Understanding Manual Exposure

<p>I recently read a good book by John Shaw "Nature Photography Field Guide" which indicated that if you are shooting in manual mode using spot or CW metering you should open up one stop if your subject is lighter than medium gray. John pointed out that the cameras TTL meter works in this way. So, over the weekend, I did some shooting in manual mode (e.g. a red covered bridge at midday) and noticed that opening up one stop overexposed all of my images. I understand that this book may not be the most current and was not written for DSLRs, but don’t exposure techniques remain the same? Has TTL metering changed and improved over time so this technique is no longer valid? I have a D90 and I am trying to improve my images. Thanks for your help!</p>

<p>The problem is likely that you did not meter what you thought you were metering. The advice to open up a stop is still very much valid. Nothing has changed with the DSLR.<br>

<br /> Can you post the example photo?</p>

<blockquote>

<p>if you are shooting in manual mode using spot or CW metering you should open up one stop if your subject is lighter than medium gray</p>

</blockquote>

<p>I don't have John Shaw's book with me, but I doubt that he said exactly that. Your subject can be lighter than mideum gray by various amounts; you cannot always open up by exactly one stop.</p>

<p>Shaw's books were written during the film era. Today with digital cameras, you are much better off using the histogram and blinking highlights to evaluate exposure. Those new tools were not available when Shaw wrote his books.</p>

<p>I happen to know John Shaw a bit. He is very up to date with current techniques.</p>

<p>Robert, nothing has changed! You just may not be metering appropriately to use this technique. Firstly, matrix, centre weighted and spot metering are all TTL on a modern SLR. However, you need to be careful in manual mode as to "what" you're actually metering. A centre-weighted metering of an average mid-tone scene (regardless of what your bridge might be) is still an average mid-tone scene so no need to adjust. As well, is your red bridge really +1 stop over mid-tone?</p>

<p>Spot metering is probably a more accurate way to perform this technique. Meter the element in your scene then adjust. One approach is to meter the brightest part of the scene you want detail in then open up 2 stops. I still shoot negative film so I tend to meter the darkest shadows I want detail in and close down 2 stops.</p>

<p>When using centre weighted you do need to be careful as to what the camera's meter is actually evaluating. If its predominantly sand, for example, then opening up 1.5 stops will probably work. Or metering at dusk and wanting the scene to look like dusk, you can close down a stop.</p>

<p>I have found in the past, John Shaw explains exposure very well. Film or digital its all the same. An old technique is to meter the back of your hand (assuming your sporting Caucasian skin) in the same light as your subject then open up a stop. You just need to really think about what your camera's metering mode is seeing; what is a mid-tone then adjust from there. Don't try this in matrix mode as that mode is trying to get a little clever and your better off using exposure variation to achieve the effect you want</p>

<p>I suspect what John said (or meant) was that if you're metering off something (presumably the subject) and it's lighter than medium grey, you need to over-expose it relative to the value the spot meter will give you - the spot meter will tell you what exposure you need to make the thing you're pointing at look mid-grey (ignoring colour), so if you're pointing at something brighter than mid-grey you have to compensate. I've heard it suggested that caucasian skin is typically a stop lighter than mid-grey, which may be whence the "one stop" came (but on a DSLR you can be more precise). If you point the meter at the scene in general but you consider "the subject" to be the paler parts it, the meter will probably not need correcting - or at least, not need correcting in so predictable a way. I hope that helps.</p>

<p>That is a good book and the exposure section is still valid. While it is true that today's technology give sus more tools, a good understanding of exposure will go a long way in using and understanding those new tools.</p>

<p>The camera may have metered something other than what you thought it should meter and/or it was more than 1-stop off from a medium tone. Keep practicing.</p>

<p>First and foremost - the advice given has NOTHING to do with shooting in manual; it is equally valid for automatic modes. That is of course only if you follow your meter's suggestion when in manual mode and zero your meter to its suggestion. Second, I strongly doubt that the advice was to always "open up one stop" as the amount will vary depending on the scene. Third, matrix metering is a bit of an exception here as it tries to evaluate the scene and may well suggest a correct exposure already without a correction. That's why it is sometimes hard to predict a metering result in matrix mode.<br>

In any case, with a DSLR you have a major advantage over film: the histogram. It tells you the distribution of luminance for the entire image and you can make educated decisions on how to set/change your exposure.<br>

