indoor low lighting question

kevinpw

Hi,

I just took some shots in an indoor low lighting situation, and although I can live with the results, there's a lot of room for improvement. I was using a combination of aperture-priority mode and full manual. Most of the time I set the aperture to f/8-11 to ensure sharpness. Because of this, I had to use a slow shutter speed to get the correct exposure (or at least my D5000 thinks so), and it set shutter speed to anywhere between 1/10 - 1/30. This slow shutter speed results in some blur obviously, but if I use a faster shutter speed then everything will be underexposed. So my questions are:

1. What settings should I use in this situation, and without flash?
2. What should I meter light off?

Thanks!

jeffreaux2

To protect against motion blur you should use manual mode....not aperture priority.....but you already discovered that.

Wedding work has put me in this situation many times.....and I am reluctant to use a flash during a ceremony....though I frequently use one for the receptions.


My advice is to set a shutter speed that you are comfortable with first. With my 17-55mm F2.8 with built in stabilization I dont mind going as slow as 1/60 or 1/30........but without stabilization I might rather 1/80 to 1/125. Next choose an aperture. A wider angle lens can handle a fairly wide aperture and still have reasonable sharpness. Again....with the 17-55mm as an example, F4 would be fine for most situations....and maybe even F3.5 or 2.8 if used at its wider end....17-30mm. Next.......start upping the ISO until the photo is properly exposed. Use your histogram to insure that the whites are exposed as bright as possible without being blown out.....and shooting in RAW can add some additional flexibility when it comes time to edit shots that are a bit under or over exposed.


Again....forget metering off something for exposure.......use the histogram and keep it set so that it is viewable as you shoot....for chimping and continual adjustment if needed.

FWIW......a properly bounced shoe mounted flash can work wonders in these situations. It's an easy way to make the gear you already own more flexible in any lighting situation.

kevinpw

thanks. one concern i had about widening the aperture is DOF. wouldn't widening the aperture result in a shallow DOF?

ChatKat

Distance
To shoot wide open in dark environments, you need some distance between you and your subjects to have the right amount of DOF. If you are too close you can get one eye in focus and one out of focus. Practice is critical.

craig_d

ChatKat wrote:

To shoot wide open in dark environments, you need some distance between you and your subjects to have the right amount of DOF. If you are too close you can get one eye in focus and one out of focus. Practice is critical.

True, but it depends on the focal length you're using. If you're trying to take close-up portraits in low light without a flash, then yeah, you are in a tricky situation. Group shots are much easier because you get more DOF at shorter focal lengths.

kevinpw

that's confusing... I actually didn't realize focal length affects DOF. I thought it's just aperture that affects DOF? So what is the formula here?

smaller aperture = more DOF
wider aperture = shallower DOF

shorter focal length (wide angle) = more DOF?
longer focal length (tell) = less DOF?

so how does these two variables combine to set DOF?

pathfinder

Focal length does not affect Depth of field if the subject's images are the same size on the film plane.

Before you flame me about this statement, read the link above - ( This is not true if the subject to image plane is close the the focal length of the lens, but very few normal images are shot at 50mm or less - only macros at under 2 inches)

The reason the DOF seems larger with shorter focal lengths, is that you are farther away from the subjects, and they are much smaller on the focal plane.

DOF is controlled by aperture selected, and by the distance from the subject to the image plane. That is why we think wide angles have greater DOF. The subject image size is frequently much smaller with a wide angle than a telephoto. To get the image size the same with a wide angle and a telelphoto, you must be much, much closer with the wide angle - and hence, your DOF is very similar to that of the telephoto with the same subject size for each lens on the film plane.

There are exceptions to the above for macro lenses shooting near 1 to 1, but that is not what this discussion has been about.

For an interesting exercise, shoot a teddy bear, or a child's wooden block with a 24mm, 50mm, 100mm, 200mm and 400mm lens such that the block is exactly 1/3 of the image plane's length. Make some nice prints, and then examine the surrounding DOF. Let me know what you find out!

craig_d

kevinpw wrote:

that's confusing... I actually didn't realize focal length affects DOF. I thought it's just aperture that affects DOF? So what is the formula here?

