Exposure & Metering

For the most part, this explanation of metering and exposure can be applied to any photographic discipline. However nearer the end, I give some tips as to how I meter for portraiture. I think most people tend to stay away from manually metering and exposing a scene because they find it intimidating at the start. So the goal of this tutorial is to demystify what really is a pretty simple idea. I hope some of you find this useful.


So where do we start? Well first off, metering and exposure is pretty easy. Yes! most of our cameras can do one or both of these automatically for us, but to get the most out of your shots, it’s really important to understand what your camera is doing when it’s metering a scene and why it makes the decisions it does.

To begin with, we really need to discuss the term 18% GREY. If we have a surface which 100 particles of light fall on to and 18 are reflected off, then that surface will be a grey colour, referred to as 18% grey or sometimes middle grey. We’ll talk about this further on, but it is pretty misleading to call it middle grey, because exposure wise it isn’t really in the middle of black (zero light) and white (completely blown).

So we know what middle grey is.. but who cares? Well we care because the light meters in our camera live in a middle grey world. Here’s where we need to branch off a little because there are two types of light meter. The first is a reflective light meter and the second is an incident light meter. Both measure light in order to help a photographer expose for a scene, but both have very different methods of metering light and as such both are used differently.

Reflective light meters are what we have installed in our cameras. They measure the light reflected back from a surface/object. I mentioned already that our cameras reflective light meters live in a middle grey world, now is the time to expand on this. If we have a pure white wall, evenly lit and we use our cameras meter, the shot will come out GREY. If we take a pure black wall, again evenly lit (or evenly unlit) our meter will again convince our camera that it’s living in a middle grey world and again the shot will come out GREY. The same would happen if we shot a grey wall. Regardless of the colour, as long as there is one tone and we expose using the reflective meter in our camera, all the shots will come out the same colour – GREY! Anything you point it at – GREY. Anything! To summarise, the reflective meter in our camera tells us how to make the thing it’s pointed at to record as an 18% grey value.

Now we know what middle grey is and we know that our cameras meters try to average a scene out as middle grey. Why on earth is this useful?. The reason is because it’s a constant – our meters behave in the same consistent way, regardless of lighting conditions. If we point our meter at a bride in a really white dress, lit by powerful strobes against a white backgroud and we point our meter at the dress… what will we get?? Yes, we get GREY (good to see you’re listening ). If we take the same bride, in the same dress and we change the lighting conditions and location, we can still meter off the dress and still get the correct GREY exposure. If you’re asking yourself “If our meters make a white dress GREY, then how do we make it white?” then you’ve completely understood what’s been said so far. If you aren’t, time to read back over it. The answer to that question comes later.

Before we move off reflective meters, there are somethings we should be aware of. Because the meters in our cameras measure reflected light, they can be fooled sometimes depending on the surface they’re being pointed at. If you’re metering a dark scene (black cat for instance) the black absorbs more light than it reflects and it can fool your camera into compensating by overexposing. Similarly white, or reflective surfaces (an ocean, water etc) can also give misleading reflective meter readings. Often wedding photographers have an extremely difficult task because they have to meter for a white dress as well as a black grooms suit in a scene – so the camera meter can be thrown off completely. One tip would be to buy and use a grey card. These are essential square pieces of card, coloured 18% grey. Basically the grey card is put in the same light as the subject we’re shooting – we then meter off the card and lock in the exposure values for the correct middle grey reading, remove the grey card and take our picture.

So, reflective meters measure reflected light and they turn what ever we’re pointing them at to middle grey. Incident light meters are subtly different. Incident light meters measure the light falling on the surface of our subject from the light source NOT the reflected light. Unlike the grey world our reflective meters live in, incident metering render any object as its true tone, so black records as black, grey records as grey and white records as white. To take the example of the bride and groom, if we use an incident light meter with the bride as the subject but then replace her with the groom, the groom will be properly exposed. Why? Well because the bride was exposed in a way thay she was rendered as her own true tone, so why not the groom in his black tux? One point I want to hammer home is that when using incident light meters, you have to point the light meter at the light that is lighting the part of the subject you want to properly exposed – NOT AT THE CAMERA

So…. Reflective meters make whatever you point them at record on film as 18% grey, regardless of what they are in real life. Incident meters make black record as black, gray record as gray, and white record as white.


So now we know what metering is, the next thing to understand is exposure. An exposure value (EV) is the total amount of light available, where as exposure (in our cameras) is the way we use that light. To explain exposure, I’ll use a common analogy of a bucket and a water tap, where water is an analogy for light. For a correctly exposed scene, we need to fill the bucket to the brim. If the bucket is not quite full of water, the scene/picture is underexposed. If we over fill the bucket, water spills away and detail is lost, which is equivalent to over exposing a scene, commonly referred to in digital as blown highlights.

