nsbca wrote:
Wait and watch. Why would I have to cut off the feet or the head?
Because you cannot take the same shot of the same subject with the same lens, at the same distance with the same settings and all, but with different sensors and end up with the same image, unless you crop the larger image down to the smaller sensor size. Assuming that you've used your full-frame camera properly, you will have filled the frame with your desired image. If your intention is to trim every full-frame image to a 1.6X crop, why use full-frame?
I understand what you're saying - you want to take a full-frame image and crop it to the same size as the 1.6X sensor to compare the two equally-framed images. And they will be identical (+/-), as expected. But that's not how we use cameras. We frame a shot to take advantage of the entire image frame of the camera. And when we do so, we will need to use a 50 mm lens on our 1.6X camera to give a similar FOV to the 85 mm lens we might use on our full-frame camera. And when you do that, you'll have to open the aperture about 1.5 stops (roughly) to get the same shallow depth of field.
10DFT wrote:
[ you want to take a full-frame image and crop it to the same size as the 1.6X sensor to compare the two equally-framed images. And they will be identical (+/-), as expected. But that's not how we use cameras.
That is what I am saying. And I do understand what you and Jeff are saying. But when I take a picture I need a standard in my head. The DOF scale on the lens (subconsciously now) is that standard for me.
When I am out in the field with my camera I don't think about how big I will enlarge the image or how far away I will have to stand looking at it. I think how sharp will this portion of the image be at 100% crop. My mind works that way especialy with wildlife. When I am sitting in a fixed blind and a 600mm lens with my subject at 100 feet (lets say a Cardinal) the image, the DOF, the subject, everything will be near to identical on an 800x800 pixel crop from subject of an image taken with the 1D Mk II and one taken with the elusive 5D FF.
Do your math and prove me wrong.
I can't readily move the blind, I can't move the bird and I can't change the focal length, so for me those variables that keep getting thown into the mix are non-existant.
nsbca wrote:
... But when I take a picture I need a standard in my head. The DOF scale on the lens (subconsciously now) is that standard for me.
When I am out in the field with my camera I don't think about how big I will enlarge the image or how far away I will have to stand looking at it. I think how sharp will this portion of the image be at 100% crop. My mind works that way especialy with wildlife. When I am sitting in a fixed blind and a 600mm lens with my subject at 100 feet (lets say a Cardinal) the image, the DOF, the subject, everything will be near to identical on a 100% crop from a image taken with the 1D Mk II and one taken with the elusive 5D FF.
...Show more →
You've described a very rational approach to the limitations of how you take pictures of cardinals etc and how you will evaluate the images when you get to see them on your computer screen, as many do, i.e., at 100%. Firstly, the DOF scales will be of little use to you because they were not calculated for your situation. Secondly, if the two hypothetical cameras (e.g. 1.6x and FF or 1.3x and FF) had the same pixel pitch, a cardinal which created a 5mm high image on either sensor and cropped to the same size on screen should look close to identical (assuming similarly high quality pixels, similar processing etc.). This is your film analogy which all have agreed is not flawed - let the same image fall on a piece of film in two different camera backs and you get the same image and same DoF , no matter how you define it. With a fixed lens, fixed subject distance you will always get the same image coming out of it and your way of looking at the situation is totally rational.
The big but is now easy to explain - that is not how the DoF scales were calculated , long ago before digital and 100% crops came along. If you go to 50% or 33% view, things will appear sharper, no? Also, if you move back, you get the same effect; things will be smaller but more items in the image will appear sharp. The whole discussion on this thread has been about this concept. If you have decided to look at 100% on a monitor with (e.g.) 90 pixels per inch resolution and your eye is 20 inches from the screen, you could make your own DoF scale empirically. Shoot an inclined ruler at 100 ft at different f stops and JUDGE FOR YOURSELF which marks are ACCEPTABLY sharp at 100% view and voila. This is the only DoF scale that will be of interest to you. I have the 500IS so unfortunately I couldn't use it
Early on the 35 mm camera makers (roughly ) did pretty much the same thing - by defining how they would print and view the image : They agreed that taking a whole 35mm frame, printing to ~8x10 and viewing at ~12 inches was a common enough thing for people to do, that they created the DoF scales you're familiar with. Pretty different from your parameters but the same concept, right? Note that the DoF scales would not have been correct predictors of what appeared to be in focus if someone had printed at 16x20 and looked at the print from ~12 inches.
For your comparison, however, the two cameras will not necessarily have the same pixel pitch, so the 100% crops of this constant 5mm high cardinal image will be different sizes on screen. The bird will be 900 pixels high from one camera and maybe 1100 from another (still with this fixed image from the hide/lens/bird set-up - if it's still there ). The bigger image will have a (slightly) smaller DoF as a result of the greater magnification - like moving closer than the "standard" 20". Jon Wienke has suggested, IIRC, that these days it makes some sense to consider the pixel pitch as the CoC to deal with this issue.
