Why Should We Care About Crop Factor?

[amolitor]

Interestingly, if we assume that we're making the same size prints from both, CoC becomes smaller for crop sensors. You're enlarging more, so you must have a higher standard for "what's sharp" at the sensor. This causes the crop sensor to have somewhat less DoF. Other factors, however, dominate, and taken all together we generally consider the crop sensor to have, in general terms, more depth of field.

In general, cropping may be considered to reduce depth of field, and a crop sensor is, as has been noted, in essential ways precisely the same as shooting with a larger sensor, and then cropping the result.

 

[Ysarex84 78]

Interestingly, if we assume that we're making the same size prints from both, CoC becomes smaller for crop sensors. You're enlarging more, so you must have a higher standard for "what's sharp" at the sensor. This causes the crop sensor to have somewhat less DoF. Other factors, however, dominate, and taken all together we generally consider the crop sensor to have, in general terms, more depth of field.

In general, cropping may be considered to reduce depth of field, and a crop sensor is, as has been noted, in essential ways precisely the same as shooting with a larger sensor, and then cropping the result.

Yes. That can sound like it's backwards but the reduction in magnification when the photo is taken is the dominant factor. That's a factor enforced in practice by the requirement to use shorter focal length lenses. You ask the question what's a normal lens focal length and a 35mm film user is going to say 50mm. Now folks using an APS-C sensor think of a 50mm lens as a portrait length lens. When I shot 6x7 I had a 58mm Grandagon that was a super wide angle. When I shot 4x5 sheet my normal lens was 150mm and a 65mm lens was super-wide.

To use the smaller media we have to reduce the lens focal length to capture the same content. Putting the whole package together including making an enlargement we develop "standard" CoC values that we attach to the media size. It really does boil down to DOF = f/stop + magnification. Magnification here is tricky because it's a complex combination of magnification calculated as focal length and subject distance as well as magnification to enlarge the image to print.

Think of it this way: Again assume a standard scene -- this time a landscape with an horizon at infinity. Maximum DOF advantage will occur with the lens set at the hyperfocal distance for the f/stop used. To maintain the exact same content in the photo each reduction of the media size will require a shorter lens focal length and at the same f/stop the hyperfocal distance will keep moving closer to the camera. Hyperfocal distance is calculated based only on lens focal length, f/stop and CoC. The media size is a factor because it forces the use of a different lens focal length to maintain the same content. The result again is more DOF in practice using the smaller media. Again using the 5D and my Samsung as an example the hyperfocal distance for the 5D with a 50mm lens at f/8 is about 34 feet. For my Samsung with a 10.8mm lens at f/8 it's 1 and 1/2 feet.

Now think of it this way: I used to use an Arca view camera that was multi-format. I had a 4x5 back for the camera as well as 6x7 roll backs and even an adapter that would let me attach my Nikon camera body. Put a 75mm Super Angulon on that camera with the 4x5 back and you have a super-wide camera. With the lens set to f/11 and the subject at 17 feet you have DOF that reaches infinity. Hyperfocal distance for that lens at f/11 is 16 and 1/2 feet. Don't move the camera. Don't even re-focus the camera. Let's just swap off the 4x5 back for the 6x7 back. That 75mm lens is now a normal lens and the DOF no longer reaches infinity. It's the same lens. We haven't even touched it but now at f/11 there's only 40 feet of DOF when just a moment ago there was DOF reaching infinity. Next we'll again swap backs without so much as touching the lens. This time let's mount my Nikon F2 body. That 75mm is now a portrait lens and set to f/11 there's only 13 feet of DOF when a moment ago there was 40. Through this back swapping we haven't so much as touched the lens. The camera has remained stationary and not moved a mm. We're cropping the photo differently with the backs, but why should the DOF change at all? The lens hasn't been changed. Then lens hasn't been moved. The f/stop hasn't been moved. And the reason is because we're changing the CoC value with the size changes in recording media.

Now this last example makes you think it's the 4x5 that's got all the DOF and the 35mm is the one with less DOF. But in practice what we're really going to do is take the same photo with both cameras and to get approx. the same angle of view on the Nikon that the 75mm gives on the 4x5 we're going to mount a 24mm lens on the Nikon. On my little Samsung that's going to be a 6mm lens.

Joe

 

[timor]

That would be interesting experiment, but it won't work this way. DoF really is for the print, not a negative. All three negatives will have same CoC, now enlarge them to 16x20. Off course smaller negs gonna loose DoF much faster. In print from 4x5 CoC on the print will be enlarged only 4 times, from 35mm over 11 times. Forget sharpness with that.