Here's the deal. DOF is influenced by four things:

1. Aperture: wider aperture = shallower DOF

2. Focal length: longer focal length = shallower DOF

3. Focus distance: nearer focus = shallower DOF

4. Circle of confusion: smaller CoC = shallower DOF

#3 is why macro photography has incredibly shallow DOF -- you're focusing on things inches from the lens. It's also why there is such a thing as a hyperfocal distance.

"Circle of confusion" is basically the maximum size a circle can be on your negative or sensor and still look like a point. In other words, you can get away with very, very small amounts of blur because it will still look sharp at the sizes at which images are commonly printed.

The typical value for the CoC with a 35mm full-frame camera is 0.03mm, or for APS-C, 0.02mm. This is based on the assumption that you want the image to look right in an 8"x10" print viewed from a few feet away (for example, hand-held or hung on a wall that you're standing in front of). You might want to choose other values if an 8"x10" print or a magazine cover is not your goal.

There is a nice online DOF calculator here:

http://www.dofmaster.com/dofjs.html

For the camera model you select, it will choose a reasonable CoC, and then you can enter whatever focal length, focus distance, and aperture you like.

kevinpw

Thanks a lot Craig. I have never heard of Circle of Confusion until you mentioned it. Shows how little I know about anything -__-

Here is an example picture that I took that may be applied to my question: http://www.kevinpw.com/Other/Mistakes/10005770_Tfw5s#684580330_rULxh


In this shot, I was on aperture-priority mode. I wanted my friend to model the necklace gift she received, with only the box and its contents in focus. My friend should be a little blurred. So I chose a wide aperture to achieve this effect, f/4.5. Because I am in aperture-priority mode, the camera chose a shutter speed of 1/25. I had ISO set to 1000. The result is:

1. My friend is blurred.
2. The top half of the box is in focus.
3. The bottom half of the box and its contents is not in focus.

I'm wondering why #3 occurred. I keep thinking that whole box should just be in focus. The only other explanation I can think of is she might have slightly moved and because of the slow shutter speed the result is a blur, but that doesn't make sense since the whole box would've been out of focus.

So what should I have done to get the whole box in focus?

Thanks!

craig_d

pathfinder wrote:

Focal length does not affect Depth of field if the subject's images are the same size on the film plane.

Before you flame me about this statement, read the link above - ( This is not true if the subject to image plane is close the the focal length of the lens, but very few normal images are shot at 50mm or less - only macros at under 2 inches)

The reason the DOF seems larger with shorter focal lengths, is that you are farther away from the subjects, and they are much smaller on the focal plane.

DOF is controlled by aperture selected, and by the distance from the subject to the image plane. That is why we think wide angles have greater DOF. The subject image size is frequently much smaller with a wide angle than a telephoto. To get the image size the same with a wide angle and a telelphoto, you must be much, much closer with the wide angle - and hence, your DOF is very similar to that of the telephoto with the same subject size for each lens on the film plane.

I've read that article, and also the Luminous Landscape article that it references. They're not exactly wrong, but I'm not impressed.

When they say that "the subject occupies the same fraction of the viewfinder", or that magnification is constant, what it basically means is that the variables of focal length and focus distance can balance each other. If you increase your distance from the subject, increase focal length to compensate, and refocus on the same point, then DOF will be about the same. However, in the real world, distance to subject and focal length are two independent variables that are controlled in completely different ways. You alter focal length by zooming or switching lenses; you alter distance to subject by physically moving the camera or the subject to a different position in space. Combining these two real-world variables is a convenience that may simplify the equation in the mathematical realm, but it does not relate as directly to the things that photographers do in the real world. It is, of course, useful to know that your DOF will not be significantly affected if you step back, zoom in a little, and refocus, but that's a special case.

Ultimately, the real problem here is that DOF is a somewhat confused concept that is generally presented in an overly simplistic manner. Most articles on the subject tell you to use 0.03mm as the CoC for 35mm cameras without really making it clear that this value relates to a set of assumptions that will be precisely true only in a minority of cases. DOF is really an attempt to be mathematically precise about something that by its nature is somewhat vague.

craig_d

kevinpw wrote:

The result is:

1. My friend is blurred.
2. The top half of the box is in focus.
3. The bottom half of the box and its contents is not in focus.

