Page 2 of 5 FirstFirst 1234 ... LastLast
Results 11 to 20 of 44

Thread: DLA is Nonsense! It should be deleted from the reviews.

  1. #11
    Senior Member
    Join Date
    Dec 2008
    Posts
    134

    Re: DLA is Nonsense! It should be deleted from the reviews.



    Never seen someone come ablaze overdiffractionlimitedaperture.

  2. #12
    Senior Member
    Join Date
    Dec 2008
    Posts
    327

    Re: DLA is Nonsense! It should be deleted from the reviews.



    Daniel better chime in here before this thread gets out of hand...or perhaps it is already too late.


    In any case, diffraction is an intrinsic physical phenomenon of passing a wave through a hole. It's there no matter how big or small the hole or wavelength is.


    But the ability to observe the extent of diffraction depends strongly on the resolution of the device that is used to measure it. While it is true that a real-world image is not composed of discrete, collimated point light sources, it is also irrelevant. If one acknowledges the two basic principles that (1) diffraction exists, and (2) the degree of diffraction is inversely proportional to the size of the aperture, then it doesn't matter what the object space looks like. Your ability to see the diffraction basically depends on the highest spatial frequency that your imaging device is able to capture.


    What do we mean by being able to observe the diffraction? After all, if some degree of diffraction occurs at any f-number, then there is some even at f/1.0 (assuming the lens is otherwise ideal). But what does it mean to have a sensor with such high resolution that we can "see" the effect? How do we know when we do see it? One way to think about it is to say that one is diffraction limited at the spatial frequency at which increasing the sensor resolution further confers no additional advantage, yet reducing the resolution causes you to lose detail. But that's counter-intuitive because we don't usually think of sensors as being of variable resolution.


    Thus, we usually phrase DLA in terms of loss of sharpness at a fixed sensor resolution when stopping down further; i.e., a particular sensor "starts to show" a loss of sharpness at the DLA and this softness increases as f-number increases. Opening the aperture faster than the DLA doesn't make it sharper, either (again, assuming an ideal lens).


    But a word to the OP: the reason why your post came across as "aggressive" is that your first post basically called the diffraction-limited aperture model nonsense, and dismissed it without providing any real evidence to support your claim. You didn't even introduce yourself or engage in those usual polite pleasantries that are considered essential to demonstrate good social skills.

  3. #13
    Junior Member
    Join Date
    Dec 2008
    Posts
    17

    Re: DLA is Nonsense! It should be deleted from the reviews.

    <p class="MsoNormal" style="margin: 0cm 0cm 0pt;"]<span lang="EN-US"]<span style="font-family: Calibri; font-size: small;"]Let me prove it.
    <p class="MsoNormal" style="margin: 0cm 0cm 0pt;"]<span lang="EN-US"]<o><span style="font-family: Calibri; font-size: small;"]</o>
    <p class="MsoNormal" style="margin: 0cm 0cm 0pt;"]<span lang="EN-US"]<span style="font-family: Calibri; font-size: small;"]A is a peace of image at any f-number. It may already be diffracted anyway. It forms a pixel signal as A&rsquo;.
    <p class="MsoNormal" style="margin: 0cm 0cm 0pt;"]<span lang="EN-US"]<o><span style="font-family: Calibri; font-size: small;"]</o>
    <p class="MsoNormal" style="margin: 0cm 0cm 0pt;"]<span lang="EN-US"]<span style="font-family: Calibri; font-size: small;"]When f-number increases, no matter how much, Image A will became B, and forms a pixel signal as B&rsquo;.
    <p class="MsoNormal" style="margin: 0cm 0cm 0pt;"]<span lang="EN-US"]<o><span style="font-family: Calibri; font-size: small;"]</o>
    <p class="MsoNormal" style="margin: 0cm 0cm 0pt;"]<span lang="EN-US"]<span style="font-family: Calibri; font-size: small;"]B&rsquo; is definitely differenced from A&rsquo;, and it will show in the digital image. It happens at any apertures and any pixel sizes. Pixel size doesn&rsquo;t limit its influence to be visible or not.
    <p class="MsoNormal" style="margin: 0cm 0cm 0pt;"]<span lang="EN-US"]<span style="font-family: Calibri; font-size: small;"]
    <p class="MsoNormal" style="margin: 0cm 0cm 0pt;"]<span lang="EN-US"][img]/cfs-file.ashx/__key/CommunityServer.Components.UserFiles/00.00.00.46.61/B.jpg[/img]

  4. #14

    Re: DLA is Nonsense! It should be deleted from the reviews.



    I don't know much about the subject, but read the other thread that clemmb linked because the topic was discussed at length there.


    The basic idea that I got from reading there is that yes diffraction occurs equally regardless of pixel size or recording method, but the idea of DLA is that as your pixels size gets smaller / density gets higher, the camera can resolve more detail and that detail becomes noticeably diffracted earlier. Thus the higher the pixel density the sooner the diffraction can be noticed. That is DLA. No? At f/11 big pixels won't notice the diffraction because they couldn't have resolved the detail that is being lost via diffraction anyway, but smaller pixels would have been able to and therefore the loss is noticed. This isn't to say that the diffraction is worse with greater pixel density.