Well, with DoF in the print one has to take in account the viewing distance to. I hope standards are set in such a way, that DoF, let say in 11x14 print, will look from FF and cropped sensor very similar.

 

[Ysarex84 78]

Well, with DoF in the print one has to take in account the viewing distance to. I hope standards are set in such a way, that DoF, let say in 11x14 print, will look from FF and cropped sensor very similar.

Of course you can't control how someone is going to look at a print. Some artists, when work is hung in a gallery will place a rope not only to keep folks from touching the work but also to help place them at a distance from which the artist wants them to view the print. The only study that I'm familiar with is one conducted using different size photos hung on gallery walls where the viewer's voluntary position was recorded. What they discovered is that within a standard range of 5x7 through 20x24 prints people on average would stand back a distance to view the photo that about equaled double the long side of the photo. If that's correct then we'd assume most people would comfortably stand back about 30 inches to view an 11x14 print.

Who gets together however in some kind of a standards organization meeting and decides for all of us what CoC values should be used by the industry is a mystery to me. I remember when I was first jarred into thinking about this was back in the late 70s when I discovered that the CoC value published by Nikon for Nikon DOF lens scales was different than the CoC value published by Leica. It was one of those WTF moments for me. I had to figure the Germans knew what they were doing and, especially since the Nikon value was more lax, those Japanese marketing bleeps were probably resting their thumbs on the scale to make it look like their lenses yielded more DOF than they really did. Now I know that wasn't the case; the Germans were just making bigger prints.

Kidding aside you could contact Don Fleming at DOF Master and ask him for authoritative references. Be interesting to know where he sourced his figures. I know he uses that Leica value for 35mm instead of the sloppy Nikon one.

Joe

 

[DiskoJoe]

I'm new to dSLR photography. I've had plenty of Point-n-Shoot cameras over the years, but have only recently jumped into and started to learn about the finer points of the dSLR realm.

I went with a MFT system (E-M5) because the IQ is great and the system is relatively light and compact. One of the first things I began reading/hearing about is Crop Factor. I've done a great deal of research and have yet to understand why anybody gives two hoots about it.

As I understand it, what we call Full Frame is in fact just a crop of the old 8x10 format, but you don't hear anybody talking about a 35mm lens being equivalent to 170mm on 8x10. (I'm totally making those numbers up!)

Truth be told, I actually find the whole thing rather confusing. The only information I've seen that offers an explanation about the significance of Crop Factor is that it gives Full Frame shooters a reference point to compare what Field of View will look like, but for those of us who aren't familiar with the 35mm FoV, why would we care?

Am I missing the point?

Thanks,

Shawn

Crop factor is important for DOF and also for distance that would need to be used for lighting. There is a lot of geometry that is involved with lighting and bouncing flash and the equivalent distance in an important factor to hitting the target properly.

 

[Ysarex84 78]

That would be interesting experiment, but it won't work this way. DoF really is for the print, not a negative. All three negatives will have same CoC, now enlarge them to 16x20. Off course smaller negs gonna loose DoF much faster. In print from 4x5 CoC on the print will be enlarged only 4 times, from 35mm over 11 times. Forget sharpness with that.

Number theory is just that -- theory. This only makes sense in practice. It's just like the algebra that you can use to calculate DOF limits. The math will always yield a number for the far limit, but at some point we just look at that number and say -- yo, that's infinity. So the whole has to be processed as a whole. Go to DOFMaster and look at Don's CoC table: Circles of Confusion for Digital Cameras Those values are all tied to sensor size because the CoC value is about the whole in practice. Run some simulations with the DOFMaster online calculator. The CoC value displays in the window and the only way you can get it to change is to alter the format/sensor size. He's got it right.

Joe

 

[timor]

I tried, but cannot run the simulation let say for 6x7 on lenses for large format. :meh:

 

[imagemaker46]

Crop factor is great when you are using a long lens and could use a little more. Crop factor sucks when you are using a wide angle and could use a little less.

 

[Steve5D]

Crop factor is great when you are using a long lens and could use a little more. Crop factor sucks when you are using a wide angle and could use a little less.

That's probably the best description I've ever read!

 

[Judobreaker]

DOF has nothing to do with the size of the sensor. The only reason people think it does is because you'd need to get closer to the subject with a full frame sensor to fill the frame. It's the different subject distance that changes the DOF, not the actual size of the sensor.

Well, you're wrong about that.

DOF depends upon 4 variables:
- Focal length
- Distance to subject
- Current Aperture
- Sensor resolution (or resolution of your print)

And what exactly focal length means depends upon sensor size. Thus DOF depends upon sensor size.

Wait... what? You're saying the focal length of my lens changes if I remove it from my D800 and place it on my D7000? That's interesting...

The focal length of a lens is always the same (unless it's a zoom lens, but then it is always the same range). The only thing changing with a different sensor size is the field of view, the physics don't magically change.
The only reason that the DOF varies is because people generally tend to get closer to the subject when they are using a larger sensor.

Crop factor is great when you are using a long lens and could use a little more. Crop factor sucks when you are using a wide angle and could use a little less.

My D800 has a more or less similar pixel density on its sensor as my D7000. This means that if I crop my D800 image in Photoshop with the same factor as a crop sensor I get more or less the same image. It's not the fact that the sensor is smaller that you can zoom in more, it's the pixel density that does the trick. Given the same density, the size of the sensor doesn't matter at all.

 

[Derrel]

>post 32>>>Think of it this way: Again assume a standard scene -- this time a landscape with an horizon at infinity. Maximum DOF advantage will occur with the lens set at the hyperfocal distance for the f/stop used. To maintain the exact same content in the photo each reduction of the media size will require a shorter lens focal length and at the same f/stop the hyperfocal distance will keep moving closer to the camera. Hyperfocal distance is calculated based only on lens focal length, f/stop and CoC. The media size is a factor because it forces the use of a different lens focal length to maintain the same content. The result again is more DOF in practice using the smaller media.> SNIP>
Joe

See, the process Joe described above was EXACTLY the reason Kodak invented the tiny "disc format" snapshot/purse/carefree film camera format back in the 1970's. Yes, Kodak invented a whole NEW format size as a way to get unlimited depth of field pictures for snapshooters! Using bigger film formats, like 126 cartridge, the minimum focus distance of the cameras at snapshot f/stop was around 4 feet, and people often shot too close, and got out of focus images. Same with 110 Instamatics with fixed focus lenses-those had focus in to about 3 feet, which was too long of a minimum focus distance! TOo many OOF snapshots!

When people photographed babies, or flowers, or close-ups of whatever using their 126 or 110 cartridge load cameras, there were MANY OOF images. Kodak realized, they needed to invent a super-SMALL format, and the Disc Format was born. Focused at 3 feet, with its ultra-short lens (I think it was 3.5mm or 3.7mm) and teeny-tiny film size, which is interestingly about the same size we today specify as 1/1.7" size (ie-more or less what we now think of as small-sensor digicam sized), the Kodak Disc Camera fixed-focus lens was set at 3 feet, which was the hyperfocal focusing distance for that lens and that size of film format at the single, fixed f/stop the cameras used, and BOOM! Deep of field from here to yonder!

FOCUS-FREE photography was achieved with the Kodak Disc Camera format--based ENTIRELY on "optics". The laws of optics. With that small a format and that short of a normal lens, there was NO need for focusing the lens...anything that the camera was aimed at would pretty much be in good focus. (This is basically the recounting of Modern Photography's former editor Herbert Keppler's account of how Kodak engineers invented the Disc format--as a way to solve an actual problem by REDUCING format size and going to an ULTRA-short focal length standard lens!)

 

[JacaRanda]

There **is a significant, noticeable and a very, real difference** between shooting with an APS-C or other smaller-sensor digital camera, and a 35mm-sized or 24mm x 36mm sensor camera( called FX by Nikon, or FF, AKA full-frame digital by many) as well as between a 35mm sensor size camera, and a medium-format rollfilm type camera, like say a 6x6 square camera like a Hasselblad. Mamiya, or Rollei,etc..

The simplest explanation is that the smaller the sensor or film size, the MORE depth of field there is at each picture "angle". Let's say we want to make a normal-lens picture--not an ultra-telephoto, tightly frames image, and not an ultra wide-angle panoramic image, just a normal picture. A point and shoot with a teeny-tiny sensor has almost infinite depth of field with its normal lens; an 8x10 view camera has very little depth of field with its normal lens; in between those extremes, there is MORE depth of field at each picture angle of view, as the sensor or film size gets smaller.

