What is a stop? - DPChallenge Forums

05/18/2005 01:42:38 PM · #1

Probably a dumb question, but I wanted to know: in exposure terminology what exactly is a stop? Does that equate to a notch on the exposure meter of a standard camera? Thanks much.

-Nick

05/18/2005 01:57:29 PM · #2

Originally posted by nfessel:

Probably a dumb question, but I wanted to know: in exposure terminology what exactly is a stop? Does that equate to a notch on the exposure meter of a standard camera? Thanks much.

-Nick

Pretty much, yeah. Back in the day each succesive aperture had a click-stop on the aperture ring, so they came to be called "stops". In time a "stop" came to refer to a doubling or halving of exposure, so you could "gain one stop of exposure" by opening the lens one stop OR by doubling the exposure time.

"f-stop" is the correct term for the size of the aperture in the camera: f:4.0, f:5.6, f:8.0, f:11.0, f:16.0, f:22.0, each delivers to the film half as much light as the one preceding it. The f-stop number is derived by dividing the focallength of the lens by the physical diameter of the opening. so in a 50mm lens, a 25mm opening would f:2.0. The same 25mm opening in a 200mm lens would be f:8.0. This is why lenses get "darker" as they get longer, OR become extremely "fat"; to have f:4.0 on a 1000mm lens requires an aperture 125mm in diameter, approximately 5 inches.

Robt.

Message edited by author 2005-05-18 13:58:16.

05/18/2005 02:44:36 PM · #3

Originally posted by bear_music:

Originally posted by nfessel:
Probably a dumb question, but I wanted to know: in exposure terminology what exactly is a stop? Does that equate to a notch on the exposure meter of a standard camera? Thanks much.

-Nick

Pretty much, yeah. Back in the day each succesive aperture had a click-stop on the aperture ring, so they came to be called "stops". In time a "stop" came to refer to a doubling or halving of exposure, so you could "gain one stop of exposure" by opening the lens one stop OR by doubling the exposure time.

"f-stop" is the correct term for the size of the aperture in the camera: f:4.0, f:5.6, f:8.0, f:11.0, f:16.0, f:22.0, each delivers to the film half as much light as the one preceding it. The f-stop number is derived by dividing the focallength of the lens by the physical diameter of the opening. so in a 50mm lens, a 25mm opening would f:2.0. The same 25mm opening in a 200mm lens would be f:8.0. This is why lenses get "darker" as they get longer, OR become extremely "fat"; to have f:4.0 on a 1000mm lens requires an aperture 125mm in diameter, approximately 5 inches.

Robt.

So it may or may not be one notch on the camera exposure meter, but it is one aperture value up from the previous aperture (f-number). Is that correct? I just want to be sure because I am doing some testing with the zone system and I need to know what they mean by "a stop" so I can run the tests correctly. Thanks much for the information.

-Nick

05/18/2005 02:56:04 PM · #4

More math for the math lovers:

It is not the next value in the aperture scale. Notice Rob's reasoning below... it doubles (or rather halves, depending in which direction you go) the amount of light.
The number is what you divide the focal length by, so the bigger the number is, the less is the aperture.
Also, the area is proportional to the square of the linear size.

This means that in order to double the area, you need to multiply the linear element by square root of 2 (approx. 1.414) Apply this to the aperture numbers and you'll see which ones are the 'full stop' values:

2.8 3.959 5.6 7.92 11.2 15.84 22.4

which corresponds to (rounded)

2.8 4.0 5.6 8.0 11 16 22

So, apertures such as 3.5, 4.5, 6.3 are NOT the full stops you are looking for
I hope this helps

Nikon AF-S DX NIKKOR 18-200mm f/3.5-5.6G ED VR II

05/18/2005 02:58:13 PM · #5

By definition, in discussing Zone System, "one stop" refers to a doubling or a halving of the exposure given to the film or, when using a light meter to measure the exposure, to a doubling or halving of the luminance read.

