[amoringello]

Nobody will know until we can see a sample and compare. This is not answerable from theory, just from practical experience.

This sort of ignorance is what is really at the heart of the matter.
This isn't magic. It isn't some form of ancient mysticism. It is physical science. It is knowable.

---------- Post added 03-17-16 at 09:31 AM ----------

OK guys, I posted a scenario.With calculations and results. There are lots, calculating the field of view by looking up the specs on the specific lens, calculating the distance from subject. Then the dof, etc.

This is why I am asking. No arguing, I really get the stuff on circle of confusion, etc. I was told to ignore the pixel to pixel stuff, I'm sorry but that is where I shoot. I know where I'm losing. I did that spreadsheet to figure out around severe and hard technical limitations.

But I also trust people's intuition or experience. People are describing something, and I want to know what it is. It comes down to physics. In my work if I find an anomaly I don't question Charles or Faraday, I look for other factors causing what I am seeing.

Not to argue, I'm really trying to help and explain in simple terms that are easy to figure out in anyone's head without deep math.

Take the DOF calculations for a crop sensor.
Take the DOF calculations for a FF sensor.

Now tape off the outer edges of your FF sensor with gaffer tape to match that of the crop sensor.

Without changing Focal length, aperture, or distance to the subject, etc...
What will the resulting image look like?
How will that compare the same image on the actual crop sensor?

Will the taped off areas magically introduce more light?
Or a change in DOF?
Will it cause more noise to be mystically introduced into those available pixels?

Quite obviously the answer is NO. No theories. Just simple real world physical properties.

---

Last edited by amoringello; 03-17-2016 at 06:32 AM.

[derekkite]

This is drifting form the original question, but DOF (Depth of Field) is a perception. Its calculation based on relatively common visual agreement. But true focus is a plane. There is no depth of focus in front of the lens.

DOF calculations are self consistent and can be used to roughly judge expectations, but they are not a law.

DOF will change if you enlarge or decrease the viewing size. It will even vary with an individual's eyesight. And is probably why people are getting so confused by FF vs Crop. They are failing to realize the FF sensor will capture a larger area of view than a crop sensor (with all else being equal). and that is affecting the calculations.

As the area of view gets larger (NOT focal length), the static elements in the view appear to become smaller. Thus DOF takes up less percentage of the sensor realestate an appears smaller.
Reverse with crop. As the area of view gets smaller, static elements in the view appears to take up more of the visible realestate and they, along with DOF, appear larger.

So if you take a photo with FF, it may print x 8x10. If you take the same photo (NO changes to focal length, aperture nor distance to subject) you may end up with a 5x7 (I'm too lazy to do the math for exact sizes, but you get the idea)

Put them on top of each other and the small print will fit perfectly and indistinguishably on top of the larger one.
BUT, blow up that small image to be the same size as the larger one, and suddenly things look larger. It looks as if you were in closer to the subject. You will see more noise, as it has also been enlarged. DOF will appear larger since the relative area that is in focus has also been enlarged.

It all comes down to the image circle coming from the lens.
Keep in mind what has to be done in the real world to the light and this stuff pretty much explains itself.
The size of the sensor cannot possibly affect the light coming from the lens. (again if physics allowed for that, the universe would probably have dissolved itself in an instant)



With that, on a crop sensor you have to open your aperture to give the same perception of DOF that would be seen on a given percentage of the surface area of a FF sensor. If you stop down on a FF you can get similar DOF as what is perceived from the smaller crop surface area.

** Again though, if you're comparing the same area (crop sensor vs. crop area within a larger sensor) no compensation needs to be done.
otherwise you're dealing with different variables, and this is not relevant to question in the initial post.



I am going to find it so sad when someone buys K1 and shoots using the Crop mode, yet expects that they will get better DOF merely because the sensor itself is larger.... ("but, but, but, the DOF chart says I get better images with a FF sensor and everyone says cropping out the middle does not affect that...") I find it disturbing that this false information is propagated with such zealousness without understanding simple principles.

So if you frame the shot the same way, either by distance or different lens the dof difference in that instance is only due to the lens focal length, not the sensor.

I see what you are saying. A photo of a field with a stripe across the middle of the frame in focus. Same shooting location, same lens. A FF shot compared to an apsc shot of the same size, say the viewing port in lightroom, the focused stripe will be narrower on the full frame shot. If you had measuring markers the actual width of the stripe would be about the same or seem a bit wider on FF, but be very close.