We have two primary ways to fill the bucket. We can turn on the tap to give us a very slow trickle of water (narrow aperture/high f-number) and leave it running for a very long time (slow shutter speed). An alternative way would be to turn the tap on wide open (wide aperture/low f-number) and leave it running for a much shorter duration (faster shutter speed). It’s really important to note that both methods end up with a full bucket/correctly exposed scene. It’s way beyond the scope of this article to discuss the implications of choosing aperture and shutter speeds as these often have different artistic affects and implications. But what is important is that we ensure we always fill the bucket, regardless of whether we’re using a huge gush of water or a drip at a time.


So how do we tie metering and exposure together? Well we’ll concentrate on reflective metering as this is what most of us will use in our cameras. Imagine we had a wall that was painted 1/3 white, 1/3 black and 1/3 grey. If we pointed our camera at the 18% grey part of the wall, the meter would give use the correct exposure value to fill the bucket/expose the scene correctly. If we pointed it at the white part, the meter would think you had a much smaller bucket (more light reflected) and would call for a smaller aperture or a shorter shutter speed. If however we pointed it at the black section, less light would be reflected, so the bucket would seem bigger and a longer shutter speed and/or wider aperture is required. All 3 surfaces try to render the subject as 18% grey.

When we look through our cameras and look at the meter. By varying the shutter speed and aperture, we are exposing to make the scene go middle grey assuming that our meter falls on the center value.


One final thing we need to discuss before getting into the practicality of how to put all this to use, is to discuss the different metering systems on cameras. The two main ones I’m going to talk about are Matrix metering (or evaluative for Canon users) and spot metering. Simply put (and it is simply put), matrix metering looks at the range of light in a scene from bright all the way through to dark. It will try and average the light in a scene so as it all becomes middle grey. Obviously if there is a huge range of tones (dynamic range) the camera will struggle, leading to whites being blown or losing detail in the shadows.

Spot metering is far more precise. It measures the reflected light at the specific point your are pointing to in the scene. This allows YOU to decide which exposure values to use in a scene. A classic example to demonstrate the differences between both systems is to put a person standing in front of a window with very strong back lighting. If we are trying to expose the scene for the person, we need to understand how each system works. The matrix system takes into account the very strong back lighting and will tone down the exposure in order to bring the whites towards middle grey. The upside of this is that you will not end up with blown highlights. However the down side is that the subject, which in reality is all we care about, is now going to be underexposed.

However, with spot metering, we can point the camera (usually the focus point) at our subject and meter specifically for the subject. Now our camera tries to bring our subject towards middle grey, leading to a correctly exposed subject, but also blows the highlights. I’d rather make sure my subject is correctly exposed though.. wouldn’t you?


Ok, so our cameras try to change the world to be a neutral/middle grey. The problem is, like DAZ we want our whites to be white.. NOT GREY. Ansel Adams “invented” an exposure system which is called the Zone VII system. It’s WELL worth doing a google on this and reading up on it in more detail. But I’ll summarise the useful bits here.

We already know that we can record a white surface as being grey using our cameras internal reflective light meter. Two stops up from middle grey is white, without losing detail. After two stops, we have pure white, which is equivalent to over filling the bucket, losing detail or blowing highlights. Pure black is about 4 stops below middle grey, which means no water at all in the bucket and no detail recorded. Now we’re ready to take our correctly exposed portraits. So, put our camera into manual exposure mode and change your meter to spot metering. My personal preference for portraits is to shoot WIDE open because I love to make sure the eyes are in sharp focus but throw everything else out of focus, making for very simple, uncluttered portraits. Using the camera, look around the scene, literally moving your camera/focal point, looking for the brightest important part on your subject.

Usually this is forheads, noses, cheeks.. or areas of tight skin. In the cases of backlit portraits, it’s often the hair. Once you decide on the brightest part of the scene in which you want to record detail, using your cameras meter, alter your shutter speed (my aperture is generally locked wide open) so as the meter is now at 2 stops above middle grey. I have now locked in the correct exposure for the scene. As I’m in full manual mode I can recompose the scene as the aperture and shutter speed will not change. Then focus and CLICK

Job done. A well exposed portrait


  1. Your post made a lot of sense, thanks, it really helped me

  2. thankyou very much for this…it has helped me understand metering and exposure much easier than it had previosly been explained

  3. Great post, nice to have the information so clear and simple to understand and is verry useful. thanks

  4. Thanks for the great post. Very usefull for making stronger BW portraits!

  5. For sure it’s one of he best lessons for me in terms of metering and exposure subject . I ahve a question for you : At the end of article were you are talking about 2 f-stops over Middle grey or 4 stops below MG , how do you determine the middle grey f-stop ?

    Thanks ,

  6. Your camera will have a built in meter which will be displayed in your view finger.. it’s a scale similar to the one below:

    < + --|--|--|--|--|-- ->

    Underneath the scale there will be a pointer of some sort. By adjusting aperture/shutter speed/ISO you can move the pointer along the scale, by letting in more or less light. If the pointer is in the middle of the scale, you are now metering for middle grey.

