Can anyone answer this question?

So can anyone tell me when you put a lens that is designed for a full frame camera on a crop sensor body like the Canon T2i is the entire glass is being used?

I got this answer from titaniumslug on YouTube which sounds like the correct answer:

It does only use the centre of full frame glass. This has the advantage that light fall off (vignetting) & softness in the corners at large apertures will be cut off. However overall sharpness (in stills only) will be less on a crop sensor. If the two cameras have the same number megapixels then the crop sensor will have smaller pixels which means it has a smaller area of the glass to deliver sharpness for each pixel. This is why on the dpreview tests the same lens scores higher on full frame.

Royalty free music by PremiumBeat.com

Products Used In This Video

Help me make more of these types of videos by purchasing gear from the links posted on my site. It costs you nothing extra, and helps support me to make more videos.

Used in this video

22 comments

  • Hey Dave,
    I’m not sure about the sharpness question but I looked at making 360 panos on a my T2i and the crop censor is very evident when it comes to glass real estate.

    At 5:15 in this video there is a screen view showing a full frame glass on a full frame camera and a crop censor camera.
    http://www.youtube.com/watch?annotation_id=annotation_270372&v=mUiw3jtErxk&src_vid=ouOEM4cKKGc&feature=iv

    I think it would use all the glass but the image would be cropped. – quality of the glass just not the full potential.

  • Sorry that last sentence doesn’t make sense.

    I think a crop censor camera would use all the light info but wouldn’t show a full image.

  • Hey Dave – as a newbie can you explain “using all of the glass”? I get it’s related to the edge but can you dig a bit deeper into it so I know what to look for?

    Thanks! – Jason

  • On a crop sensor camera, only a portion of the image captured by a full frame lens will actually fall on the sensor. It also means the image output from the camera is magnified, on a t2i, times 1.6. I don’t agree with the previous answer on sharpness. The t2i has an 18 Mp sensor. The 5d has a larger sensor, spread over a larger area of real estate. Note that a 24-70 lens designed for a crop sensor camera should delivera focal length of 24-70 while a full frame 24-70 lens comes out to (24*1.6-70*1.6) or 38.4mm-112mm.

  • One thing to note, for video, output is o ly about 2mp per frame, so the video image quality of a t2i is comperable to a 5d, unless you are in very low light, then the superior sensor on the 5d shines through. For still images, the difference in image quality is more apparent.

  • @Jason,

    Think of it this way:

    Your looking out your window (lens) and you see everything, the clouds in the sky to the fresh cut grass. In this case you’d be seeing through all of the window (using all of the glass).
    Then someone comes along and places a piece of cardboard with a small hole cut in the center over your window. With the cardboard you can no longer see the clouds and grass because your not looking through the full window. The only view is through the little hole. (Your not using the full glass)

    Crop sensor cameras (like Dave’s T2i) have the same principle. Think of the window as the sensor.

    Hope that helps,
    John-Mark

  • Consider a single point in the subject in view – the tip of a leaf, for example. All light rays which emanate from that point, and which strike ANYWHERE on the front of the lens, are refracted by the lens so that they all fall on a single point on the film/sensor plane. That’s the magic of what a lens does, and that’s what makes the image. For a point which appears in both the full-frame and crop-sensor images, exactly the same glass is used to create that point, assuming you are shooting at the same aperture in both cases. So, all the glass is used, regardless of the size of the sensor, for each point that appears in the image. It is only stopping down, ultimately to a pinhole, that reduces how much glass is used in making the image.

    There may be differences in the sensors, such as quality, pixel pitch, etc., which cause one to *capture* the image better than the other, but the image which falls on the crop sensor is exactly the same as the image which falls on the corresponding portion of the full-frame sensor. The same “glass” is being used for both.

  • Okay that makes sense but why does a crop sensor “crop” a fish eye and not a full frame lens?

  • Hi guys,

    “Cropped Sensor” cameras DO NOT use all the glass on a lens designed for a “Full Frame” sensor. The easiest way to envision it is by looking at this image, it does a good job of illustrating it:

    http://upload.wikimedia.org/wikipedia/commons/4/49/LensCropFactor.png

    Also, look at lens reviews for FF lenses, such as this one:

    http://www.dpreview.com/lensreviews/nikon_50_1p4g_n15/3

    They always test it on both a FF body and a “cropped sensor” body, explaining the difference in performance as it relates to the sensor not utilizing the edges of the lens glass.

  • @Jamie,

    that wiki pic is what I thought a cropped sensor did. (Like my window explanation. you can’t see the grass or the clouds. just the tree.)

    Thanks for sharing,

  • I’ll just say again that for a given lens, wide open, the only difference is that on a crop sensor camera the lens doesn’t have to be good at handling light rays originating from the periphery of the field of view – because those areas of the field won’t be “seen” by the crop sensor.

    But it’s still true that all of the glass is used by both cameras. Every point on the front element of the lens receives rays from every point in the field of view, and the lens refracts them so that the image is formed on the sensor. The ray which emanates from the very center of the field of view and strikes the very outer edge of the lens front element is refracted so that it strikes the very center of the sensor, regardless of the sensor size. The same thing happens to all the rays of light which emanate from that point in the field and strike anywhere on the front element. They all get refracted to fall on the center of the sensor. So, all of the glass is used.