If you want the same proportion of the frame to be in focus you would have to stop down the FF.

Thanks.

[amoringello]

So if you frame the shot the same way, either by distance or different lens the dof difference in that instance is only due to the lens focal length, not the sensor.

I see what you are saying. A photo of a field with a stripe across the middle of the frame in focus. Same shooting location, same lens. A FF shot compared to an apsc shot of the same size, say the viewing port in lightroom, the focused stripe will be narrower on the full frame shot. If you had measuring markers the actual width of the stripe would be about the same or seem a bit wider on FF, but be very close.

If you want the same proportion of the frame to be in focus you would have to stop down the FF.

Thanks.

In short, correct. D

If when you say "same size" you do mean what is visible in the viewfinder then you must change Focal length or distance to subject, and this will require some other attribute to change if you wish to maintain balance.
Either you accept the relative change in DOF, or YES you can modify your aperture to put the DOF back in line with where it is desired. That could be done by stopping down the lens on a FF camera or opening the aperture on a crop camera.

It is all give and take. DOF is defined by Aperture, Focal Length and Distance to subject.
You cannot change one parameter without affecting another. But you can compensate and change other parameters to get things back. As a result when you change aperture you also affect the exposure triangle. So you may need modify shutter speed or ISO to maintain a proper exposure with the new aperture. This may or may not be acceptable.
And this is where you can start to decide if FF or Crop is going to best meet your needs.

[derekkite]

So if I have a critter in that stripe in focus, I'm not interested in the whole frame, and likely will crop. The detail comes from the pixels in focus, the number of pixels, and how many of them I have to throw away due to noise.

I don't have to stop down any more or less than apsc, I want the bear body in focus, it is 10 feet long, I'm at this distance, this is the fstop I need. I set the circle of confusion about the same size as the Bayer pixel.

I'm now interested in noise. Say my bear is 1600 pixels across on the K1, on my k3 it is 2000 pixels across, or a bit less. 4*4=16mpx, 5*5 =25mpx. Close enough for a napkin calculation. Every pixel that is noise is a loss of resolution. They are scattered throughout the image section, not randomly but for the sake of argument.

I have fewer pixels of the details I want. There are real factors outside the sensor physics that matter. Shutter speed is critical, likely higher Iso.

Are there better noise characteristics at a pixel level?

---------- Post added 03-17-16 at 07:18 AM ----------

Same size, viewing the image on the screen scaled to the screen size. Or printing it the same size, say 5*7 or whatever. And yes if you want the same scene, a different focal length will have different dof characteristics.

---

Last edited by derekkite; 03-17-2016 at 07:23 AM.

[amoringello]

So if I have a critter in that stripe in focus, I'm not interested in the whole frame, and likely will crop. The detail comes from the pixels in focus, the number of pixels, and how many of them I have to throw away due to noise.

I don't have to stop down any more or less than apsc, I want the bear body in focus, it is 10 feet long, I'm at this distance, this is the fstop I need. I set the circle of confusion about the same size as the Bayer pixel.

I'm now interested in noise. Say my bear is 1600 pixels across on the K1, on my k3 it is 2000 pixels across, or a bit less. 4*4=16mpx, 5*5 =25mpx. Close enough for a napkin calculation. Every pixel that is noise is a loss of resolution. They are scattered throughout the image section, not randomly but for the sake of argument.

I have fewer pixels of the details I want. There are real factors outside the sensor physics that matter. Shutter speed is critical, likely higher Iso.

Are there better noise characteristics at a pixel level?

---------- Post added 03-17-16 at 07:18 AM ----------

Same size, viewing the image on the screen scaled to the screen size. Or printing it the same size, say 5*7 or whatever. And yes if you want the same scene, a different focal length will have different dof characteristics.

I am not sure exactly what it is you're asking... but that sounds way out of scope from the original post and is probably best suited in a new thread.

[jsherman999]

This comment shows that you are not understanding the most basic concept of the issue.

Actually, he understands it fully.

...
Stop making it out to be more complex than it is. We're not cropping. we're not enlarging.
I have no idea where the concept of enlarging or cropping ever came in

The initial question was about using a crop lens on a FF camera - assumption being that they're cropping in post or letting the camera's auto-crop work, vs. printing or displaying images with blacked-out corners.