  7. Thanks Ciaran !

  8. great post, thanks. I use a hand held meter measuring incident light.But I’ve been told to point towards the camera not the light source as you say I should. Does the dome on top of the light meter average out the reading?.Will you explain the contradiction here? Thanks

  9. Hi Geoff… I guess start off by looking at this lighting diagram ( http://www.thewonderoflight.com/misc/Strobox.png ). It’s a simple setup, where the photographer is side lighting the subject. This setup is also something you can use to “prove” this concept to yourself. For obvious effect, ensure the light is very close to the subject, so that the fall off is dramatic between lit side and shadow side.

    So, the subject is lit from one side and it is the only light source. If I place the meter on the side of the subjects head pointing towards the soft box, I will measure the light output from that box falling on the side of the subjects face. I always meter for the highlights, so I dial the required aperture into my camera. and take the shot. In this case, I will definitely retain all highlight detail but as the light travels across the face towards the shadow/unlit side, depending on how close the light is to the subject, I will potentially lose out on shadow detail.

    However, if I was to take your approach and position the light on the subjects forehead, pointing back towards the camera, I am now exposing for what ever light is falling on the centre of the subjects face. For sure the dome will adjust for some of the side lighting, but only marginally. If I take the shot using the aperture setting from that meter reading, I would now blow out the lit side of the face (something I would want to avoid)

    Obviously the further away you have your light, the less fall off you will have and the less of a dynamic range across the face you will have, so the positioning of the meter matters less. But the point is here, is that it matters less, so if you place it incorrectly (i.e. back towards the camera) then the inaccuracy of the reading matters less. If you had metered correctly in the first place, there is no inaccuracy. Similarly as the light becomes more diffused and less directional, the direction you point the meter matters less (again because the difference will be small between a correct and incorrect reading).

    I think it’s more about understanding how and what you metering for than just following blind instructions. Rather than doing something because someone told you to do it (including me), you should reallly experiment and satisfy yourself that you understand WHY YOU are doing what you do.

    As an aside, there are times when I certainly do point the meter back at the camera – but these specific cases call for it. So it’s on a case by case basis that determines how I meter my shots. But for a tutorial point of view, for people starting out, I think it’s “safer” to give the advice to point the meter towards the main/key light.

    Hope this helps?

  10. Many thanks Ciaran, you explain this so well.
    I’ve tried this approach so far.
    Meter at the side of the face that has light falling on it and note the reading, then meter at the shaded side using the same aperture as before and the one I want to use in camera. The meter is pointing towards the camera both times. Then I average out the 2 time readings. It kinda works but as you suggest I’ll experiment some more.

  11. Averaging out readings can work at times, but it does rely on two key elements.

    First and foremost, the dynamic range has to be low (i.e. fall off has to be very small between lit and shadow side). If I was to meter the lit side, for f11, the shadow side for 5.6 and used an “average” of f8, there is still a stop difference between my average and my highlight.. so I would still blow the highlights using an “average”

    Secondly it relies on people being able count in stops and being comfortable averaging them. My experience is that most people struggle with the concept of what a stop of light is, let alone being able to count and average them πŸ˜‰

  12. I’m feeling really ignorant cause everyone gets it but me. Where I am lost is that you are metering from the existing lights. I use cool lights (Flash) How can you meter if the flash isn’t going off? If I meter for existing light, then take the shot, it’s all blown out from the flash tubes. I don’t have an external meter. I use the built in meter from my Nikon D90. I was using Matrix metering but I can understand now why you would use Spot instead. So, what I have been doing is trial and error but according to my built in meter, I am WAY underexposed. I just ignore the meter and shoot with the setting giving the better result. Usually, f/3.5 and 1/125 at ISO 200. Please explain how to use internal meter with Flash. I have two Opus OPL-L150 flash heads and they are triggered via my SB600 flash unit mounted on the camera and I have my subject behind a Storm Cloud Muslin. The Opus heads only have adjustments for flash power output by percentage I guess.

    Thank You for all of the insight and help. It was a great article and has helped with everything except the Portrait stuff.

    • Either increase your flash or change your exposure. Fire a few test shots till you get it right. This way you don’t have to rely on technology and you’ll get a natural feel for it after a while.

      • You probably missed the bit where the article doesn’t discuss the use of flash at all? And as for changing your exposure.. that’s pretty much what the article does discuss. But thanks for the valuable “take it and see” advice.

  13. A Continuous lighting system is called “hot lights” because they are on the whole time and they end up getting hot. Flash is not known as a cool light, but in fact a strobe!

    And the reason for your confusion is that your in camera meter is an ambient light meter, not a flash meter. It can only measure light that is there all the time. A strobe, once fired, only happens for a fraction of a second and this output can not be read by your cameras ambient light/reflective light meter. If strobe is the way you intend to light most or indeed a lot of your shots, then you need to invest in a hand held flash meter. I suggest you read my Flash 101 article, which maybe of assistance?

  14. I understand the cameras meter. I know what to look for. I amlearning to just use manual mode.

    My question is how many people use a gray card or other items to set their exposure?

    I have had a few people tell me that gray cards are a thing of the past that they were meant for film cameras.