    Jamie, I don’t think your wiki drawing reflects how a lens actually works. The optics section of a good physics textbook would show it better. Or, the Wikipedia entry for lens (optics) has a drawing about 1/3 the way into the entry, under “Imaging Properties” which shows multiple rays of light from a single point striking different points on the lens.

  • @Wayback

    I don’t think we’re disagreeing with you.
    Yes all the glass is used but the net result (on a crop) is that less glass (or info) is used.

    Look at the video link above. (5:15) though the lens “technically” uses the full glass it crops the top and bottom of the image. ( light rays travel in straight lines)

    My physics book has an image that is like the one mentioned.

    When working with a cropped sensor the net result is that the information from the top and bottom of the glass is not gathered.
    If it was gathered then I’d have a full frame sensor.

    Jason, If you working with video then all of this won’t really matter unless you want to use a special lens.

  • Sorry, all the glass on the front of the lens may be used, but on a crop sensor camera, the outside portion of the image projected inside the camera will fall outside the area of the sensor, resulting in the image appearing at a focal length 1.6 times longer than the the numbers on the lens. You are correct that corner distortion and light fall off issues become virtually unnoticable, because the sensor doesn’t see the edges or corners of the projected image.

  • What will matter is that the focal length stamped on the lens will not be what you see on a crop sensor camera. Cor Canon, multiply the focal length by 1.6 to get the focal length for the image that is projected on a crop sensor. Pretty important to note for lens selection. Because most shoot video using manual focus, you can save thousands of dollars buying older used manual lenses, just do the math to calculate true focal length.

  • Wayback:

    Actually, the issue of how a lens works is a red-herring here, due to a misnomer. It would be proper to say, “does a cropped sensor cover all of the imagine circle?” rather than, “does a cropped sensor use all of the glass?” The latter is always taken to mean the former, when mentioned in the context of FF vs cropped.

    It does not matter one bit about how a lens works. What matters is how much of the resultant imaging circle (the resultant image that hits the plane of the imaging sensor) is filled with the sensor. I could use a pinhole camera and have the same issue of using more or less of the imagine circle, depending on the size of the film I use.

    The issue of the lens comes into play because it determines the size of the imaging circle. A FF lens will produce a circle that is larger than a lens meant for cropped sensors. If you use a FF lens with a FF sensor, you use the most of a large imaging circle. If you use a FF lens with a copped sensor, you use a smaller portion of the same large imaging circle, in effect throwing away the edge portions of the imaging circle. If you use a lens meant for a cropped sensor (Cannon EF-S or Nikon DX, for instance) with a FF sensor, you will get an image in the shape of a circle with black all around it because the sensor is bigger than the imaging circle.

    Therefore, the answer to the question, “does a cropped sensor use all of the glass?” (which we know in context to mean, “does the sensor cover all of the imagine circle?”) is no, a cropped sensor does not make use of “all of the glass.”

  • Sorry, I meant “imagING circle”, not “imagine circle”. Darn that auto complete.

  • Well, Jamie, in the context of the question Dave asked, and the video he presented, I think he was certainly already aware that the crop sensor doesn’t use all of the image, but I understood him to be aksing whether, literally, all of the glass was being used when an EF lens is used on a crop sensor body. And in that context the answer is yes, it is all used.

    John-Mark:

    “When working with a cropped sensor the net result is that the information from the top and bottom of the glass is not gathered.”

    I’m sorry, but that’s just not right. Information from the periphery of the SCENE is not gathered, but all the information from the center of the scene which hits the top and bottom of the glass *IS* gathered.

  • Great and informative info, still not sure what it means in the practical world though. I thought a FF lens performed better on a crop sensor camera.

    From these answers it’s clear that the corners of the lens are not used which is a good thing but since the whole crop sensor must be filled through a smaller portion of the lens glass this means that some quality (sharpness?) is lost in the process. The sum would maybe mean that we could expect similar or less sharpness from a FF lens on a crop sensor but less vignetting?

  • You would think there would be a quality loss, but it’s not so. The larger sensors just have more pixels, and I think the number per square inch remains the same. There certainly are sensor differences, but it’s not based on density. The GH2 seems to produce video comperable or superior to the 5D2, and the Panasonic has a much smaller sensor. Again, sharpness is virtually unaffected.

  • @Wayback

    six of one half dozen of another. I don’t disagree that light from the subject in the center, that travels through the edge of the lens and hit sensor uses all the glass.

    I’m speaking in lay mans terms (sorry). A crop sensor does not work like a FF sensor.

    No offense taken. we’re both looking at the same results. (I’m just not looking at it “scientifically”)

  • @ John-Mark, Jamie et al

    Thanks for the discussion and comments. It’s instructive for me to read all of the explanations. What I know about crop sensor versus a full is explained well by Philip Bloom in this Vimeo Video about 1:10-1:45

    introduction-to-dslr-cameras Here he displays what a crop sensor using a 50mm compared to a full frame while taking video. Clearly there is a difference in what is captured and displayed.

    @ Jamie – how you explained things makes complete sense to me so I think the video shows an example of the same principle.

    Thanks everyone!

Signup for Dave’s Newsletter (not working)

  • 4hourcourse1.jpg