amehta Loc: Boston

Good, I get to practice explaining this without using any math. :-)

First, DoF is not an absolute value, it is a calculation based on a completely human quality: what is sharp enough? That's where the circle of confusion comes in, where we can start to tell that something is not sharp because it starts to have visible size, instead of being a point or razor thin line. We're not going to see this on a sensor, we're only going to see this on a displayed image. That CoC size is measured in the display setting: how large is the image, how far away is the viewer, how good is the viewer's eyesight. The viewer doesn't care whether the picture was taken on a FF sensor, APS-C sensor, medium format, or tiny cell phone camera. But, once a value for the CoC is determined, there is a corresponding size on each sensor. The images from different sensors had to be magnified different amounts to make the same size display, and that same magnification number shrinks the C (the CoC value on the sensor). Since the C is different, for the same focal length and aperture, the DoF is also different.

f8lee Loc: New Mexico

@lovitlots, it in fact turns out that sensor size DOES make a difference, because the amount of enlargement to achieve a given final image size increases as the sensor size decreases. The so-called "Circle of Confusion" limit is different. Of course, since different sensors (of the same physical dimension) can have different sized individual photo sites, I would think the calculation would have to take that into account as well as the CoC might differ depending on hoe many microns each photo site is.

But get a DoF calculator and check it out yourself.

SonyA580 Loc: FL in the winter & MN in the summer

Thanks to all for the interesting and informative comments! This old dog learned some new tricks.

GeneS Loc: Glendale,AZ

Try the calculator then try the preview button and a tape measure. this will show you the DOF and answer the question for you. The preview button shows you what is in focus

amehta Loc: Boston

Yes, and take some pictures of the tape measure at different apertures and distances.

SonyA580 Loc: FL in the winter & MN in the summer

OK guys. As soon as I get back to FL and thaw out, I'm going to do some research on DOF. Thanks again.

lovitlots Loc: Tottenham, Ontario, Canada

If the image in each view finder is the same then the dof is will be the same simply because the camera position will be different. We're talking about comparing the same picture produced by either camera with the same lens. If I put both cameras in the same spot and you have to enlarge the ff picture to match the cs picture the dof will still be the same for both pictures.

f8lee Loc: New Mexico

What you say is true when discussing film, but as I suggested use a DOF calculator (I used the DOF Master app on iOS) which is how I got the numbers I reported in my post above. Same focal length, aperture and distance from subject in both cases - the only difference was setting to FX and DX. To repeat myself, I believe this has to do with the fact that the DX image needs more enlargement to achieve the same final image size and so it's CoC needs to be smaller.

Yes, if I take a photo with a 14MM lens from 1000 yards and crop it to show exactly what a 600MM lens would show from the exact same position and aperture, the two images will be precisely the same (save for "grain"). But this is a different situation.

SharpShooter Loc: NorCal

Sony, I tried to warn you. I don't think you're gonna get outa here that easily.
Sit back, strap yourself in, and enjoy the ride. ;-)
SS

amehta Loc: Boston

Ok I put the following in the DoFMaster.com calculator:

D7000, f=50mm, f/4.0, d=15'; D = -1.90', +2.54'
D800, f=50mm, f/4.0, d=10'; D = -1.26', +1.68'

The DoF distances are the same percentage of the distance to the subject, but since the distance for the DX camera is larger, the DoF distances are larger.

amehta Loc: Boston

Yes, you've just put the D800 in DX mode to mimic the D7000. Then, yes, the DoF will be the same, since you're not actually using the whole sensor, you're using part of the sensor. It all comes back to "size matters".

lovitlots Loc: Tottenham, Ontario, Canada

I see what the calculator states but what I'm not sure of is does the distance you set both cameras up from the subject actually change by 1.5 times. I don't know the answer to that. I know what looks obvious but I have no way to test it nor can I find any info on the net.

aaronwolf

This wont be quick but it will be painless. When I started shooting 35mm in 1961 at age 12 I went into the library and started reading books on the laws and physics of optics. All of the basics have all but been forgotten and 99.9% of pro photographers don't know them. First of all what is a normal lens. This is simple. Remember the Pythagorean Theorum in high school geometry. That defines the length of the hypotenuse of a right triangle. a squared plus b squared equals c squared. The hypotenuse if you don't recall is the longest side of the right triangle it is the diagonal of your sensor frame. So with 35mm full format you have a 24mm by 36mm frame size and if you use the Theorum you will find that the diagonal length of that frame is about 43.24mm which is the absolute normal lens for full frame 35mm. You might ask why is it then that I have always been told it is 50mm. The answer is that it was by convention as its simpler to remember 50mm than 43.24mm. In fact Kodak was one of the last to follow that convention with its 35mm full frame film cameras using 45mm lenses on its Retina rangefinder cameras. The normal lens is regarded as the focal length that gives a perspective closest to human eyesight. Not the angle of inclusion but the closest perspective to human eyesight. The angle of inclusion of the human eye is much wider than the 43 degrees coverage of a 50mm lens. I think the best way to speak of perspective is to think of the concept of field compression that one sees with long telephoto lenses or the field decompression seen with wide angle lenses. The concept of field compression or flattening of the field plays in importantly in depth of field and I can talk about that in another post.
Now think of taking a normal 50mm lens and projecting the identical image on an APS-C sensor with a size of lets say
16x24mm. Using the Pythagorean theorem you can determine that a normal lens for that size sensor is 28.8 mm. So using a 50mm lens on this smaller sensor is using a lens which is longer than its normal lens and is therefore a short tele on the APS-C sensor. The crop factor is simply found by dividing 28.8 into 43.2 and that is 1.5. Bingo. Now think about a 3 dimensional subject projected onto a flat surface (our sensor). If that object is quite large such as a 20 foot long auto being photographed from 10 feet away and we are using a 50mm lens and we focus on the passenger in the front seat at say f2.8 we will not see a sharp grill in front nor the rear bumper of the car. The image seen by the sensor is in a single plane cut through a 3 dimensional projection. There is a blurry unfocused part of the projection in front of the sensor and a potential blurry projection behind the sensor which is not realized because the sensor is opaque. That miniaturized 3 dimensional projection of a 20 foot deep subject is just as blurry 1mm in front of or 1mm behind the sensor irrespective of whether the sensor is full frame, APS-C or a sheet of 4x5 inch film. So the depth of field of the projection at the sensor is unchanged however in comparing the full frame 35mm sensor with the APS-C we have a new perspective, i.e that of a 75mm lens rather than a 50mm lens. Things now look closer to the camera using the smaller sensor and this results in an APPARENT increase in depth of field in the final image because of the change in perspective. The depth of field of the miniature 3 dimensional projection at the sensor is unchanged in going from a full frame to a smaller sensor. Now since its xmas eve I will call it a night and hope that Santa sends me a Phase One medium format camera with an 80 megapixel sensor.....I wish.

SonyA580 Loc: FL in the winter & MN in the summer

Holy DOF Batman! I'm going to print this one and hang it on the wall. Thanks Aaron

amehta Loc: Boston

Here's how you can test it, if you have a zoom lens:
* Take a picture of a brick wall from 3 meters away (I3). Then take a second picture from 2 meters away (I2).
* On your computer, crop I3 by 1/3, so if the original is 4500x3000, the cropped would be 3000x2000 (I3c).
* Compare I3c and I2, both in fill-screen mode, so the difference in the megapixels is not a factor. If they look nearly identical, then moving 1.5x counteracts the 1.5x crop.

Do this in two ways: measure the distances from the back of the camera (the sensor) to the wall, and from the front lens to the wall.