I'm wondering why #3 occurred. I keep thinking that whole box should just be in focus. The only other explanation I can think of is she might have slightly moved and because of the slow shutter speed the result is a blur, but that doesn't make sense since the whole box would've been out of focus.

So what should I have done to get the whole box in focus?

According to the dofmaster DOF calculator (which is based on all the usual imperfect assumptions, but it'll do for now), 44mm at f/4.5 with a subject distance of about two and a half feet (my guess for how far away the word "Swarovski" was from the camera) results in DOF of just under three inches, from 1.38" in front of the focus point to 1.52" behind it. The box looks to be at least 4"x4" in size, so the front part of it, and her fingertips, are closer to you than the near edge of the focused area.

Even so, it's not very blurry; it looks sharp if I view it at SmugMug's "Large" size, and not bad even at "X3" size. At "Original" size, I can see the blur easily, but on my display that's equivalent to a 29"x44" print. Standard DOF calculations tend to be based on an 8"x10" print.

To improve this picture, for one thing, I would have asked her to tilt the box toward the camera a little more for a better view of the necklace pendant. Then I would have focused on the point of the necklace just above the pendant. Since you're already at ISO 1000 and 1/25 sec., I'd probably leave the aperture alone rather than risk motion blur. In better light, f/6.3 would probably have been sufficient to improve focus on the box a bit without making the girl sharper than you wanted.

pathfinder

craig_d wrote:

I've read that article, and also the Luminous Landscape article that it references. They're not exactly wrong, but I'm not impressed.

When they say that "the subject occupies the same fraction of the viewfinder", or that magnification is constant, what it basically means is that the variables of focal length and focus distance can balance each other. If you increase your distance from the subject, increase focal length to compensate, and refocus on the same point, then DOF will be about the same. However, in the real world, distance to subject and focal length are two independent variables that are controlled in completely different ways. You alter focal length by zooming or switching lenses; you alter distance to subject by physically moving the camera or the subject to a different position in space. Combining these two real-world variables is a convenience that may simplify the equation in the mathematical realm, but it does not relate as directly to the things that photographers do in the real world. It is, of course, useful to know that your DOF will not be significantly affected if you step back, zoom in a little, and refocus, but that's a special case.

Ultimately, the real problem here is that DOF is a somewhat confused concept that is generally presented in an overly simplistic manner. Most articles on the subject tell you to use 0.03mm as the CoC for 35mm cameras without really making it clear that this value relates to a set of assumptions that will be precisely true only in a minority of cases. DOF is really an attempt to be mathematically precise about something that by its nature is somewhat vague.

Thanks for reminding me about Michael's link concerning DOF and focal length. I saw a similar article published on Pop Photo 25 or more years ago. The concept is still true. One factor that does contribute to greater DOF, that was not mentioned, is format size. Smaller sensors intrinsically offer more dof at similar apertures.

I agree that most discussions about DOF wander off into mathematics, and my concern is the final print on the wall - not at 8x10 inches, but 16 x 24 or so.

craig_d

pathfinder wrote:

One factor that does contribute to greater DOF, that was not mentioned, is format size. Smaller sensors intrinsically offer more dof at similar apertures.

Not really, but I understand why you think so. It's a statement that has practical merit given the aesthetic biases of photographers, but that isn't actually technically correct as such. There's nothing about the sensor itself that changes DOF. What changes DOF are the way the photographer relates to it and what he does with the images he produces with it.

The only intrinsic property of a smaller format, relative to a larger one, is that it has a narrower field of view for any given focal length. Not only is this self-evident, but from it you can derive most of the other differences. A narrower field of view implies the use of shorter focal lengths and greater subject distances to obtain a desired framing, both of which result in greater DOF. But the desire to obtain a particular framing is a property of the photographer, not of the format.

At the same time, the desire to print images at a particular size implies that a smaller format requires a smaller CoC value (since the image will require greater magnification), which decreases DOF.