    To just set my exposure by pointing it at subject and take the photo then look at it and check the histogram.

    Is this correct is this really how I should be setting my exposure??


  15. It’s definitely disingenuous for someone to say that gray cards were meant for film cameras ❗ The metering on todays digital cameras is largely the same as the film cameras of yesterday, so the same theory and indeed practice applies. One obvious advantage of digital is that you can afford to take test shots, so do not need to nail exposure. This leads towards people taking an image, checking a histogram, making adjustments if necessary and taking it again.

    Whilst this is a perfectly valid way to work, so is obtaining the correct exposure the first time. A gray card is just one way of ensuring you do this. Another way is by using a hand held incident light meter and a third, is understanding precisely how the reflective meter in your camera works as well as understanding what the dynamic range is in the scene you’re capturing and the part of that range you’re interested in capturing.

    So if you understand the cameras meter, then you do not need a gray card, but similarly you do not need to check the histogram.

  16. Thank you so much. I’ll start again taking pictures. Millie

  17. Hi, brilliant post thank you, I too like to shoot wide open for portraits (i’m on a sony system) Can I presume that instead of altering the shutter speed once finding my bright point I can use my exposure compensation toggle which covers two stops? How would this system work for bride and groom (very contrasty situations) lock on the dress?

  18. Exposure compensation “fools” your meter into believing something is darker or brighter than it actually is. It basically says: Take the real reading, then compensate by a certain amount in a certain direction. So in reality there is no value using it here. If your aperture is fixed and assuming your ISO is also fixed, the only setting left is the shutter speed. Whether you set it manually or do it by means of compensation, the result will be the same.

    As for the white dress and black clothes on the groom, yes you meter off the dress. Assuming it’s not too bright a day and the dynamic range is within the range of your camera, everything will be fine. If the scene has a range much bigger than the camera, then the dress will be exposed correctly, but the grooms clothing could be blocked up.

  19. You just took me from:

    1) Take test shot.
    2) Check for blinking highlight warnings.
    3) Dial down by “feel”.
    4) Go back to 1.

    To doing it right in the first place. Dialing in a highlight to +2 with my in-camera spot meter works wonders! Wonder why I never either heard that before or figured it out… It’s a funny feeling realizing that although I thought I understood something, I actually didn’t really get it at all.

    This really helped me a lot. I can choose exactly which highlights I want to expose and which to let go- without wasting time with trial-and-error. Thank you!

  20. In the digital world, there’s nothing wrong with taking test shots until you’re happy with your exposure. Taking an image is costs nothing, so this is a valid approach.

    But you’re right, there’s no substitute for doing it right in the first place πŸ™‚ I’m glad it helped you.

  21. very good

  22. Great explanation, thanks

  23. Hi, I understand the concept, I haven’t tried it yet though but I wondered if this works for all scenes not just portraits? For example if I am shooting at the beach and the reflection off the water is the brightest do I spot meter the water at +2 and then recompose? Or are there different values to set depending on the scene, time of day, sunrise, sunset, etc…?

  24. Yes, exact same concept for landscape and time of day doesn’t matter. Meter off the brightest point that you want to retain detail in and use that as your white point.

  25. How about if the sun is in the photo, I’m sure that would be the brightest area and I didn’t think you should point the lens directly at the sun and it’s not something you should probably look at directly in the viewfinder. How do you handle that situation?

  26. Whilst taking a picture, it is bad practice pointing the lens into the sun (or any direct light) as it will result in lens flare and lack of contrast. But for the purpose of metering it’s fine to do so (watch your eyes!!)

    A good tip I once heard… if you can not comfortably look into the sun, then you won’t be able to photograph it – it will just be too bright. So if it’s in your scene, chances are you won’t be able to control the exposure anyway.

  27. So, if I have a 18% Gray card… Can I use that in place of the last option of spotting off a bright spot on face? So, I’d point at grey card center the meter remove and shoot?

    Thanks guys/.

  28. Yup… That would work perfectly fine

  29. Thanks Ciaran for the awesome info! Have any more you’d like to share?

  30. Nice information, thanks.

  31. thanks for this! it was super helpful.

  32. Thanks Ciaran, great explanation!!

  33. Hi there, I’m not sure if your still responding to this article. Anyway what if I’m taking a portrait of someone under incandescent or in low light – what if the brightest highlight was already mid tone? This would overexpose by 2 stops – how would you meter here?

  34. If the brightest point in your shot is midtone, then the shot is under exposed.

  35. Sorry I disagree – if I take portrait shot of someone in a living room with tungsten lighting and I want the exposure to match what I see with my eye- keeping any shadow detail – and the low light levels – your method would over expose.

    Exposure can be very subjective – I might want to snap my girlfriend by candle light -this does not mean she is under exposed – only if you think like a camera meter and want her brought up to middle grey. You could then set your white balance to 2500k – and then she would look like she was in broad day light – hardly capturing the ambience of the moment.
    The way I see it metering off the brightest part of someones face means that part has to be just that – bright -or to be more precise, 2 stops above middle grey to get the right exposure. For that the light hitting your subjects has to be constant, just like the middle grey world.