A bit of a digression, but I hope it helps illustrate why folks with the original DoF concept didn't agree with some of your logic. BTW, do you ever print your images
AJSJones wrote:
BTW, do you ever print your images
Quite often, but I rarely think of print size when I'm out with the camera. Resolution, sharpeness, subject matter, crop, custom frame prices and venue are more often contributing factors in the resultant print size then any thoughts that were going through my head when I set the aperture.
slin100 wrote:
Encapture, I do not agree with this explanation.
In an image technically only one distance is in perfect focus. All other distances are out of focus by various amounts. The near and far distances at which the blur is perceived as being out of focus is what defines DOF. Therefore, magnification plays a central role in determining DOF because it directly affects the perceptibility of blur.
DOF is not just some fixed quantity inherent to a lens. DOF is entirely a subjective matter. The DOF scale on a lens is set based on the manufacturer's judgement but someone with a more discerning eye might not agree and move the DOF markings closer together.
A practical consequence of the effect of sensor size on DOF is that one should use the DOF markings on a lens approximately 1 1/3 stops wider than the actual aperture when using the lens on a 1.6x crop camera. ...Show more →
You forgot viewing distance. If you keep the viewing distance proportionally the same (which is normal the case) then only the big three affect DOF.
nsbca wrote:
What has anything you have just said have to do with Depth of Field? If I am standing 20' from a 50' high white plaster wall with my 180mm 3.5 mounted on a tripod set at f/3.5 and I first attach my 1Ds to the lens and take an image. Then, leaving the lens set the way it is and the tripod in the same position, I take another image with my 1D Mark II has the DOF changed?
DOF, I will repeat, is the area in front of and behind the focus point that is in focus.
Lastly, how come DOF scales ask for focal length, f/stop setting and distance from subject to arrive at an acurate DOF. Never once in my life did I ever see a DOF scale the asked how large the intended print size is....Show more →
Well said. Any argument counter to this is erroneous.
You guys are too funny, no wonder so many of the intelligent canonites have been spending so much time over at Nikon.
It's astounding that so many know so little!
I can't believe that there are still people comparing croping in PS to straight from camera.
DOF scales also ask film format as well by the way.
Cheers,
Jasin.
jasin wrote:
You guys are too funny, no wonder so many of the intelligent canonites have been spending so much time over at Nikon.
It's astounding that so many know so little!
I can't believe that there are still people comparing croping in PS to straight from camera.
DOF scales also ask film format as well by the way.
Cheers,
Jasin.
You guys over there at Nikon are too funny. Why can't you all talk about something more then Canon?
Keith Wong wrote:
I think I need to refocus the debate.
The fact is - larger sensors are better. I'm sure that all of you believe this to some extent. This is one of the reasons why our DSLR's outperform P&S cameras, and why digital MF backs outperform DSLR's.
The point of contention here (I think) is whether a FF sensor has enough of a performance advantage over a 1.5 crop sensor. The two are not dramatically different in terms of size - not to the same magnitude as say, comparing a DSLR sensor to a P&S sensor. What's more, there are other ways in which to improve the performance of smaller photosites - better microlens and electronics design. I can see where the disagreement is, and I don't think it's going to be resolved one way or another in the short term.
I personally want a FF sensor, except that I can not get over the cost hurdle just yet. If the 5D turns out to be all that has been rumoured, it will definitely be on my shopping list. ...Show more →
jasin wrote:
Huh! hey Keith I see the LARGER sensor really works for you!
Oh thats right you don't have one so what gives you the right to comment.
Stick to reading buddy!
Oh and Alan, right on Tiger!
Cheers,
Jasin.
Seems the 5D is all the talk in Nikon land as well.
All those who believe DOF is controlled by Aperture, Subject Distance and Focal Length only, move to the left of the room<
OK.
Now all those who think that the sensor size also affects the DOF move to the right>
OK.
All those on the left of the room are correct. All those on the right of the room, are well, incorrect.
That's just the way it is, you can't change the laws of Physics - if you are on the right and still think you are correct - sorry!
The bit you seem to be stubbornly missing is that the size of the final print is a factor. DOF calculations all assume a standard print size and viewing from a standard distance. If you shoot with a cropped sensor (and for this purpose it doesn't matter if you use a DSLR or an APS-C camera and digitize the negative) then you need to enlarge more to get your standard print size. This extra enlargement reduces DOF.
CW did you get your FACTS from the back of a pack of honey smacks or what?
I think I will leave it to you guys to work out , oh and nsbca don't you mean F5?
Cheers,
Jasin.