There are two kinds of light meters: reflected light meters and
"incident" light meters. The latter are commonly used in cinematography, and they measure the brightness of the light falling on the subject. We can define this most practically in terms of stops: "this lighting is one stop brighter now that I've moved the key light in closer."

Zone System is based on precise measurement of the light reflected from key areas of the subject. We meter the dark areas and the bright areas and see how many stops aprt they are. We then adjust our processing of the negative to expand or contract the tonal range of the negative so that all the zones contained within the image can be expressed fully within the relatively limited tonal range that a print on paper is capable of achieving.

A reflected light meter works on the presumption that the surface it is metering is a "Zone 5 gray" (18% luminance). So if you meter a black wall and shoot it full frame, then meter a white wall and shoot that full frame, both at the indicated exposure, and process both films the same way, the negatives will be indistinguishable from each other.

Therefore, to get a "correctly exposed" negative of the black wall (you would want it on Zone 3 in the print, probably) you'd have to expose for 2 stops less than the metered exposure. To correctly expose the white wall to Zone 7, you'd need to give 2 stops MORE exposure. This means, effectively, that if you wanted a shot of a wall that's half-black and half-white to be "perfectly" exposed, you's need to make sure that there was precisely a 4-stop difference between the two meterings, and if this was not the case you'd push or pull your film in processing to make the tonal range of the resultant negative nominal for the previsualized print.

I used to teach Zone System, feel free to ask me questions about it.

Robt.

kyebosh

Canon EOS-20D

Canon EF 80-200mm f/2.8L

05/18/2005 02:59:33 PM · #6

Well a stop of light is not just the Fstop value. 1/30 and 1/60 are exactly one stop apart. Just the same as ISO 100 and 200 are. Often in post processing you could say "I underexposed by 1 full stop" and that has not that much to do with the Fstop values used, it has to do with more than just that.

BobsterLobster

Fujifilm FinePix S3 Pro

05/18/2005 03:00:09 PM · #7

Originally posted by bear_music:

Didn't know that, very useful info. How does it work for zoom lenses though?

kyebosh

Canon EOS-20D

Canon EF 80-200mm f/2.8L

05/18/2005 03:03:46 PM · #8

"Didn't know that, very useful info. How does it work for zoom lenses though?"

works the same way actually... if you look at a lens where the F stop value changes when you zoom, you can physically see the size of the shutter change while zooming. Funny thought that I don't see any change in my 80-200L 2.8...

edit: as he said the F stop is just a ratio of length of focus and size of maximum shutter opening.

Message edited by author 2005-05-18 15:04:43.

05/18/2005 03:05:56 PM · #9

It works the same way, you'll just have to know (read from the barrel) what focal length you are at.

This is why some lenses list the range of aperture values. For example, my Sigma 18-50mm has a range of 3.5-5.6, and it translates to 3.5@18mm to 5.6@50mm. Same with my 75-300mm F/4.0-5.6. It starts at 4.0@75mm and closes to f5.6 @ 300mm.

You can try this in manual mode: turn the zoom to the shortest f-length and then open the aperture as widew as you can. Then start turning the zoom lens, and watch the aperture jump up as you lenghten the lens.

That is, unless you have a constant aperture lens (I don't know about nikkor lenses, but most canon L lenses are fixed aperture on the full zoom range.) e.g. 70-200mm f/2.8. There are exceptions of course.

Canon EF 70-200mm f/2.8L IS USM

05/18/2005 03:07:45 PM · #10

Originally posted by kyebosh:

This is correct; you may add a stop of exposure by doubling the exposure time, and vice versa. There is a joker in that deck, though, and it's called "reciprocity failure"; every film has a maximum length of exposure beyond which doubling the time of exposure does NOT double the exposure the film receives. Typically, higher-speed films (high ISOs) encounter reciprocity failure sooner than slower films do. It's particularly evident in color transparency films.

