture can allow more light in. But which way was it? Big is more, no, less, no, yes, right? You spin that round to the smallest number and the largest hole. Damn, it's still too dark. Help! You drop the shutter speed down to 1/1000 s. The LEDs are happy, and there's no more flashing lights, over than those on the police cars. They're nearly here. A bit of rough focusing and ker-click. You have the shot. And I'll just F6 this little version of GTA2 [by the way, I feel I should warn you, as there is a very tempting chain of Elvises on the pavement directly behind you, that you might want to jump soon].
So what happened there? You started on f11 with an exposure length of 1/60 s, and ended on f3.4 and 1/1000 s. How's that work? (Other than it being jolly handy that the buildings across the road cause exactly the same amount of shading of the white police cars as the shop window did of the veil). At a set light level there is a shutter speed that works best with a certain aperture, for example 1/60 s with f11. But if you want to adjust the shutter speed, to avoid blurring for instance, then you are able to adjust the aperture to compensate. If you halve the exposure time, and the amount of light hitting the film, by going to 1/125th s, then at that light level, you will underexpose the image. To compensate you must somehow double the light coming in, back up to the original level. Hence you use the aperture to do that. Fortunately, it has been designed so that each notch it turns towards the largest opening doubles the area of the hole at the previous notch. So to counterbalance the change from 1/60 s to 1/125 s, you need to move the aperture setting from f11 to f8. As one gets physically bigger the other must get physically smaller [although remember that the aperture settings are the inverse of the size of the hole, and so a big f-stop is a small hole].
So how many options do we have at this fixed example light level? My camera will do the following pairings: f22 1/15, f16 1/30, f11 1/60, f8 1/125, f5.6 1/250, f3.5 1/500. Why on earth do we need 6 options to take one set of conditions? Well, we have already had to cope with high-speed subjects. I also mentioned before the belief that a shorter exposure reduces the apparent saturation of the resulting print. So we have one trade-off: sumptuous versus quick.
However there is also another effect of changing the f-stop. It changes the depth of field, which is the range of the focused area. Manipulation of the depth of field is a very common technique used in photography. I'm sure you can think of examples in which one item in an image is in focus, and another blurred. It is used to throw the attention onto a specific subject. Imagine you're a private detective [or deranged stalker if you prefer]. You are watching your target through a window. She is standing in the kitchen, with a cluttered work surface behind her. A man comes in, throws something onto the work surface, and they kiss. You ought to be taking a picture right now.
But, you've just noticed that this is your last shot on that film [happens a lot doesn’t it?], so you'll only get one go. You want to capture the action clearly, so you want a fast shutter speed. So you pick f3.5 and whatever the camera suggests, which is 1/250 s this time. You try to set up the focus, and manage to get the window frame clearly. Well, it will be useful for showing which house this is, but it is not what you really want. Change the focus, and you've got the couple, mid-embrace. Great. But what was that packet he threw on the counter? Focusing beyond the couple you can see it is a brown envelope with what looks like money in it, and an address on it, which you can't make out through the viewfinder. So, do you show the location, the action, or the goods? The camera is only letting you show one of these, as the depth of field is so shallow that the rest are out of focus.
But surely there must a way round this. And there is, and you've probably guessed it by now. Adjust the aperture. Due to very complicated reasons, the aperture changes the depth of field, which means it changes how much of a scene one can have in focus at once. The narrower the aperture, the greater the depth of field. So if you want to have that window frame, the couple, and the envelope all in focus, you need a narrow aperture, and so a large f-number, such as f22. But this means that you will have to adjust the shutter speed to cancel out the change in incoming light. So how many notches will you have to move it? You moved the aperture 5, so… yep, 5. In which direction? You have less light coming in, so you need a slower speed. So, from 1/250 s, you should end up on 1/8 s. Which is slow.
So we have another trade-off - depth of field versus speed, which in this light level is not a good choice. So what should you do? An eighth of a second is a long time in photography, and so you will need a tripod (or a very steady hand). But it is either that or using several different shots to show each item, and you have only got the one frame left on the film. It is up to you. You could of course feel the need to ensure the liaison occurs earlier in the day, that that cloud is not there, or perhaps that they will not notice that the bush outside the window has grown 3 metallic legs, as well as to two human feet.