  36. I really don’t care if you agree or not.. it’s fact.

    If the brightest part in your shot is midtone, then the shot is under exposed. Of course, you may want the shot to be under exposed and that’s where subjective taste comes in.

    The brightest point (of interest) in a shot, should not be two stops under.

  37. To qualify this. If you want the brightest part of your image to be grey, then expose the shot exactly as you describe.

    Assuming you’re shooting JPG (or will end up as JPG), you’re only using 4 bits of information, giving just a possible of 16 shades of grey if mid tone is the brightest point. For me, I can certainly differentiate (by eye) more than that, even in low light.

  38. Hi. Really hope you are still answering questions on this, Ciaran! I was so excited to read your tutorial on this, and tried this technique out today. I set my camera to manual and spot metering, and with my aperture wide open I searched for the brightest important part of the scene with my focal point. I found it on my little girl’s cheek and added the 2+EV. Although it was definitely brightest on her cheek, it wasn’t HUGELY bright, just brighter than the rest of the scene. It was overcast, weather-wise. Took twenty pictures, and all came out way over exposed. Took some more with just +1 EV and they were better. Took some on 0EV, middle grey, and they looked fine. Feel really confused and frustrated as I just can’t seem to get the hang of metering, spot or otherwise. Matrix is okay until I want to shoot in a wood on a sunny day or the beach etc. Just would DIE to be able to pretty much guarantee the right exposure every time. Please, any help would be so appreciated x

  39. Hmm… it depends on what you mean by adding 2+EV??

    Let’s assume ISO is fixed (at 100 as an example.. but could be any ISO).
    Camera is in manual
    You have set your aperture wide open and this is now also fixed (as you’re shooting in manual)
    There is no exposure compensation set on camera.

    So now, the only control you have over actual exposure is shutter speed.

    With spot metering, point your camera/focal point/meter at the brightest point (your daughters cheek). Now using shutter speed, dial in a value/speed where the meter is 2 stops above middle grey.

    Be careful! Most cameras only have meters which show 4 stops (two stops under and two stops over middle grey), so it wont show the difference between 2 stops over or 3, 4, 5 or even 6 stops over.

    As a tip. I choose a shutter speed that meters for middle grey and then six more clicks (assuming 1/3rd of a stop increments) and you’re there.

    When people say EV, I tend to think exposure compensation? I’m not sure if it’s that you’re referring to?

  40. Wow, thanks for your speedy reply x Really new to this, and am learning so much stuff that I find myself dreaming about it! I selected f3.5, ISO 100, and pointed the camera at my daughter’s left cheek. Adjusted the shutterspeed to be +2 on the meter, it read 1/250, i lock this exposure, then repositioned and pressed the button. I thought I was being clever saying +2 EV, hehe, obviously using the wrong terminology! The arrow was definitely pointing to the 2 and hadn’t gone over the meter limit. Since my last message, the sun has come out and I’ve taken a few more pictures, this time metering off the bright sunlight on my girl’s hair. I spot- metered and added the +2 and this time the exposures were perfect. YES! SO I am thinking, due to my previous set of unsuccessful pictures, that if the exact spot I am metering off is not very bright nor very dark, it should be okay for me to lock the meter at 0 to +1 and if it’s something dark I want to focus on then I set the meter to -1 to -2?! Kind of calculating the different tones in your brain and working it out from there. Taking photos from each little 1/3 on the meter and tons of practice should help me get my head around this eventually! I find it fascinating that you could take my camera now and get a beautiful, sharp, creatively correct picture of my daughter, while mine are pretty average! Makes me strive to learn and learn x

  41. Glad it’s working πŸ™‚

    One thing to watch for is that with spot metering you have to be pretty precise where you’re pointing it. If the bright point is small in the frame, it’s easy to point outside it, which will throw the exposure readings off.

    And yup, you could also apply this to shadows like you say. But generally we care about retaining detail in highlights rather than concerning ourselves about blocked up shadows. Also true black is actually 3~4 stops below ,I’d grey.

    The best of luck with your journey πŸ™‚

  42. Thanks, dude. One final question if that’s okay… Do you use spot metering all the time? Incidentally, just read your blog entry about losing focus, and think you would make a brilliant teacher to eager little newbies. You explain things so clearly, makes it much less daunting. x

  43. No.. I don’t use spot metering all the time.

    I only use it if I’m shooting natural light (no flash – in which case I’ll use an incident flash meter) and even then, only if there is a lot of contrast in the scene.

    If there’s not and the dynamic range is will within the capability of the camera/sensor, then matrix metering works just fine.

    Also, if I have an assistant on the day, I’ll often get them to use a hand held incident light meter and forget about in camera metering all together.

  44. Thanks and Very good explanation and easy to understand.

    I have been reading a lot about photography in recent times and practising it at the same time.

    the one that i read recently is about zone system by Ansel Adams, pretty good one and interesting thats its valid in digital world too.