When we were shooting available light architectural interiors under very dim light (often unavoidable due to architectural lighting and interior design that's impossible to light adequately with strobes or floods) our nominal exposure for a 4x5 color transparency might be, say, 16 minutes. One stop "over" would be more on the order of 50 or 60 minutes, and one stop "under" might be 5 minutes.

Fortunately, Polaroid's 55 P/N 4x5 film very closely mimicked the reciprocity characteristics of Ektachrome film, so we'd set up the camera, calculate our best-guess exposure, and start exposing a test shot as we moved around prepping the room for the final shot. Some of these outtakes on polaroid were very funny, with arbitrary ghost images of people wafting about them in unexpected ways.

Robt.

Plexxoid

Canon EOS-30D

05/18/2005 03:07:50 PM · #11

Zoom lenses are much more complicated, but I will share by saying that exposure is Tv x ISO / Av^2. As long as the number formulated stays the same, the ISO, Av, or Tv may all change in respect to eachother.

Also I discovered that 1/3rd stops are 2^(1/6) times more than the previous one and that each stop is a square root of 2^x because of the two dimensional area of the openning. Example: F11 is actually the square root of 128, but I doubt the need for the precision.

Yeah, I got bored one time and wrote it all out, would've been easier to learn that from a book, but I just figured it out on my own...

Message edited by author 2005-05-18 15:13:36.

Mr_Pants

SiPix Blink

05/18/2005 03:15:50 PM · #12

Originally posted by srdanz:

More math for the math lovers:

This means that in order to double the area, you need to multiply the linear element by square root of 2 (approx. 1.414) Apply this to the aperture numbers and you'll see which ones are the 'full stop' values:

2.8 3.959 5.6 7.92 11.2 15.84 22.4

Shouldn't the second number be four exactly even if not rounded?

Nikon AF-S DX Zoom-Nikkor 18-70mm f/3.5-4.5G IF-ED

05/18/2005 03:15:55 PM · #13

Good quality zoom lenses have the aperture physically linked to the zoom ring, so as you extend or contract the zoom at, say, f:4.0, the aperture does, indeed, expand or contract slightly so the exposure data remain constant. Most zoom lenses, however, vary in their maximum aperture, relatively speaking, as you can't get any larger than wide open. The "L" lenses get around that, but I don't know exactly how. I'd assume they limit the size of the opening at shorter zoom lengths to the indicated ratio, and expand beyonf that to physically maximum aperture at the greatest extension of the zoom range.

One thing's a fact: a zoom lens that does NOT automatically adjust aperture size within its range is essentially useless for Zone System photography. Parenthetically, when we used the view cameras at extreme extension for close-up work (like in the photography of architectural models, or for product shots of small objects) we had to manually calculate the decrease in effective f-stop by making precise measurements of the actual length of the extended setup and applying a formula to that.

Robt.

DannyM

Nikon D90

05/18/2005 03:46:09 PM · #14

The Zone system is fascinating. I learned a little about it but that was many years ago. I believe we also took into consideration type of film, type of developer, type of paper, enlarger position, enlarger lens aperture etc. Accounting for as many variables as possible made for consistent results. Very difficult to do with roll film however. It's ideally suited for view cameras but it was a valuable learning experience nonetheless.

05/18/2005 03:56:17 PM · #15

Originally posted by DannyM:

Absolutely. Every exposure meter has to be calibrated. You buy your film in bulk, all of the same emulsion batch, and test the film to find its true ISO speed. Each shutter needs to be calibrated and the variances from nominal recorded. (with view cameras, the aperture is built into the shutter, so you're calibrating both aperture and shutter speeds for each shutter) Developers must be tested and then mixed religiously. You have to use distilled water to mix your developers or day-to-day variances in water quality from the tap will affect your processing. These are the main considerations in getting properly exposed and processed films. Issues with papers and enlargers are less critical, because the film is still there and you can futz around. However, nominal processing of the film DOES need to be done with a specific paper in mind; the "perfect" negative for a glossy, RC print is different from the one for a matte-finish paper, for example, and then there are the contrast characteristics of different papers to factor in.