Coincidentally, they have been kissing for a very long time, have they not?
Moving on, or possibly zooming on. Yes, you have got it; it is now time for the audible warnings switch. Or maybe the zoom instead.
Lenses come in all manner of, well, everything. There are 50 mm such-and-such's, 200 mm whatever's, et cetera. But what does it mean? The number expressed in millimetres reflects the focal length, between the final internal lens and the film surface. So in a 50 mm lens, the final internal lens is five centimetres away from the film. Some lenses have more than one focal length. This allows one to zoom in and out, by adjusting the magnification of the image. My camera ranges between 28 mm and 70 mm [it also has a Macro setting, but that can left alone for now]. This is not a huge zoom, but I cannot afford a huge zoom lens.
What effect does the zoom have? Let's get back to that kitchen window. At 28 mm, the view through the window occupies the middle third of the frame. At 70 mm, it fills two thirds of the frame. If we move closer until it fills the frame and then zoom back out, the hole fills just under half the frame. So if we decide that we want the frame of the window to match the frame of the image, we move closer, and use the zoom. But hang on, moving closer to that menacing man, who is in the process of dropping off that dubious package, whilst we are clutching a camera with which to record to evil deeds…not today thank you. Shall we just stick where we are, in this nice comfortable holly bush, and let the camera do the work?
Whilst I'm doing zoom, I may as well mention the dreaded parallax. Parallax is the variation in apparent position which occurs when our eyes move relative to two objects. So for example, you are driving your car. You think you are doing 70 mph, as that is where the needle appears to be over on the face of the speedometer. You lean your head to the right, so it would be out of the window if you are English, Indian, Australian or Kenyan [or anywhere else we invaded], and over towards the passenger seat if you are American. Now look at the speedometer. The needle appears to be somewhere over 60 mph. Oh dear, driving below the speed limit - we can't be having that. So you accelerate until the needle appears over 70 mph once again. Ok, so our American friends have scavenged whatever they want out of the glovebox, while those of us who drive on the right [as in correct, not as in not left] side of the road decide that inhaling exhaust fumes at 70 mph is not all that fun. Reverting back to our normal driving positions, and we look once more at the speedometer. Oh no! Now we are doing nearly 80 miles per hour! My goodness, if we go any faster, we will find ourselves time-travelling [Whilst I would never break the law (much), I have discovered that Hollywood lies. 88 mph is not a magical number. I did not find myself 50 years hence. Although I wasn't driving a Delorean, so might be why]. Getting back to the point, if we move in comparison to two objects [the needle and the printed dial], they apparently move compared to each other.
Take two equal sized objects, such as two pencils. Hold both upright. Hold your right arm straight ahead of you, with the pencil sticking up at the end. Place your left hand on the crook of your elbow, holding the other pencil up. Shut your left eye. Keeping you arms still, move your head until they line up. How big is the apparent distance between them? How large are they relative to each other? Um, well you should be able to see the nearer one, so it was a bit cruel to ask.
Now, open your left eye, and shut your right. And how did you manage to read this, and look at the pencils at the same time? Wow, the power that is multitasking. Once again how big do they seem comparatively, and how far apart do they appear? Well, you can see both for a start. Now flip back to using your right eye. And back to the left. Yes, you are playing peek-a-boo with a pencil. But you are also discovering your own internal parallax. Which actually provides the foundation for 3D vision, but I am not going to get into that now.
So what where the differences you saw? When you could see it, the nearer pencil appeared larger. Why? Perspective, and once again I am not going to cover this now. If you do not know what that is, go and watch the Father Ted episode with the toy cows, and those which are merely far away. Suffice to say, the nearer one is to something the bigger it appears. So in camera terms, the more of the frame it will fill.
And what about the distance between them. Well, we know that your right arm is unlikely to have shrunk and grown during the experiment, so the actual distance between the pencils remained the same. But your view of the pencils did change. In one they were so near as to be indistinguishable, and the other they appeared a far distance apart. But what I forgot to do was to get you to look at them as you moved your head gradually. So we will do that now.