    To me an analogy that makes sense is lens quality. Viewing images at 100% from a low pixel density camera can and does look sharper than the same setup (lens, sensor size, etc) with higher pixel density, because you are resolving more detail and can notice more problems. However if you take that higher pixel density image and lower the resolution or view it at the same size as the other image, it will look the same or better, not worse.


    Basically if your sensor is better, it more quickly stops being the bottleneck on image quality as diffraction sets in or when using a lower quality lens. This is true of almost anything, it's just a matter of finding the weakest link in a chain of making high quality images in this case.


    Somebody correct me if I'm wrong here.






  5. #15
    Junior Member
    Join Date
    Dec 2008
    Posts
    17

    Re: DLA is Nonsense! It should be deleted from the reviews.



    Quote Originally Posted by wickerprints


    Daniel better chime in here before this thread gets out of hand...or perhaps it is already too late.


    In any case, diffraction is an intrinsic physical phenomenon of passing a wave through a hole. It's there no matter how big or small the hole or wavelength is.


    But the ability to observe the extent of diffraction depends strongly on the resolution of the device that is used to measure it. While it is true that a real-world image is not composed of discrete, collimated point light sources, it is also irrelevant. If one acknowledges the two basic principles that (1) diffraction exists, and (2) the degree of diffraction is inversely proportional to the size of the aperture, then it doesn't matter what the object space looks like. Your ability to see the diffraction basically depends on the highest spatial frequency that your imaging device is able to capture.


    What do we mean by being able to observe the diffraction? After all, if some degree of diffraction occurs at any f-number, then there is some even at f/1.0 (assuming the lens is otherwise ideal). But what does it mean to have a sensor with such high resolution that we can "see" the effect? How do we know when we do see it? One way to think about it is to say that one is diffraction limited at the spatial frequency at which increasing the sensor resolution further confers no additional advantage, yet reducing the resolution causes you to lose detail. But that's counter-intuitive because we don't usually think of sensors as being of variable resolution.


    Thus, we usually phrase DLA in terms of loss of sharpness at a fixed sensor resolution when stopping down further; i.e., a particular sensor "starts to show" a loss of sharpness at the DLA and this softness increases as f-number increases. Opening the aperture faster than the DLA doesn't make it sharper, either (again, assuming an ideal lens).


    But a word to the OP: the reason why your post came across as "aggressive" is that your first post basically called the diffraction-limited aperture model nonsense, and dismissed it without providing any real evidence to support your claim. You didn't even introduce yourself or engage in those usual polite pleasantries that are considered essential to demonstrate good social skills.
    <div style="CLEAR: both"]</div>
    You gentlman, i respect your opinion "that phrase DLA in terms of loss of sharpness at a fixed sensor resolution when stopping down further; i.e., a particular sensor "starts to show" a loss of sharpness at the DLA and this softness increases as f-number increases. Opening the aperture faster than the DLA doesn't make it sharper, either (again, assuming an ideal lens)."


    but my opinion may be for a real lens, large aperture brings more geometry aberration. narrow down the aperture increases the sharpness in this aspact when lost sharpness by diffraction. the best IQ comes at the balance.


    And for ideal lens, the only loss of sharpness came from diffratction, then it "starts to show"at the largest aperture, as the graph shows above. And and softness increases as f-number increases.





    This is the traditional theory on film camera. I think itapplys to digtal camera too. DLAmay notbe caculated by pixel density.

  6. #16
    Junior Member
    Join Date
    Dec 2008
    Posts
    17

    Re: DLA is Nonsense! It should be deleted from the reviews.



    Quote Originally Posted by StapledPhoto


    I don't know much about the subject, but read the other thread that clemmb linked because the topic was discussed at length there.


    The basic idea that I got from reading there is that yes diffraction occurs equally regardless of pixel size or recording method, but the idea of DLA is that as your pixels size gets smaller / density gets higher, the camera can resolve more detail and that detail becomes noticeably diffracted earlier. Thus the higher the pixel density the sooner the diffraction can be noticed. That is DLA. No? At f/11 big pixels won't notice the diffraction because they couldn't have resolved the detail that is being lost via diffraction anyway, but smaller pixels would have been able to and therefore the loss is noticed. This isn't to say that the diffraction is worse with greater pixel density.


    To me an analogy that makes sense is lens quality. Viewing images at 100% from a low pixel density camera can and does look sharper than the same setup (lens, sensor size, etc) with higher pixel density, because you are resolving more detail and can notice more problems. However if you take that higher pixel density image and lower the resolution or view it at the same size as the other image, it will look the same or better, not worse.


    Basically if your sensor is better, it more quickly stops being the bottleneck on image quality as diffraction sets in or when using a lower quality lens. This is true of almost anything, it's just a matter of finding the weakest link in a chain of making high quality images in this case.


    Somebody correct me if I'm wrong here.