Using manual mode can be of advantage - but the same amount of flexibility is given in any of the automatic modes by dialing in exposure compensation; these are just two different routes to the same result. If, for example, your meter suggests f/8 and 1/125s, then there is no difference whether you set this exposure manually or let the camera do it (actually, there is as in manual mode the camera might be restricted to third, half, or even full stop increments whereas in automatic modes, usually any value is possible). If you now want to add one stop you can either manually set f/8 and 1/60s (or f/5.6 and 1/125) or tell the camera to expose +1EV by dialing in that compensation when in auto mode. Just be careful when using exposure compensation when in manual mode - the setting applies a bias to the meter.</p>

<p>In today's world, the camera gives you all those options: spot, center, matrix; A, S, M and even P. None is better than the other, they are all just tools. Personally, I never use P as it is the hardest one to interfere with - but to me A and S and M are equivalent (and that's why you almost never find me in M mode as I can make the same adjustments as easily when in A or S mode). I am also very comfortable now with the matrix system and hence never use center and very infrequently feel the need to employ spot.</p>

You said that you opened up a stop, but from what initial exposure value? And how did you calculate that initial value?

 

If you calculated your initial exposure using the camera's internal meter, and if the bridge was very light in color, then

adding an extra stop of exposure might make sense. If you calculate the initial exposure using the Sunny f/16 rule then

opening up an extra stop makes no sense at all. The extra stop is a fudge factor that corrects a "wrong guess" by your

meter when you point it as a highly reflective object. If you did not calculate your exposure using a reflective lit meter, no

fudge factor is needed.

 

Other tips:

 

Use the blinking highlights display to warn your of overexposure.

 

Use the histogram to warn your of overexposure or gross underexposure.

 

If the shot matters to you, bracket your exposures. I.e. in addition to what you believe to be the correct exposure, take

one or two "too dark" exposures and one or two "too light" exposures. If your main exposure is off, bracketing gives you

insurance.

 

And don't let anyone tell you that bracketing is unnecessary or for dummies. Jay Maisel brackets his shots. If it's good

enough for Jay, it's good enough for you and me.

<p> <br>

Thank you all for your responses which have helped. I hope that I will eventually be able to reprogram my brain to think that if a scene is overly bright (sand, snow) you need to add exposure to what the TTL meter is reading. While I understand how the meter works, this just doesn’t come natural to me or seem logical. Histograms to me is a technology that provides a good guide but is in no way a guarantee of a correct exposure (even if it is bell-shaped peaking just right of center!). So, I think bracketing is what I intend to use, especially if I am shooting a distance from home, and until I gain more experience.<br>

I do have a follow-up question: I shoot a lot on the water say of a wood duck filling one quarter of the frame (300mm). How does water affect the TTL meter and how do you compensate for this? Thanks again.</p>

<blockquote>

<p>so this technique is no longer valid?</p>

</blockquote>

<p>that technique is never valid, and no technique is right for all situations. If you don't know when it's right and when it's wrong, then you shouldn't be in control. Feel free to practice to figure out when it's over and when it's under (and by how much); that's when you can start using manual mode</p>

<p>In manual mode when you are using the spot meter the meter tells you how the camera sees the brightness using your current iso, aperture and shutter speed settings.</p>

<p>0 = right in the middle means that whatever you have spot metered will be as bright as middle gray in the photo.<br /> +1 = one stop brighter than middle gray, so light gray or so.<br /> +2 = two stops brighter than middle gray, which is a light light gray.<br /> +3 = three stops brighter than middle gray, which is usually white.<br /> -1 = one stop darker than middle gray.<br /> -2 = two stops darker than middle gray.<br /> etc, etc.</p>

<p>So if you point it at a white swan and you want the swan to be properly exposed with detail in the feathers you need to change your settings so the spot meter reads somewhere between +2 to +3.</p>

<p>When you have dialed in a setting you can move the camera around and see how the meter changes depending on what you point it at and that will tell you how the brightness of the scene looks.</p>

<p>PS. You absolutely need to disable auto iso if you want manual control.</p>

<p>If you are going to use spot metering in the manual mode, one of the best books you can buy is the Practical Zone System by Chris Johnson. That blanket statement that you should open up one stop on something lighter than middle gray is very ambiguous. Is the value Zone VI or Zone VII (two stops above 18% gray). If you open up by only one stop, you are still overexposing by one stop.</p>

<p>The principles outlined in this book are (generally) applicable to black and white or digital exposure with the caveat that you cannot "push" or "pull" a digital exposure in the same way you can with a black and white negative</p>

<p>My 2c. <br>

After getting into digital and now enjoy slide film, I have been using spot meter, with digital I was too lazy, hey why spot meter when I could take 2 sec to take the picture and evaluate the histogram, haha. </p>

<p>In terms of metering, I have shot a good couple of films side/side now with my digital, using the same settings, you can use the digital to teach yourself. </p>

<p>Re: the palm of the hand, I read that book recently from the library and I think he does mention over a stop bigger on the lens. But really, metering is really subjective to the person, I find it easier to just trial and error, get a digital and spot meter at diff areas in the photog you enjoy. Like I do landscapes so you spot meter in diff areas and learn from that and try diff adjustments out. I also be careful with the sky, some books do say meter at blue sky but the sky can have so many diff tones (so does green for that matter). Also depending on the scene sometimes you may want to overexpose if it does not incl sky like water streams. Night photog, all out the window I personally spot meter somewhere *I* think is grey and +1 (like street lit). If in deep shade maybe +2.</p>

<p>I learned exposure from the very John Shaw's book and I found what have been said in the book are great practical guides. Be sure where You are spot metering and you may like to check more than one point in the scene, then determine how many stops (or fraction of a stop) you want to adjust. Be careful about strong color. Bright red may not be really lighter. 2 cents from a newbie using spot metering and M mode exclusively for more than two years on a D200.</p>

<p>Regarding the duck...if you are spot metering the duck and your area of coverage includes only the duck, but not water, and the duck is a typical mallard, just use your meter reading. If you have water included in the metering area, you need to know whether the sun is reflecting off the water yoward you and affecting your exposure, or if the sun is at your back and creating "black" water. Best advice, since you're using digital rather than film, is to bracket a couple of shots, check your histogram until you get a feel for how to compensate.</p>

<p>This is the standard problem with using reflectance metering. You have to second-guess the environment (almost always wrong unless you have lots of experience), or meter from a gray card.</p>

<p>Simple solution: Use an incident light meter and forget about reflectance. Use the exposure shown by the meter. That exposure should render the scene as you see it. You can increase exposure a bit to bring up shadow detail, or decrease exposure to improve detail in highlights.</p>

<p>Any old incident meter will work. You don't need the latest and greatest hi-tech monster.</p>

<p>If you buy a used one, check its accuracy using the "sunny 16" rule. Set an ASA of 100 (or 125) and measure sunlight (point the sensor directly at the sun) on a clear day around noon. It should show 1/100 (or 1/125) sec @ f/16. There will be a slight error if the ASA and shutter speeds are not the same.</p>

<p>When using the meter, point the sensor at the camera (if you're at the subject position), or in that direction if you're at the camera.</p>

<p>- Leigh</p>

 

<blockquote>

<p>Robert Melone wrote:<br /><em>I recently read a good book by John Shaw "Nature Photography Field Guide" which indicated that if you are shooting in manual mode using spot or CW metering you should open up one stop if your subject is lighter than medium gray. John pointed out that the cameras TTL meter works in this way. So, over the weekend, I did some shooting in manual mode (e.g. a red covered bridge at midday) and noticed that opening up one stop overexposed all of my images. </em></p>

</blockquote>

<p>As others have pointed out:</p>

<ol>

<li>it's not as simple as some sure-fire +1 f/stop rule of thumb (but I understand you've abridged the procedure)</li>

<li>manual or [semi]automatic mode doesn't make a difference -- exposure is still exposure, no matter whether you set it manually or your camera did it for you</li>

<li>neither the time of day (morning/evening vs noon) nor the conditions (sunny vs overcast) matters</li>

<li>for most precise metering of a subject, use <strong>spot</strong> and then zoom-in and/or move in closer to it so that to further limit lens's field of view when taking the reading -- otherwise you can't be sure what are you really measuring</li>

<li>the single greatest challenge is to know the reflectance of your subject -- how did you figure out that red bridge was 1-stop lighter than medium gray?! -- either it wasn't (or you measured it wrong as per #4 above), or your meter is not calibrated properly</li>