Lay both pencils so they overhang the edge of a table, about 10 cm apart. If possible get them perpendicular [at right angles or 90o] to the edge, and parallel to each other. Shut one eye, and stick with using only that eye. Now lay your head on the edge of the table, a fixed distance away from the nearest pencil, using a ruler or your hand-span to help you (just don't poke your eye out). Can you see both pencils? Probably not, or only just. Move your head so it is halfway between surface of the table and a point directly over the first pencil, check the distance away from the first pencil. Now what do you see? The separation of the pencils has increased. Now move until you are directly above the first pencil, still keeping the same distance away. The separation should appear to be about double what it was at the midway [45o] point. Now move back round the curve you have just come up, rotating around the nearer pencil. Watch how the separation wanes. It appears to change slowly at first, then speeds up in the middle, and then slowing again. It follows a sinusoidal pattern, but as that is a bit too complicated for now, I will leave it.
Anyway, despite rotating evenly around the pencils, the change is not uniform [and I have made it a bit less uniform by using a pencil as the axis, not the midpoint between them, but I thought an invisible point might be a bit hard to stick to, especially when one of the items it is derived from cannot be seen].
So by now you probably understand that as one's point of view shifts, the view one sees shifts. If one moves around two items, the view changes. If one moves towards two items, the view changes. Exactly the same effects work on the camera. Stand down the lane from two gateposts flanking a path off the road, and they will appear together, as a slight change in the rhythm of the fence which runs through your shot. Walk up the lane, and they diverge. Stand opposite the gateway, and they are even more widely separated, showing the track down to the farm beyond. Walk through the gate, and one post passes to one side of you, and the other to the other side. You would be hard pushed to get both in the same shot.
So what has all this stuff about kitchens, cars and gates shown us? That if you move the camera, the composition within the frame is effected, and so the image is not the same. But what if you want the angled gate to fill the frame, but you happen to be standing down the lane. You could try walking closer, but then you would be looking down on the gate. You could then kneel, but then the brambles just this side of the gate become too prominent, and the nearer gatepost is much larger than the further one. You cannot take quite the same picture. You cannot move the camera and expect the same image. You cannot use your feet to zoom. You can get closer, but that is not the same. Zooming allows one to crop an image and enlarge it in one go, but without the usual faults enlarge often causes.
But on the other hand, zooming, because it is operating on items far away, invariably has a long depth of field. And so causes flattening of the image, with the background and much of the foreground being equally in focus. So it makes it harder to use the plane of focus to pick out a certain detail, whilst discarding the rest within a blur.
But I think I might have done zooming to death, so I will start on the next item. The exposure compensation control. Which I have never quite got round to using, despite there being instances when I should have done so. This setting allows one to adjust to exposure to under or over-expose the entire image. Why on earth should anyone wish to do this? The example where I could have used this feature was when I was at a friend's birthday party on boat going up and down a river. There was seating facing inboard along the gunwales of the boat. I tried taking pictures of my friends. Most of them are dark ovals with halos of hair in the middle of a stunning picture of the estuary behind them. Oh dear. What I should have done was use the dial to adjust the shutter speed by a notch or two. Which would have left me leave my friends looking normal, but with an overexposed and washed out looking view through the window behind them.
The opposite scenario would be if one were taking a picture of someone in a spotlight. Let's say that it is your eldest child's first school play, and they are playing the lead role [unlikely I know, but not everyone can be Sheep Number 3]. The school are really piling the pressure on, and your child is giving a monologue, as a solitary spotlit figure against a dimly lit set. It's beautiful, and you want to treasure the moment forever [so you have some embarrassing blackmail material for use on your offspring's potential partners]. You point the camera at you child. He, she or it is a slender figure, who only reaches two thirds of the way up the image. The camera's inbuilt light sensor decides it is quite dark, and so adjusts the suggested shutter speed up to a reasonable level, to allow the average light levels to match the requirements of the film.