Really an anal process, to tell the truth, but the end results are worth it.

Robt.

05/18/2005 04:03:28 PM · #16

Originally posted by AJAger:

Originally posted by srdanz:
More math for the math lovers:

This means that in order to double the area, you need to multiply the linear element by square root of 2 (approx. 1.414) Apply this to the aperture numbers and you'll see which ones are the 'full stop' values:

2.8 3.959 5.6 7.92 11.2 15.84 22.4

Shouldn't the second number be four exactly even if not rounded?

Mr_Pants

SiPix Blink

05/19/2005 02:46:22 PM · #17

I assumed that the sequence would start from 1 (root 2 to the power of 0). I know that root 2 is an irrational number, but surely we are talking about successive powers rather than multiples ?

Originally posted by srdanz:

Nah,

the square root of 2 is not a rational number, so any multiples of it cannot produce round numbers like that. And I started from 2.8. If I started from 2.0, I would have gotten the 4, 8, and 16, but the 2.8, 5.6, and 11.2 would look different...
2 2.83 4 5.66 8 11.31 16 22.63 etc.

Originally posted by AJAger:

Originally posted by srdanz:
More math for the math lovers:

This means that in order to double the area, you need to multiply the linear element by square root of 2 (approx. 1.414) Apply this to the aperture numbers and you'll see which ones are the 'full stop' values:

2.8 3.959 5.6 7.92 11.2 15.84 22.4

Shouldn't the second number be four exactly even if not rounded?

Panasonic Lumix G 14mm F2.5 ASPH

05/19/2005 02:54:30 PM · #18

DOH!

I was thinking too much of the 2.8 lens I am drooling about these days.
Of course the ratios start at 1.

Edit: I was of course showing different powers of sqrt(2) here. It is not multiples of it bad choice of words.

Message edited by author 2005-05-19 14:56:12.

Olyuzi

Panasonic Lumix DMC-GX7

05/19/2005 03:58:57 PM · #19

Is it a good idea to make a test of your camera's dynamic range by photographing a 18% gray card at varying f-stops (1/3rd of a stop apart) in both directions of over and underexposure until you get to pure black and pure white? In this way, you can know just how a photographic subject's exposure range fits into your camera's and can then adjust your exposure settings accordingly.

05/19/2005 05:30:03 PM · #20

Originally posted by Olyuzi:

I'm not 100% sure on this, but I don't think it would be meaningful. "pure black" and "pure white" are dependent on the output settings anyway. It's possible you can accomplish what you want by shooting a series on ards that were split vertically into 18% gray and pure white, 18% greay and pure black, metering for the gray, and the proceeding as you suggested and watching the values converge. I've honestly never thought about it though.

Robt.

Message edited by author 2005-05-19 17:30:23.

05/19/2005 05:45:19 PM · #21

Originally posted by bear_music:

Originally posted by Olyuzi:
Is it a good idea to make a test of your camera's dynamic range by photographing a 18% gray card at varying f-stops (1/3rd of a stop apart) in both directions of over and underexposure until you get to pure black and pure white? In this way, you can know just how a photographic subject's exposure range fits into your camera's and can then adjust your exposure settings accordingly.

I'm not 100% sure on this, but I don't think it would be meaningful. "pure black" and "pure white" are dependent on the output settings anyway. It's possible you can accomplish what you want by shooting a series on ards that were split vertically into 18% gray and pure white, 18% greay and pure black, metering for the gray, and the proceeding as you suggested and watching the values converge. I've honestly never thought about it though.

Robt.