As far as 'enlarging' goes, unless you're actively trying to keep your crops smaller all the time in output - display or print - your crops are getting 'enlarged' when you view or print them. Unless you, say, crop an 8x10 to 5x7 and then leave it at 5x7 thereafter and make sure you never allow yourself to view it at the same size as the original was in the preview pane.

I will re-iterate - cropping (and enlarging) DOES change the DOF in the resulting image. You can test this yourself, I can demonstrate it, some of the links that have been provided have images that demonstrate it, and some of the links describe why this is happening in more detail. If you have doubts (and that's OK) please follow the links.

Cropping/enlarging also affects the perceived noise in the resulting image, which, again, you can test yourself or consult multiple external sources that confirm this.

So to track back to the OP's question - using a crop lens on a FF camera (assuming you're cropping the results to aps-c dimensions and displaying at the same size, which is why you asked) turns your FF camera into an aps-c camera, in noise, DR, DOF, and FOV. Please, please don't listen to anyone who tries to tell you otherwise.

If you buy a FF camera thinking that cropping to aps-c doesn't affect things because it's still an 'FF camera', you've just wasted your money.

Now, here's something to consider - some 'aps-c' lenses cover more than the aps-c image circle, so you may only need to crop a little in post, like 1.2x (Some Nikon cameras have a 1.2x and 5:4 or 3:2 auto-crop option built in just because of those lenses,) and in some lens cases you may not have to crop at all if you can live with some slight vignetting. In those cases, your 'aps-c' lens becomes just a FF lens with 1.5x wider FOV, but (probably) fairly weak edges.

.
.

[osv]

Same size, viewing the image on the screen scaled to the screen size. Or printing it the same size, say 5*7 or whatever. And yes if you want the same scene, a different focal length will have different dof characteristics.

yes, viewing it at different display sizes can appear to affect the dof, which adds to the confusion... the advanced version of the cambridge in color calculator, for example, takes print size and viewing distance into consideration A Flexible Depth of Field Calculator

combine that with cropping/enlarging and things become complicated in short order.

in certain situations, focal length can appear to change background blur radically, without affecting the actual dof itself:
"Background blur
A common source of confusion in DOF-related discussions is the issue of background blur. To illustrate the relation between DOF and background blur, or rather the lack of it, two photographs are presented in Fig. 3. Here, Gromit was captured with a 100-mm lens and a 28-mm lens on a 35-mm camera. The image magnification is approximately the same at M=0.12, and the F-number is f/4 in both cases. Despite the fact that Gromit appears at the same height, the two pictures are very different. First, the perspective differs as the photographer needed to come closer to the subject with the 28-mm lens. Second, the 28-mm lens shows more of the background because of the wider field of view. Third, the 100-mm lens seems to yield more background blur. (Does it?)
...Neither the absolute nor the relative background blur should be used as a criterion to judge the depth of field. Much of the confusion in DOF discussions arises because people base their judgment on out-of-focus parts of the image. DOF should not be judged from background blur. The luminous landscape puts considerable effort into examination of background blur, but in doing so the definition of DOF is forgotten."
Depth of field

[amoringello]

Actually, he understands it fully.


The initial question was about using a crop lens on a FF camera - assumption being that they're cropping in post or letting the camera's auto-crop work, vs. printing or displaying images with blacked-out corners.

Now you're just making stuff up!
Did you actually read the original post?
Your assumptions are never even mentioned in follow-up posts by the OP.


You have not comprehended the basic premise of the question.
Without that, everything else you say is simply white noise and tl;dr.
(further, you have no basis to confirm another person's understanding)

[Gimbal]

Now you're just making stuff up!
Did you actually read the original post?
Your assumptions are never even mentioned in follow-up posts by the OP.


You have not comprehended the basic premise of the question.

If that was so wrong, perhaps you could explain what the original question was really about?

[amoringello]

That has been covered over and over again.
Would my repeating myself once more make a difference?

[derekkite]

I plead stupidity. I have learned quite a bit from this discussion, and you expressing it again helps.

[enoeske]

That has been covered over and over again.
Would my repeating myself once more make a difference?

I, too, would like to see what you think the OP was about.

[amoringello]

I, too, would like to see what you think the OP was about.

Super! Glad to hear it!
Read the prior posts.

[redrockcoulee]

Actually, he understands it fully.



The initial question was about using a crop lens on a FF camera - assumption being that they're cropping in post or letting the camera's auto-crop work, vs. printing or displaying images with blacked-out corners.