Note that word average. Three little piggies each own one house. The first house has a value of £3,000 pounds, as it only built of straw. The second, being built of wood, is worth £6,000. The third, a stone house, is worth £60,000 [it is in the back of beyond, hence the low asking price]. So there is a mean [average] value of £23,000. So when Hurricane Wolf huffs and puffs and blows all the houses down, how will the three pigs respond when the insurance company pays out the average cost of an house in the area? Two of them will be delighted, being much richer than before, and one will be a bit miffed, as he has just lost nearly forty thousand pounds. [And no pointing out that the insurance company would probably opt for the median value of £6,000 or claim the pigs were not covered for anything officially known as "the Big Bad…"].
So averages can be unfair. Some get too little, and some too much. Cameras cannot average out the light that hits film, as doing so would cause every image to be various shades of grey. But light metres do only use the average. So, the averaging light-metre would instruct the camera to use a shutter speed which lets too much light hit on one part of the film, and not enough on the rest. The greater the disparity, such as the smaller the child in the frame, or the stronger the contrast, the more the average is skewed from the desired exposure. Think of DOS. One small white chevron in a sea of black is not going to make much impact on the average, which will be very dark grey. And so to recreate this very dark grey, the automatic response of the camera is to adjust the shutter speed so most of the field of view would come out very dark grey. So the DOS screen would have an odd blob of blinding white in the upper left-hand corner. And the shot of your firstborn would allow you to see the ghost of you child, in an all concealing white haze. But you can pick out the screws on the scenery behind the child.
So to stop this happening, we need to drop the exposure down a bit, by either using the adjustor dial, or by ignoring the advice of camera, and selecting a shutter speed which is faster than that the camera suggests. We do this and the child appears to float above the darkened depths, like the angelic being that we known them to be.
There is also (according to the manual, which I cannot check, having lost it again), a button by the self-timer control, which allows one to use the exposure recommended for one view on another. For example, if one wanted to take a shot of the roof of a house against the sky, but did not want the roof to the silhouetted, then one could set the exposure whilst the building fills the frame, and then swing the camera upwards, so the composition is right. This would allow the roof to be properly exposed, whilst the sky would probably become over-exposed.
However, despite having used this button, it miraculously seems to have become part of the self-timer, and caused me to take an unintended shot of a pile of boxes. So I won't be pressing that again, until I find the instructions.
Which brings me on to the self-timer. It allows one to delay the point when the photograph is taken. Usually just long enough for the people in shot to begin to wonder if the thing is still working, and so they begin frowning and muttering amongst themselves. It also allows enough time for any faults in tripods to become apparent, as the camera starts sagging, or rolls off the convenient perch one had found for it. How it can take so long, and yet not be long enough for the last minute changes one invariably needs, I do not know, merely that it does. I have never successfully used it, and usually only manage to set it off accidentally.
Now onto something for which there might conceivably be slightly more use. Switching between single and continuous shooting. If one is taking shots of a scene which is changing rapidly, continuous shooting might prove useful. If there can only be one chance to take pictures, one may as well cram as many in as possible. So instead of only having one shot of a rally car crashing, during which a spectator fleeing blocks most of the view, one could have a series, thereby increasing the likelihood of getting a decent shot. Also a series of similar, but temporally displaced photographs can prove very effective visually [and for demonstrating how horses gallop]. Having said that, film is expensive, as is developing, which given I am not a freelance photographer capable of selling my pictures, means I having used this feature in earnest, but only by mistake.
And so onto the last adjustable feature: the audible warnings switch. Which sits on mute the whole time, as it flattens the batteries, scares off wildlife, and the beeps and whines irritate me beyond measure.
Speaking of scaring off wildlife, I need to get a flash, as well as learn how to use one. This because most of the times I want to take pictures, the shutter speeds end up too low, and I get blurred images.
That appears to be all the variables on my camera, but I have hunch that this is a bit too long winded for most people. So here is a recap of the important stuff.
- Cameras. Viewfinder: separate viewer from shutter lens. Rangefinder: separate lenses, use to find distance, then adjust camera lens. Single Lens Reflex: proper camera, shutter and viewing screen share lens.
- Film speed: Fast = faster subjects, but grainer and less rich. Use slowest practical.
- Shutter speed controls exposure length. Inverse exponential scale, as 1/X second. Bigger number = faster, for use in bright conditions, or action shots.
- Focus. Split image most accurate [align]. Microprism for soft shapes [reduce speckles]. General field for quick estimate [least blurred].