    But i have a question here, relating your explanation here, with zone system, when we increase / decerease the f-stop, according to zones, the exposure is affected, then the final image is affected and its overexposed or underexposed by the f-stop numbers. relating it to your example of bucket with water, in this case it will be half empty or overflowing.

    how do we treat this, i am not able to co-relate the concepts here.

  45. The points are explained very well but I think the author might be confused about what “18% grey” actually means. An 18% grey card is firstly “neutral grey” because it reflects all wavelengths of visible light equally. This defines it’s color. It is second “18%” because it reflects only 18% of the light which lands on it, regardless of what wavelength that light is. The cameras metering system actually DOES NOT CARE about the color of the card. It is only the LUMINOSITY of the card which matters to the metering system. In other words, it matters that it is 18% but it does not matter than it is grey!

    The fact that is is neutral grey becomes important when setting white balance, but it is completely irrelevant to the metering system!

    When the author says, “To begin with, we really need to discuss the term 18% GREY. If we have a surface which 100 particles of light fall on to and 18 are reflected off, then that surface will be a grey colour”, this is untrue. While it is true that an object reflecting 18% of the light which lands on it will have 18% luminosity, it is untrue that this will define it’s color. It’s color will be grey if it reflects all wavelengths of light equally – simple as that.

    • Sorry, had to chime here. You ended your reply with “simple as that” when in actual fact your rather technical and unnecessary reply will only serves to complicate and to confuse people. Ciaran’s reference to 18% grey is widely used AND accepted. To introduce luminosity and what the the camera “does not care” about is superfluous.

      • How does correcting incorrect information count as “superfluous”? The meter *doesn’t* care what color the item is as it *is* only looking at luminosity. I could have an 18% pink card and it would still meter the luminance correctly.

        Additionally, saying, “Regardless of the colour, as long as there is one tone and we expose using the reflective meter in our camera, all the shots will come out the same colour – GREY! Anything you point it at – GREY. Anything!” is incorrect. I have beige walls and I metered my test shot and guess what: the wall was still beige in the test shot.

        In order for that to happen, you’d have to discard the color information present, and the metering system just doesn’t do that. What is true about this statement, and perhaps is what the author meant (I haven’t read the site so maybe this is a black/white photography site? I came across the article while looking for exposure information), is that if I meter for an item of any color and record as or convert to black and white, the item I metered should come out 18% grey if it was metered “correctly”.

        But once you get away from 0% luminosity (black) or 100% luminosity (white), then all sorts of colors come into play and they’ll only record as grey if the item actually is grey or your color balance is wrong. (For an idea of how this works, play around with the HSL converter at http://serennu.com/colour/hsltorgb.php – leave your lightness at 50% and change the Hue/Sat values: all sorts of colors, all 50% L.)

        So yes, the only advantage of an 18% *grey* card rather than an 18% pink or blue or purple card is for color balance; for luminosity, the color doesn’t matter a lick.

  46. Hi Ciaran
    I understand about different lights giving off different values etc and I understand how to under and over expose through hours of trial and error. I am still learning the technical side of my photography but the one thing I do not get is this “grey card” thing. In your example you say the camera sees all grey, understood. Then you say later on to use grey card next to white dress and focus on the card then take photo of dress and it will work. I do not understand this one bit. If the camera sees grey, why focus on grey? Very confused beginner, by the way I prefer night photography(see flickr account) so metering is something I should fully undertsand really!!

  47. wonderful…. tutorial as well as the comments, questions and answers πŸ™‚

    • Camera, incident and spot meters only see in black and white (and the 254 in between shades if grey)

  48. might be a good read … but white text on a black page is stupid, it ruins eyes for next page in normal white.

    Change it to white with black text.

  49. Hello

    I have read so many articles but still struggle with metering. Would you be able to help me out further?

    I understand that if you have a high contrast scene the camera cannot expose for the highlights and the shadows and you must choose. It made sense that a backlit portrait, even if the sun isn’t in the photo, we must override and add exposure to expose for our subject’s brightest part. If we use evaluative metering, do we just take the camera’s suggestion and add +2 …. Eitherwise how does the camera know what we are pointing at? Or use the spot meter directly on the brightest spot and lock the correct exposure suggested without overriding?

    Now, what if we move our subject into shade or it is an overcast day. The light falling on our subject is equal to the overall scene so will the camera metering still be confused? Will we need to override?

    What if our subject is filling the frame versus not filling the frame, how does this affect our metering whether in shade, overcast day or backlit?

    Wedding day. …. Same questions as above. Will the dress always come out as grey whether it be backlighting, shade or overcast day? What if the bride fills the frame versus doesn’t fill the frame? How do all these variables affect the meter using evaluative?

    Thank you so much for your help. I just keep spinning my wheels on this topic.

  50. You’re over thinking things πŸ™‚

    Long story short.. a camera has a limited dynamic range.