I'm not sure. Ansel Adams, in his book "The Negative", suggests at the very beginning of his book to create exposures for each of his ten zones by metering a wall and creating 10 separate exposures of the same wall. The "18% zone" (Zone V) equates (or SHOULD equate) to 18% gray. Metering with a camera, this means that shooting the "correct" exposure suggested by the camera would give you the Zone V gray.

With this Zone V exposure value as a reference, you can then increase by 1 stop and get Zone VI exposure for the same wall. Increasing the stop by 1 again would give Zone VII. Vice versa for going from Zone V to Zone IV. The higher the Zone number, the lighter the tone. So Zone 1 is basically completely black, and Zone X is basically completely white. Theoretically, this is how it should work. I'm going to attempt the test this afternoon with a digital camera. This test with a film camera may work differently. I don't know if this test will even work with a digital camera, but theoretically it should work.

Anyway, this is why I was wondering exactly what a "stop" is, so I can try this zone test.

-Nick

Message edited by author 2005-05-19 17:46:52.

05/19/2005 05:48:24 PM · #22

Slightly off-topic, but, nfessel, could you please edit your Bio and remove those rrrrrrrrrrrs, or at least put some line breaks so that I could look at your photos easier (that's what your page is for, to look at the photos, right?)

thanks,

05/19/2005 05:51:12 PM · #23

Originally posted by srdanz:

A Firefox bug that I wish they would fix. Darnit! :) Annoys the heck out of me about this textbox issue. I removed the scrolling r's.

In Firefox, if you don't put line breaks in your textboxes, Firefox won't insert them, so the thing just scrolls forever. Very annoying and I can't believe they haven't fixed it yet.

Internet Explorer inserts the line breaks automagically. :)

Message edited by author 2005-05-19 18:00:07.

05/19/2005 05:52:00 PM · #24

Originally posted by nfessel:

Originally posted by bear_music:

Originally posted by Olyuzi:
Is it a good idea to make a test of your camera's dynamic range by photographing a 18% gray card at varying f-stops (1/3rd of a stop apart) in both directions of over and underexposure until you get to pure black and pure white? In this way, you can know just how a photographic subject's exposure range fits into your camera's and can then adjust your exposure settings accordingly.

I'm not 100% sure on this, but I don't think it would be meaningful. "pure black" and "pure white" are dependent on the output settings anyway. It's possible you can accomplish what you want by shooting a series on ards that were split vertically into 18% gray and pure white, 18% greay and pure black, metering for the gray, and the proceeding as you suggested and watching the values converge. I've honestly never thought about it though.

Robt.

I'm not sure. Ansel Adams, in his book "The Negative", suggests at the very beginning of his book to create exposures for each of his ten zones by metering a wall and creating 10 separate exposures of the same wall. The "18% zone" (Zone V) equates (or SHOULD equate) to 18% gray. Metering with a camera, this means that shooting the "correct" exposure suggested by the camera would give you the Zone V gray.

With this Zone V exposure value as a reference, you can then increase by 1 stop and get Zone VI exposure for the same wall. Increasing the stop by 1 again would give Zone VII. Vice versa for going from Zone V to Zone IV. The higher the Zone number, the lighter the tone. So Zone 1 is basically completely black, and Zone X is basically completely white. Theoretically, this is how it should work. I'm going to attempt the test this afternoon with a digital camera. This test with a film camera may work differently. I don't know if this test will even work with a digital camera, but theoretically it should work.

Anyway, this is why I was wondering exactly what a "stop" is, so I can try this zone test.

-Nick

This is all true, and basic, except that photographic paper can't actually depict 10 fulls tops of luminance, so absolute black in printing is Zone II and pure white in printing is Zone VIII or IX, depending ont he paper.

My point is that making these tests on a digital camera are pretty meaningless because there's no base from which to work, you don't have any image at all until you run it through software and a monitor, and the monitors are all calibrated differently. Hence, I suggest attempting it with some "bipolar" cards so you have a built-in, valid reference on each exposure.

Robt.