In last week's video on why 50mm Prime Lenses are NOT Portrait Lenses I had a ton of questions about the crop factor of smaller sensor DSLRs. This video will show you what is actually happening with your camera and dispel the myth that your lenses are magically longer with a crop sensor. Keep shooting!
Nikon D3s on DX Mode
Nikon D3s on FX Mode
Nikon D3s on FX Mode, Cropped
Comments
Greg I would like to thank you for your great videos. I enjoy them very much, thanks.
I sure would hate it if it happened anyway, considering everything fits in my bag nicely now...
Have you tried the same experiment with a real DX body instead of simulating it on a FX body? I'm wondering if the body construction differences could interfeere over the final result.
Thanks for the video, it is great!
Maybe an alternative, or additional, way of showing the difference would be to take a photo on a FX camera with a 300mm lens, and one in DX mode at 200mm. The difference in DoF should show that the two are NOT that same.
a lot of people don't get that you are only seeing a smaller part of the same image on a crop, and when you look at it on the screen it looks larger because you are not seeing all of the available image just the inner part. It's a good illusion of extra reach with a lens.
He says that the size of the chip does not make the lens longer. That is true, however its never really explained that the crop factor depends on the chip size, and on the fact that the images are displayed in the same size relative to each other. Also, showing the full-frame image and the cropped small-frame image is bad, because it confuses people. He should have shown the uncropped FX version, and the uncropped DX version. Then you see that the latter image is indeed zoomed in. Thats a fact. Its also a fact that this is not due to the lens getting "longer" but to the chip measuring a smaller area of the image that is displayed by the lens.
Also, I don't see differences in bokeh. Thats probably because he used one and the same camera, but I'm not sure about that.
All in all, I don't see the problem with crop-lenses. I use a CF 2.0 olympus myself, and I never get the full-frame machismo of the nikon/canon people. If I don't have to print my images on 3ft canvas, I don't need all that frame size. I like shooting wildlife and love to have the extra "zoom" on my 200mm lens by the crop factor. Its all about what you want to do and not whats the biggest camera.
If my lens-supplier states, that my 50-200mm lens is 100-400mm EQUIVALENT in fullframe, then what's wrong about that statement?
The real interesting thing would be, what else changes (DoF, distortion), but that was not covered in the video, because you used one and the same lens, just changed the chip-size.
You confused me, because the title of the video and the first few minutes suggested, that a 50 mil lens produces the same images on full frame and on 2x (1.3x, 1.6x etc.), which it DOES NOT, apparently. So sorry, but I didn't get the point of the video...
Nicely explained, you now deed to explain some of the differences between FX and DX, done to reduce the size of lens, cameras, weight and costs with digital technology.
Kevin
I think you dropped the ball here in terms of explaining the objective of the video.
It is brilliant that you show what the actual effect of different sensor sizes is.
However, what I believe you should have explained is that the lens perspective does not change with the small sensor. You show that with the comparison. Consequently, the 50mm is not giving you the right perspective for head shots even with a small sensor.
On the other hand, the net effect on the small sensor is a shallower angle of view, which is achieved by using a longer focal length. So, you did not bust any myth, since there is not any, but you demonstrate how the facts have been misinterpreted all along. Think along the lines of "How to lie with statistics".
But, you did not take into account pixel density.
You can lay the part of the image you are interes5ed in over more pixels on a DX versus an FX camera,
D7000 versus D700 for instance, using the same lens.
This is not a lens proprity, gut a camera properity.
What actually changes is the angle of view.
I have wondered what the IQ difference would be between the cropped full frame v/s the smaller sensor...not exact calculations but in theory for example a 23 MP picture on a full frame cropped to the 1.5 or 6 (loosing about a third by cropping) compared to an un-cropped picture of a smaller sensor of approximately 18 MP.
I couldn't find any EXIF material on your examples...
Thanks, Sabe
In order to take identically framed photo in FX you need to be closer to the subject (more distortion) compared to the DX (less distortion).
Excellent demonstration.
Is it safe to say that this is sort of what happens with a point and shoot when one uses the "digital zoom"? It's actually cropped, and "resized" in the point and shoot camera.
In the DX mode, the "cropping" happens on the sensor so to speak because of a smaller area of exposure on the sensor surface.