</ol>

<p>As someone has mentioned before, it takes a lot of practice to judge the tonality of things, and the only way to develop that expertise is to carry an 18% gray card with you, place it next to your subject (or at least in the same light as your subject), then measure both and compare the readings: if, relative to the gray card, your meter tells you that you're over-/under-exposing your subject by x f/stops, then your subject is by that much lighter/darker than your reference card. Chances are, your bridge would have shown the same reading, or it might have even been slightly darker. So-called <em>medium-gray</em> appears much lighter than what most people intuitively think it would. Read again: people not accustomed with the subject matter usually perceive 18% gray card (or equivalent in a different color) as lighter than the middle tone!</p>

<p>As for testing your in-camera meter: make an evenly lit still life arrangement (just place various light and dark objects as well as your 18% gray card somewhere in shade to avoid glare and harsh shadows), measure on the card and expose accordingly, and if it looks good, then it's calibrated properly.</p>

<p>Finally, just to reiterate what was said before, do learn how to read the histogram -- it's a *great* advantage.</p>

<p>Enjoy your learning experience!</p>

<p>P.S.<br />The only John's advice that doesn't apply to digital is calibrating the light meter by adjusting ISO -- that setting doesn't affect film in any way, but it's like loading a faster or slower film into the DSLR, figuratively speaking.</p>

<blockquote>

<p>the single greatest challenge is to know the reflectance of your subject -- how did you figure out that red bridge was 1-stop lighter than medium gray?</p>

</blockquote>

<p>The elephant in the room that everybody is avoiding is chromatic sensitivity of the reflectance metering system. </p>

<p>Two very different colors which might yield the same density on a B&W negative may NOT show the same exposure when checked with a reflectance meter due to non-linear chromatic response of the sensor.</p>

<p>The only solutions to this problem are:<br>

1) have a filter installed and the meter re-calibrated, as was commonly done to the Pentax Spotmatic for use with the Zone system, or<br>

2) do extensive testing on your own.</p>

<p>The issue is non-trivial, and can result in significant exposure errors.</p>

<p>And unless you have decades of experience doing so, don't even think about guessing the reflectance of a particular subject or color. You can do this as a learning experience, and you'll find yourself coming closer as you practice. I've been shooting B&W for over 50 years, and I still don't try to second-guess the meters.</p>

<p>- Leigh</p>

 

<p>Try the Black Cat experiment. That is photograph 3 black items (it does not have to be a cat) and 3 white items and deviate a little from your initial exposure by(+/-) 1 or 2 fstops. Then look at your viewfinder and see which of the images matches the original the most. I haven't sone this experiment in a while but last time I did, the camera does seem to compensate for the (+/-) deviations at times. </p>

<blockquote>

<p>...the camera does seem to compensate for the (+/-) deviations at times.</p>

</blockquote>

<p>The camera knows nothing about the subject. It sees all the light coming through the lens and tries to set the exposure to produce an average 18% gray image.</p>

<p>That's why it "seems to compensate at times".</p>

<p>- Leigh</p>

<blockquote>

<p>Leigh wrote:<br /><em>The elephant in the room that everybody is avoiding is chromatic sensitivity of the reflectance metering system.</em><br /> <em>Two very different colors which might yield the same density on a B&W negative may NOT show the same exposure when checked with a reflectance meter due to non-linear chromatic response of the sensor.</em></p>

</blockquote>

<p>...but we're not talking about B&W here, are we? :S<em> </em><br /> Besides, isn't the difference mostly due to B&W film's sensitivity to colour?<br /> Personally, I've never experienced this problem: light/medium/dark gray vs light/medium/dark green vs light/medium/dark red has always been all the same to me. Isn't that what the Zone system is all about?</p>

<p>As for learning to recognize tonality, it's not THAT hard in typical situations. Just test and memorize some common "reference subjects" such as grass or whatever else that is frequently featured in your photographs (not necessarily as the main theme) or at least handy (you can measure light off something you don't include in the photograph as long as it's in the same light, hence the trick with your own palm).</p>

<p>And finally, usually you don't have to guess or otherwise determine your subject's reflectance very accurately for a couple of reasons: first, as long as you don't blow the highlights and/or completely block the shadows, you can recover quite a lot in RAW conversion; second, let's don't forget photographer's intent! -- we don't have to record things exactly as they are, and sometimes we even cannot because of the limited dynamic range when something has got to be sacrificed, so I may choose to render something that in fact is light shade of blue as dark, if that's the look I want to achieve (so I just expose it @ -1EV or so without worrying what it's actual reflectance is as compared to the 18% gray card).</p>