(sigh) I really should write my own series of articles on these issues, because there are a lot of flawed write-ups out there, including that Cambridge tutorial. He treats magnification as if it were a fundamental property of lenses, which it isn't. Magnification derives from focal length, so it's absurd to say that focal length isn't related to DOF but magnification is. Furthermore, lenses in the real world do not have a user-friendly way of altering magnification and subject distance in a coordinated way (for obvious reasons), so it's more useful to the photographer to tell him that focal length and subject distance both influence DOF. This way, he knows that he can compensate for a change in one by changing the other.

kevinpw

wow this is a lot of information to take. I'm not sure I understand everything you guys have said. I'll need to read your posts a few times and practice as well. I'd rather not get too much into the math and technical formulas to get DOF. Perhaps it's better if I just shoot an exposure at a few different apertures to compare and learn.

what I did get from your posts is that focal length affects DOF but not as much as aperture does? next time I will try to stand a few more inches away from the subject in this kind of shot, and use different apertures.

kevinpw

while we're talking about low-light indoor situations, i'd like to ask about flashes. i dont own one and ive never used one, so i know zero about them. i am almost sure that i will be at a place with even less lighting this weekend and i'd like to be prepared. perhaps i will rent one of those flashes.

my question is, how do i use that kind of flash to achieve best lighting? what i hate about my internal flash is the light is very harsh. people's faces become full of blemishes, like a pool of oil, and it's just gross. what do i do with these external flashes to make sure the light is soft and nice?

Thanks

craig_d

The simplest way to use a flash and have reasonably nice lighting is to turn it so that the light bounces off a white wall or ceiling to illuminate the subject. This spreads out the light and makes it come in from another angle instead of firing directly at the subject from just above the lens. If no large white surface is available for bouncing, a small diffuser mounted on the flash can soften the light somewhat, though bouncing is usually better.

I don't know Nikon's product line, but with Canon, the cheapest model of external flash unit cannot be rotated, which basically makes it almost useless. Make sure whichever one you rent can rotate at least 90 degrees in each direction, and tilt to point straight up as well. Also make sure you know what kind of batteries it needs (and how many) in case the rental company doesn't provide them. Flashes typically do not run off the camera's batteries.

Photographic lighting is a whole art form in itself. There is so much you can do with it. Pros working in a controlled situation may have several flash units all controlled by a master unit attached to the camera's hot shoe. Some of the flashes may fire at different intensities or might be aimed through umbrellas to further diffuse the light. This is just one example of the kinds of things that are done.

kevinpw

hmmmm bounce it? so let's imagine this. imagine you're in a nice, fine dining restaurant. it's dinner time and each table is only dimly lit with small lights and some candles. it's a long table full of people, anywhere between 10 - 20 people. one side has their backs against the wall, the other side has their backs to other tables and people.

how would I use the flash to give soft light to and from both sides?

craig_d

kevinpw wrote:

hmmmm bounce it? so let's imagine this. imagine you're in a nice, fine dining restaurant. it's dinner time and each table is only dimly lit with small lights and some candles. it's a long table full of people, anywhere between 10 - 20 people. one side has their backs against the wall, the other side has their backs to other tables and people.

how would I use the flash to give soft light to and from both sides?

I don't think I would want to blast other diners in the face by bouncing the flash off them. Also, if you're in a candle-lit restaurant, using a flash will completely change the look of the scene by drowning out the illumination from the candles.

My inclination in this sort of situation would be to go for the widest-aperture lens I could get my hands on and the highest ISO level I was willing to tolerate. I might use a flash on a low power setting (either put the flash in manual mode or apply negative flash exposure compensation on the camera), with a diffuser mounted on it, to add a little light without completely overpowering the candles. In that case the flash would not be bounced, but just pointed directly at the subject (possibly tilted up a little -- experiment to see what works best for you).

Generally speaking, if you have light coming from one side, you can use a large, flat, white object (a large piece of paper or plastic will do) on the other side, just outside the frame, to bounce some of the light back again. Or get a second flash and slave one to the other, assuming one of them is capable of serving as a flash master.

kevinpw

ok. i think slave-master is beyond what i want to do right now. i dont think i wanna carry all that gear around to a restaurant, not like im getting paid to take the pictures it's all for fun. plus, id rather learn to use one flash properly before i use more than one.

thanks