    If a scene has a certain amount of contrast – a good range from black through to white, but is within the range of the camera, then the exposure should be perfect – the whites should be bright, the blacks should be black.

    If the scene has too much contrast, then the camera either has to sacrifice the highlights, or the shadows, or a bit of both. Generally it underexposes so as to retain highlight detail, so the shadows are sacrificed and become blocked up.

    If you move a subject under a tree or into shade so that there is very little contrast, then the shot will be grey. It always tries to turn things grey if it can, or find a compromise as close to it as possible.

  51. Hi Ciaran, Just an update on recent Matrix meters in Nikons. They are colour aware and the very latest cameras can identify skin tones and will try and expose for them if they are under the focus point. I shot a wedding and the guy in the dark salmon suit kept being overexposed until I realised that the camera thought it was an art-nude shot. πŸ™‚
    As I understand it, the camera measures the brightness and colour at many different points in the scene. It weights the importance of the focus point and distance and compares to 300,000 known shots and “estimates” the final exposure. It also still seems to give a nod towards protecting highlights but less so than earlier matrix meters.

  52. wonderful article

  53. If metering in manual mode is difficult, try going a head and putting your camera in aperture priority (Av) mode. This way you can “lock” the correct exposure when metering off of the brightest highlights. For example, taking a portrait of someone under natural lighting and the sun is high lighting the side of their face and hair. If it’s a large enough area, then use spot metering.

    If you’re metering off of the brightest high lights, the subject will be under exposed. This is where exposure compensation comes in. Its that (+/-) button on the back of the camera. Usually over exposing by +1/3 or +2/3 stops using exposure compensation brings back the details in the dark/shadow areas. However, you must pay attention to how bright the subject is you’re metering off of. If you’re metering off of the clouds, or say 15 degrees right or left of the sun, expect to have a silhouette; trying to use exposure compensation will not do you any good because the shutter speed you’re locking in will be way to fast to bring the subject back to a correct exposure. After locking in the correct exposure for the brighter highlights; usually by pressing in the (*) button on the back of the camera for canon users, then recompose, focus and take the shot.

    Consult your manufacturers handbook on where to find the AE/AF lock settings located in the menu. It works wonders.

  54. Thanks for the time and effort to explain this topic!!

  55. Thank you for such a clearly written article!

    • You’re welcome. Glad you found it useful

  56. Thank you for the fantastic article.

    One question:

    Upon googling, it seems there is some confusion about camera manufacturers (specifically Nikon) calibrating the on-board meters to ~12-13% luminance (not reflectance). Does this mean that both my Nikon D700 and D800E have on-board light meters that are calibrated to ~12-13% gray? And if so, what does it mean in terms of the process you describe? Does it mean I need to compensate by opening up 1/2 stop to keep everything correct? i.e. my hottest highlights that I want to maintain detail would be metered at +1.5 stops as opposed to +2 stops?

    Before running into the ~12-13% threads I noticed my shots seemed overexposed on my D700 but I’m also quick to consider user error. I should mention I use fully manual mode, manual focus, with spot metering — none of the fancy matrix multi sample 3D color tracking stuff.

    Thanks again Ciaran; and thanks in advance for any info you may have on this.


    • For sure, different camera manufacturers have a slightly different idea for what middle grey is. I can’t comment with any certainty on what value Nikon (or other manufacturers) use.

      My guess is, if you’re using spot metering, the over exposure is more user error than camera error and isn’t directly linked to the fact that the camera does or doesn’t use 18% as middle grey. I find spot metering error prone simply because the size of the frame being used for consideration in exposure, is so small. It’s easy to wander away from the point you’re ideally trying to meter from. When I use spot, I get REALLY close to the subject I’m shooting, filling a large percentage of the frame with the area I’m metering for, lock in exposure, and then back off for the actual framing. As long as the light isn’t changing, this method works fine.

      As for 1/2 stops, normally most people go in steps of 1/3rds – certainly I do. If you’re finding you’re always over exposing, then it’s definitely worth either dialling in some exposure compensation, or compensating manually for it. As a Nikon shooter, I’ve never found the need to do this myself.

  57. Hi Ciaran I’m going to be shooting a family who will be wearing White Shirts and Black Pants, in this instance do I still Meter off the white shirt and add 2 stops using my shutter speed or will this result in the shirts being white but the pants being grey? there will also be a lot of white wooden posts in the background as well.

    • It depends entirely on the contrast in the scene! The more contrast there is, the more black the blacks will appear if exposing for white.

      If it’s lower contrast, then they’ll appear muddier. I would always expose for the highlights, ensuring I retain detail there and then deal with boosting contrast later if it were required.

  58. Great article!! Easy to understand

  59. Nice, thank for share it,, I am 100% manual and use spot metering all time..

  60. Before making my comment, understand that I have been a professional landscape photographer for almost fifty years. I’m shoot almost exclusively in large format (4×5″ and 8×10″) and I understand exposure very well. I’m a Zone System photographer so I have to. I’ve also written magazine articles on related subjects involving metering, spectral response in reflected light meters, etc.