- Aperture. Smallest number, biggest hole. Each notch bigger doubles incoming light. Allows flexible shutter speed, and depth of field trade-off.
- Depth of field. Greatest with furthest focus, smaller [higher f-stop] aperture. Approximately1/3 before, 2/3 behind optimum focus.
- Zoom. Change dominance, enlarge subject, without perspective or parallax problems.
- Exposure compensation. Graduated, persistent override of automatic light sensing. Alternative to manual shutter speed based adjustment.
- Self-timer. Delay shutter release, e.g. for self.
- Single/Continuous shooting. Series can be effective, also scattergun approach.
And now for the quiz.
1. A magpie habitually lands on the roof of your house, just above a window. You get a stunning view of the splayed wings as it comes in to land. It is a bright, yet overcast day. For ASA 100, f8, my camera suggests using 1/60 s. Which of the following should you use?
A. ASA 100, f3.5, 1/500 s, 28mm lens, fd 7 m.
B. ASA 200, f8, 1/50 s, 70mm lens, fd 0.9 m.
C. ASA 200, f11, 1/1000 s, Macro lens, fd 1.2 m.
D. ASA 400, f11, 1/125 s, 28mm lens, fd ∞ m.
E. ASA 800, f5.6, 1/2000 s, 70mm lens, fd 0.9 m.
2. I am using 1/125 s, f8, with ASA 200 film. You have ASA 100 film, and your camera is jammed on f16. What shutter speed should to use?
A. 1/8 s.
B. 1/15 s.
C. 1/30 s.
D. 1/60 s.
E. 1/125 s.
3. Select the lens with the widest field of view.
A. 20 mm, f11.
B. 30 mm, f3.5.
C. 40 mm, f22.
D. 50 mm, f5.6.
E. Macro, f3.5.
4. You are using ASA 200, f5.6, 1/8 s. What are the conditions?
A. Strong direct sunshine.
B. Strong reflected light.
C. Overcast sky, but bright.
D. Unlit room.
E. Night.
5. Family portrait time. Your family are sitting at a table, with 3 down each side, with you standing at the head to take a photograph (you have them well trained). Select the best settings for ASA 100.
A. 1/2000 s, f3.5.
B. 1/1000 s, f5.6.
C. 1/250 s, f11.
D. 1/125 s, f16.
E. 1/60 s, f22.
Oh, do you want answers now? Really some people. Ok, highlight the section after this, as the text should have been set to match the background colour.
S
P
A
C
E
R
1: e. Fast film, fast shutter speed, zoomed in to fill the frame [the bird is very predictable], close focus. Aperture is being driven by shutter speed. Maybe not ideal, but the best of the bunch.
2: b. ASA 100 is twice as slow as 200. So that the shutter speed needs to halve, to 1/60 s. Then it is 2 stops from f8 up to f16, so the shutter speed needs be quadrupled [doubled twice]. So, 2 x 2 x 1/60 s = 4 x 1/60 s = 4/60 s = 1/15 s.
3: a. The shorter the focal length of the lens, the wider the field of vision. Aperture is irrelevant.
4: d. Slow shutter speed means a long exposure. Long exposures are generally bad, as things [including the camera] move, and therefore would be avoided if they could be. f5.6 suggests that the aperture is as open as possible whilst still retaining the depth of field. ASA 200 is fairly normal speed film, but no conclusions should be drawn from its use, as it could just be what is in the camera. So the camera is set up to make almost maximum use of light. Which suggests there is not much. But 1/8 s is not really suitable for use during the dead of night. So it is probably inside a building, either with poor daylight levels, or on a dull day.
5: e. The family start quite close to the lens, and finish a few metres away. Turning your uncle into a soft focus blur is not a good idea, so you need a long depth of field. The house is not big enough to have everyone in focus at near infinity, and you need to choose the aperture which allows the greatest depth of field. Which is the smallest hole, and highest f-stop, or f22.
Comments, queries, objections and corrections can all go in the comments section below. You could email me, but if there's a counter-argument then it's probably that other people know about it.
References
- Spillman R, 1971. ### Complete + links ###
Well that was fun.
Anyhoo,