As far as 'enlarging' goes, unless you're actively trying to keep your crops smaller all the time in output - display or print - your crops are getting 'enlarged' when you view or print them. Unless you, say, crop an 8x10 to 5x7 and then leave it at 5x7 thereafter and make sure you never allow yourself to view it at the same size as the original was in the preview pane.

I will re-iterate - cropping (and enlarging) DOES change the DOF in the resulting image. You can test this yourself, I can demonstrate it, some of the links that have been provided have images that demonstrate it, and some of the links describe why this is happening in more detail. If you have doubts (and that's OK) please follow the links.

Cropping/enlarging also affects the perceived noise in the resulting image, which, again, you can test yourself or consult multiple external sources that confirm this.

So to track back to the OP's question - using a crop lens on a FF camera (assuming you're cropping the results to aps-c dimensions and displaying at the same size, which is why you asked) turns your FF camera into an aps-c camera, in noise, DR, DOF, and FOV. Please, please don't listen to anyone who tries to tell you otherwise.

If you buy a FF camera thinking that cropping to aps-c doesn't affect things because it's still an 'FF camera', you've just wasted your money.

Now, here's something to consider - some 'aps-c' lenses cover more than the aps-c image circle, so you may only need to crop a little in post, like 1.2x (Some Nikon cameras have a 1.2x and 5:4 or 3:2 auto-crop option built in just because of those lenses,) and in some lens cases you may not have to crop at all if you can live with some slight vignetting. In those cases, your 'aps-c' lens becomes just a FF lens with 1.5x wider FOV, but (probably) fairly weak edges.

.
.

Seeing as you are using large format formats I think the point is if you take a photo with the 8X10 and then put on 5X7 back on the camera and then take a second photo, same lens and distane from subject, which would be better if you are only interested in the portion of the image that is covered by the 5X7 negative. If the lens, distance and film (sensor quality) are the same would it make any difference. No difference in DoF at all. If you trimmed off the 8X10 negative and printed off both the trimmed 8X10 and the full negative 5X7 to the same size on paper you would not be able to tell the difference apart, you would not even have to make adjustments on the enlarger, For the 8X10 you trimmed the image after the capture to match the 5X7.


Doing the same thing on digital with switching cameras for the one and using FF in crop mode for the other there would be no difference in DoF as enlargerment is not a factor, the image would be the same size depending only on sensor pixel density.

Does this relate to the question as that is how I read it.

[enoeske]

"Read the prior posts."

1.)

...And really no relevance to the Crop v.s. FF issue to begin with.

Okay, so you agree this is about crop vs FF...

2.)

The original post was regarding the use of a crop lens so the question is likely regarding the crop area of the K1 sensor, only.

Okay, you also agree that by mentioning a crop lens, we are talking about the cropped portion of the FF sensor...

3.)

...I don't know why some people get hung up on enlarging an image when comparing effects of light on the FF sensor vs crop.
Quality within the same area on a FF vs Crop has NOTHING to do with enlarging an image.

But I'm viewing them at the same size on my monitor -- so the smaller (crop) image DOES have to be enlarged. Otherwise I could never view it at the same (full screen) size as the FF image.

4.)

We're not cropping. we're not enlarging.
...
But AGAIN, that has NO relevance to the initial post.

Wait, yes it does. You said so in your first post. This IS about cropping. And by your third post, its also about enlarging.

5.

...Until the basic concept is understood, all these arguments talking about cropping and enlarging have absolutely no bearing or relevance.

umm...but...that's all this is about.

6.)

Using only the area of the cropped sensor, will result in an identical image. (minus any difference in technology...

Yes, that is what we said just a few posts into the thread. There won't be much, if any, improvement to using The K-1 in crop mode instead of the K-5iis.

...So if you take a photo with FF, it may print x 8x10. If you take the same photo (NO changes to focal length, aperture nor distance to subject) you may end up with a 5x7 (I'm too lazy to do the math for exact sizes, but you get the idea)

Put them on top of each other and the small print will fit perfectly and indistinguishably on top of the larger one.
BUT, blow up that small image to be the same size as the larger one, and suddenly things look larger. It looks as if you were in closer to the subject. You will see more noise, as it has also been enlarged. DOF will appear larger since the relative area that is in focus has also been enlarged.

So...cropping and enlarging?

I'd would like you to clarify what you think the OP is about because you are sending all kinds of mixed messages.