In simpler terms; like having a smaller picture frame to fit 2 full size photos on,so you have to "cut" one picture down to size to fit the smaller frame giving the illusion of zoom on the same subject.
Does that accurately describe it?
Thanks, Perry
You will see that the DX lens doesn't "zoom in closer", or "multiply 1.5x" I.e. making a 300mm equivalent to a 450mm lens, rather you lose zoom in the demonstrator. This is a useful tool to determine what lens I buy in the future.
I think Greg has belabored the point that a DX lens and camera are not the same class as an FX platform, and what really happens with the DX sensor is that it "crops" the image much in the way that the "digital zoom" on a point and shoot camera does, giving the illusion of zoom. There is a reason that FX cameras sell for a higher price than the DX cameras do. The FX camera has a larger sensor than the DX camera, therefore it's able to capture more information and yield a higher resolution and even give a closer capture than it's DX counterpart using the same size MM lens.
I don't claim to know everything about photography, that's why I follow Greg Cazillo's lessons. I have learned a great deal from his tips, and instead of debating him for weeks on end trying to prove him wrong, I actually get out there (when my camera gets back from it's annual cleaning), and I put what I've learned to use.
I realize that Greg is a portrait photographer for the most part, and most of his tips are geared toward that premise, and I am an outdoor/nature and wild life photographer; so some things are going to be a little different in my approach and settings than someone in a studio with their subject sitting on a stool or standing in front of them.
I just think that much more can be learned if you open your eyes and ears and shut your mouth. If you don't agree with Greg, please get your own website and "enlighten us", or go on emulating some ditzy brunette who's a fashion train wreck on a lame car insurance commercial, but please let's move on already.
Let me add here: yes, I know that FX and DX are different things, and that larger CCD-chips are more expensive than smaller ones. My whole point was that there is a difference between cropping the sensor and zooming digitally, and using a "real" crop-sensor camera with real crop-lenses.
One more thing: I have great respect for what Greg is doing and the time he's investing into those videos. But that does not mean that one should not mention possible mistakes or inaccuracies.
I like to learn new stuff, and get into details of things - sorry for annoying you. But then again, you wouldn't have to read the comments section now would you?
http://www.youtube.com/watch?v=_uUH27HAybU
the "equivalent" is the important thing. 50mm FX will be 100mm equivalent on DX in terms of the view angle. But what about image quality? As far as I understand it, DX lenses are made for DX bodies for a reason. Me would interest if chromatic aberration becomes a problem when you use 2x or 1.6x crop factors on those lenses...
I am in total agreement and you have proven it well.
However there is one other factor (at least) that should be considered.
For a full frame 20MegaPixel sensor you are covering the full lens area with 20MP. If you have a crop sensor that is also 20MegaPixels then you are covering the cropped area of the lens with the same 20MP, which potentially (italicized) gives you the ability to create the same physical size picture (in. x in.) as if you had zoomed in more or used a longer focal length lens on a non-cropped sensor. However, you are still only getting a center portion of the glass and then enlarging that center portion - and the glass is what delivers the light for the photo to the sensor and that's why the glass is so important.
Relatively it is the density of megapixels vs physical sensor size for which people indicate that a 50mm lens on a 1.5x sensor is like (italicized) a 75mm lens for a full frame sensor.
But when you are enlarging digital pixels information or anything really you will lose quality and effective clarity. However depending on the end usage for the media it may not be that big of a deal.
I think you said it below a little differently but yes...
Equivalent View Angle does not mean Equivalent Image Quality.
Maybe it's more clear to exaggerate what's happening to understand it better:
If I made a 20MP sensor that has a 1000x crop factor (super tiny chip). My 50mm lens is like a 50000mm lens but I'm probably not going to get good details of the lunar surface when I enlarge the image to visible proportion.
When you are dealing with ratios closer than this outrageous example it might not be so pronounced but the image qualities will be different. Are they 5% different? 10% different? Noticeably different to your intended audience? If they are noticeably different then it might not be accurate to say the lens is 1.5x longer range than it really is.
Your point is absolutely valid for still photography; however, when shooting video with a DSLR, you get the benefit of a quasi-longer lens because it's not practical to crop every frame of a video.
John