Oh and CW sorry I lean to the right but try this on for size!
Why smaller image sensors have greater DOF
1. For a given field of view (picture angle), the lens focal length must be proportional to the diagonal of the image sensor. For example, the diagonal of 35 mm film is 43 mm. A lens with focal of 43 mm, mounted in a 35 mm film camera, will provide a field of view of 53º. To obtain the same field of view (53º) in a sensor witch diagonal is half that of 35 mm film, we must use a lens witch focal length is 43/2 mm. (By the way, a lens with focal length that is equal to the diagonal of the sensor is called a “normal lens”, as it provides a field of view that is approximately the same as the human eye).
2. In the formula of the hyperfocal distance, the diameter of COC must be proportional to the size of the image sensor. The reason for this is obvious: the image from a smaller sensor must be more enlarged than an image from a larger sensor. So, in order to obtain the same COC in the printed image, the COC (in the sensor image) must be smaller for the smaller sensor.
From what was told in the previous paragraphs, it can be demonstrated that, for a given lens aperture and a given field of view, the hyperfocal distance is directly proportional to the sensor size (that is, a smaller sensor has smaller hyperfocal distance and, as a consequence, larger DOF). I leave the demonstration of this statement as an exercise to the reader.
cwphoto wrote:
That's just the way it is, you can't change the laws of Physics - if you are on the right and still think you are correct - sorry!
With a Horseman kit in your sig, I'd expect you to get it right
The whole point of this thread has been that DoF is NOT a law of physics. It is a result of the properties of the human eye to perceive what is acceptablysharp and its limitations.
It is NOT an intrinsic property of the captured (on the film or on the hard disk) image, it is a combination of the captured image and how it is presented and seen. If I were to print a 4x6 image of a receding fence taken with shallow DoF (aka some in focus and some out of focus) on a 16 MP FF camera for example and ask you which fence posts were acceptably sharp, you would give a different answer than someone looking at a 100% view on screen or on a 16 x 24 print.
Here we go
If you don't look at an image it has no DoF.
OK
In order to determine DoF you have to look at the image.
OK How you look at it will determine what you see as sharp.
OK?
Or not?
AJSJones wrote:
The whole point of this thread has been that DoF is NOT a law of physics. It is a result of the properties of the human eye to perceive what is acceptablysharp and its limitations.
Does that then mean that your 8x10 print will only have half as much DOF when viewed by a half-blind person
Perhaps the definition of DOF is just too narrow and there should be more than one definition. You are taking DOF from a three-dimensional world and putting it onto a two dimensional piece of paper. Sounds to me like you just lost all your DOF. Unless you are using three-dimensional paper.
Seems to me like thousands upon thousands of photographers have been producing beautiful photographs for the last 100 years without having to worry the least bit about all this technical jargon over DOF.
Sure, you can teach students all this technical stuff about sensor size, lenses, COC, etc., but so what? A 747 pilot does not need to know how everything on a 747 works to be a good pilot or to operate that model plane efficiently. I am sure many photographers have been using DOF very efficiently without having to know a single fact about COC.
Whether or not sensor size changes DOF is really quite insignificant in the larger picture. The photographer has far more control over DOF through the use of other factors.
Imagemaster wrote:
Perhaps the definition of DOF is just too narrow and there should be more than one definition. You are taking DOF from a three-dimensional world and putting it onto a two dimensional piece of paper. Sounds to me like you just lost all your DOF. Unless you are using three-dimensional paper.
DOF would still apply in this case, but you'd need a whole bunch of new formulae.
Seems to me like thousands upon thousands of photographers have been producing beautiful photographs for the last 100 years without having to worry the least bit about all this technical jargon over DOF.
Many of these same photographers came up with the math and/or were deeply interested in the amount of DOF they wanted in their prints.
Sure, you can teach students all this technical stuff about sensor size, lenses, COC, etc., but so what? A 747 pilot does not need to know how everything on a 747 works to be a good pilot or to operate that model plane efficiently. I am sure many photographers have been using DOF very efficiently without having to know a single fact about COC.
You would be utterly amazed by the amount of technical detail a 747 pilot needs to know about his aircraft before he's allowed to fly it. The theory of how it flies and the circumsances in which it will not are definitely on the list.
Whether or not sensor size changes DOF is really quite insignificant in the larger picture. The photographer has far more control over DOF through the use of other factors.
The photographer has no control over sensor size (except by cropping the completed image to give an effective reduction) but the bigest factors are always focal length, aperture and distance. The OP wanted to know if these were the only factors though and the answer is no.
). The bigger image will have a (slightly) smaller DoF as a result of the greater magnification - like moving closer than the "standard" 20". Jon Wienke has suggested, IIRC, that these days it makes some sense to consider the pixel pitch as the CoC to deal with this issue.