Here is one of the best articles I've seen on the issue. One that's accurate, and factual, unlike many pieces written about this issue.
Depth of Field and the Small-Sensor Digital Cameras - photo.net

SOME BULLET POINTS:

1. For an equivalent field of view, the small-sensor camera has at least 1.6x MOREdepth of field than a full-frame camera would have - when the focus distance is significantly less then the hyperfocal distance (but the full-frame format need a lens with 1.6x the focal length to give the same view).
2. Using the same lens on a small-sensor camera and a full-frame camera, the small-sensor image has 1.6x LESS depth of field than the full-frame image would have (but they would be different images since the field of view would be different)
3. If you use the same lens on a small-sensor camera and a full-frame camera and crop the full-frame image to give the same view as the digital image, the depth of field is IDENTICAL
4. If you use the same lens on a small-sensor camera and a full-frame camera, then shoot from different distances so that the view is the same, the small-sensor image will have 1.6x MORE DOF then the film image.
5. Close to the hyperfocal distance, the small-sensor camera has a much more than 1.6x the DOF of a full-frame camera. The hyperfocal distance of the small-sensor camera is 1.6x less than that of a full-frame camera.

3 & 4 are all I need to know or think about.

 

[imagemaker46]

DOF has nothing to do with the size of the sensor. The only reason people think it does is because you'd need to get closer to the subject with a full frame sensor to fill the frame. It's the different subject distance that changes the DOF, not the actual size of the sensor.

Well, you're wrong about that.

DOF depends upon 4 variables:
- Focal length
- Distance to subject
- Current Aperture
- Sensor resolution (or resolution of your print)

And what exactly focal length means depends upon sensor size. Thus DOF depends upon sensor size.

Wait... what? You're saying the focal length of my lens changes if I remove it from my D800 and place it on my D7000? That's interesting...

The focal length of a lens is always the same (unless it's a zoom lens, but then it is always the same range). The only thing changing with a different sensor size is the field of view, the physics don't magically change.
The only reason that the DOF varies is because people generally tend to get closer to the subject when they are using a larger sensor.

Crop factor is great when you are using a long lens and could use a little more. Crop factor sucks when you are using a wide angle and could use a little less.

My D800 has a more or less similar pixel density on its sensor as my D7000. This means that if I crop my D800 image in Photoshop with the same factor as a crop sensor I get more or less the same image. It's not the fact that the sensor is smaller that you can zoom in more, it's the pixel density that does the trick. Given the same density, the size of the sensor doesn't matter at all.

Is this what I said? If I use a camera that allows me the option of using my 300mm and gives me a 460mm to shoot with, or I use a camera with my 300mm that gives me a 300mm, and I crop it in photoshop I will end up with exactly the same size image as the first camera?

 

[Derrel]

Yes, those are pretty good ones. But number 5 is worth looking at closely; in the REAL WORLD what happens is that as sensors get smaller, and their corresponding lenses get shorter, in ACTUAL living situations, at home, at work, in the park, in the back yard, whenever-those small-sensor cameras and their short lenses approach, or hit, the hyperfocal distance, what those cameras create are deep depth of field pictures.

HERE is the issue, as I see it: With a crop-frame d-slr, when shooting indoors in a garage studio, to shoot a group of six people in a horizontal row, you need to use a short lens, like around 19mm to 28mm. WHat do you get? DEEP depth of field, so that the background paper or wall is IN SHARP FOCUS!

With an FX camera, and the SAME,exact group of people, you can just begin to throw the background paper out of focus, so it is less of a distraction. Moving to medium format film, the background will be well and truly de-focused.

One of the things that "non-photographers" often like are selective focus photographs. Those types of pictures are, for the most part, IMPOSSIBLE to make with small-sensor P&S or cellphone cameras.

One of the things that advanced street/journalist shooters LOVE is the ability to use a small-sensor camera, like a 4/3 sensor or even a compact P&S camera with good lens, to get HUGELY DEEP depth of fieldf pictures, without the need for super-critically placed,exacting focusing distance setting!!!!

THE kind of Henri Cartier-Bresson-like "street" images, the Garry Winogrand-like 28mm street images...that whole school of street reportage/documentary/decisive moment/near versus far kind of shooting, is made easier with a SMALLER-format sensor tool. When the smallish 4/3 sensor cameras first made by Olympus hit the market, the documentary/street/subway/restaurant/photojournalistic type shooters were SOOOOOOOOOO Happy, due to the small cameras, short lenses, and the deep depth of field type pictures that those TOOLS made possible, in actual, real-world places.

If you need to shoot across a cafe table AND still maintain sharp, clear, focus 20 feet behind, you CAN do that with an iPhone camera. You can NOT do that with an FX Nikon and 24mm lens. Nope.

 

[amolitor]

And what exactly focal length means depends upon sensor size. Thus DOF depends upon sensor size.

Wait... what? You're saying the focal length of my lens changes if I remove it from my D800 and place it on my D7000? That's interesting...

Actually, he didn't say that. Go read it again.