    That said, your article is good but you made one surprising mistake: Colored objects will NOT turn out gray as you say. Color is not affected by improper metering technique except that it will come out lighter or darker. However, the density of the film will come out the same no in your example assuming the meter you are using is good spectral response. Many, if not most, reflected light meters, especially some spot meters, unfortunately, do not have good spectral response. That is, they are more sensitive to some colors and less to other colors. But that’s not the issue here.

    It’s true that if you shoot a white wall, a gray wall, and a black wall while using the exposure the meter recommends that all results will turn out to be middle gray. But that’s because the density is the same. Color, on the other hand, doesn’t change except to become more or less saturated. For example, say you have a very light red wall, a middle red wall, and a very dark red wall. When metering as in the above example, they will all turn out to be middle red. Well, they may not match exactly but the red will still be there.

    Here is where you go wrong:

    “Regardless of the colour, as long as there is one tone and we expose using the reflective meter in our camera, all the shots will come out the same colour – GREY! Anything you point it at – GREY. ”

    Think about it: If we do the test you mention first, then you are, in effect, saying that all the pictures will turn out grey. They won’t. The red wall with approximately 18% reflectance will come out red just as it was in reality. This makes it obvious that your above statement is incorrectly. Going on, if we shot a very light red wall, it would come out middle red. And if we shot a dark red wall, it would come out middle red. They would be lousy reds but, nevertheless red.

    The point is that 18% gray applies to the DENSITY of the film, not the color the exposure produces.

    But here’s something can cause a food fight among photographers because probably not one in a hundred know this: Reflected light meters are not calibrated to 18% gray. In other words, they are not calibrated to assume that what they are “seeing” reflects 18% of the light falling on it. That’s a myth. Reflected light meters are actually calibrated for approximately 11 to 12% reflectance. If you push an engineer for a company that makes cameras and/or reflected light meters, they will admit this. The 18% gray card was really made for the graphics arts industry. Photographers just took to it because it’s a known standard and therefore very useful. It’s also usually close enough. The difference is explained by the little known K-factor that is built into reflected light meters. Ansel Adam’s alludes to the K factor in The Camera but even he didn’t fully understand it. Besides, when calibrating to the Zone System, you are taking all variables into account and it makes no difference what he meter is calibrated for. You do film tests to find your proper exposure index. As long as a meter has linear response, it makes no difference how it is calibrated. A photographer may have two spot meters, for example and he uses the same film/developer combination but one meter must be set to E.I. 400 while the other meter must be set to E.I. 100 to get the same exposures. In other words, the tests eliminate the K-factor and other variables.

    Somewhere I have an old article from, I believe, the long defunct Camera & Darkroom that explains how the myth of the 18% gray calibration in reflected light meters came about.

    Other than the error I’ve mentioned, your article is good. In fact, I came across it when looking for articles explaining how reflected light meters work and how they can be fooled by very light or dark subject matter that I could pass on to a photographer who is new to this whole idea.

    • I apologize for the typos in my response above. I blasted the comment out very quickly but skimmed over it after I posted it and saw my typos. I hate that! πŸ™‚ I hope it’s understandable. Also, I want to expand on a point that I made in more detail. If you shoot a very light red wall, a middle density red wall and a very dark red wall, they will, as I said, not come out gray. But they will not be the same. The very light red and very dark red will come out in sickly reds but red nonetheless. And, again, as I pointed out, if different exposures of red always come out as gray, as you say, then the properly exposed middle red would come out as gray too. Obviously, that’s not the case. The colors will remain for the other two but they will be darkened or lightened. The red with 18% reflectance will come out as you see it, not gray, assuming your film records colors accurately. In short, exposure affects density, not color. You don’t change a color by merely over or under-exposing but you do change it’s density. Think about this: If what you are saying was true, various parts of a “normal” scene having a wide range of reflectances would come out gray. As you know, it doesn’t. Color photography wouldn’t work if it did. For example, suppose you shot a seen that had various objects of the same exact color – say red – but some of those objects are in full sun, others are in open shade, and yet others are in full shade. A garden would be a good example. Since there is only one exposure made, the red flowers in deep shade would receive far less exposure than the flowers in open shade. The flowers in bright sun would receive far more exposure than the flowers in open shade. But did any of them turn out gray? No. The only thing different in the three groups of flowers is differing exposure, just as in your example.

      Again, good color photography would be impossible if what you said was true.

      Keep in mind that I’m talking about film.

  61. Wow, thanks so much for the great explanation! I have been researching the concept of middle grey and your post made perfect sense. I think I got it now. Thank you for taking your time to explain this concept in a simple and clear way so a beginner can understand.

  62. Spot meter does not work for off center subject cause while you recompose your exposure changes. Is that true. Expouse lock and focus might work but that is another step you have to keep in mind when use spot meter.


  1. Shoots - Page 2 - [...] the meter lock, and where to find your histogram. Understanding Camera Metering and Exposure Exposure & Metering…

Care to share your opinion?