    <div style="CLEAR: both"]</div>
    ye, I somehow agree with you.


    but in my opinion, diffraction shows not only as on dot becomes a airy disk, or on single line became wider these kind of details, but the change of contrast in every single detail among the image. And alse, airy disk is only 84%energy aera (i don't know how to strict define it in english), the actual diffraction spread far larger.


    As diffraction influence all the details, how could the size of pixel matter.


    Actual the <span class="trans"]sensitivity to the contrast ofeach pixel would influence the visiblilty of the diffraction.

  7. #17
    Senior Member
    Join Date
    Dec 2008
    Location
    The Netherlands
    Posts
    2,304

    Re: DLA is Nonsense! It should be deleted from the reviews.



    Hmmm I don't really see your point and all this technical stuff isn't helping me understand it all without reading it a couple times[:P]


    "DLA is nonsense! It should be deleted from the reviews."


    Alright...first of have you tried the images sharpness at different apertures? So could you notice a decrease of sharpness beyond the DLA?


    I personally do see the difference and I always try to keep the aperture at the DLA at maximum, even with macro where a larger DOF is welcome. I like to get a sharper image over a softer but with greater DOF image.


    Anyhow, I don't think it should be deleted, because it proved it's use to me and I thank Bryan for putting it in it's reviews. I think I'm not the only one. And I believe 1 extra column in a list doesn't hurt anyone... perhaps 1 now[:P]


    But please continue the great debate on whether it is good or not or right or wrong, I might be reading it all more carefully if I have more time later this week[A]


    Jan

  8. #18

    Re: DLA is Nonsense! It should be deleted from the reviews.



    My experiences with the current high res cameras (50D, 5D2) is that DLA isn't something to worry about in real world use. I shoot with confidence at f16 on the 50D and f22 on the 5D2 and have no problems. If you need to stop down, go for it.

  9. #19
    Senior Member
    Join Date
    Dec 2008
    Location
    Vancouver, Washington, USA
    Posts
    1,956

    Re: DLA is Nonsense! It should be deleted from the reviews.






    Quote Originally Posted by wickerprints





    Daniel better chime in here before this thread gets out of hand...or perhaps it is already too late.








    After your excellent post, I don't think there's anything I could possibly add to the discussion. Of course, that's never stopped me before. []





    Quote Originally Posted by pin008


    DLA is Nonsense! It should be deleted from the reviews.





    I respectfully disagree. It is not nonsense, and I like having it on the reviews. It tells me what f-numbers are capable of the highest sharpness (depending on the lens, of course) as well as the point of diminishing returns (in the context of a multi-camera comparison).





    Quote Originally Posted by pin008


    there is no limit from the pixel size and density.





    Yes, there is. Every time I try to set my 24mm L II to f/32, a deep voice booms "thou shalt not stop down!" and a horde of crows materialize out of nowhere, pecking at me until I set it back to f/1.4. [] While there's nothing to physically restrain the photographer from using any particular f-number that they are capable of (and I don't mean to imply that you said there was), the sharpness of the resulting photo does have a limit, as well as a range of f-numbers over which the sharpness will be limited.





    Quote Originally Posted by pin008


    This theory assumes the image is a perfect dot and nothing else, and compares its airy disk with pixel size. If the airy disk is larger then pixel size, then the diffraction is visible.





    As wickerprints said, it also applies to continuous surfaces.





    Quote Originally Posted by pin008


    There&rsquo;s no such thing called perfectly pixel sharp images.





    I think there is. The most extreme definition of "perfectly pixel sharp" that I can think of would be 95% MTF at Nyquist with no sharpening, and that can be achieved with sensors that are designed improperly (without OLPF) at low spatial frequencies. But I'm sure Bryan had in mind a more typical (and reasonable) definition (e.g. no drop in contrast at Nyquist due to diffraction that is visibly noticeable after a small amount of sharpening) -- which is what others said in the thread.





    Quote Originally Posted by pin008


    The size of pixel doesn&rsquo;t limit anything. Digital sensor and film have no difference here.





    Say someone is shooting f/64 macro photos on their 6 MP rebel and upgrades to a 7D in order to take advantage of the smaller pixel size. They will be in for a shock when they see that the modern 18 MP images are no more detailed than their ancient camera.





    Contrast that with the example of a photographer who shoots portraits at f/5.6. When he upgrades from the 6 MP Rebel to the 7D, the linear resolution can be almost doubled.





    Between the two examples is someone who shoots at f/22. There will be an increase in resolution, but since f/22 is narrower than the DLA of the 7D, returns will be diminished.
    <div></div>

  10. #20
    Junior Member
    Join Date
    Dec 2008
    Posts
    17

    Re: DLA is Nonsense! It should be deleted from the reviews.



    According to the DLA theory which should be like graph A. pictures lose sharpness quickly after DLA, so DLA is an important critical value.


    but what i see islike graph B.





    [img]/cfs-file.ashx/__key/CommunityServer.Components.UserFiles/00.00.00.46.61/C.gif[/img]

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •