gdanmitchell wrote:
Seriously folks, what possible value is coming from continuing this back and forth? There's nothing new to add and no minds are going to be changed.
I agree. But its too bad as it really is an interesting topic to me if we focused on the physics of it and why lenses have diffraction at different apertures. Once you think of light as a wave form, it gets easier to understand or even visualize with water flow. But that's no fun when we could be calling each other names like a bunch of little boys.
SGinNorcal wrote:
I agree. But its too bad as it really is an interesting topic to me if we focused on the physics of it and why lenses have diffraction at different apertures. Once you think of light as a wave form, it gets easier to understand or even visualize with water flow. But that's no fun when we could be calling each other names like a bunch of little boys.
Honestly the average physics degree holder doesn't seem to know much about optics. Ive talked to physicists, I've dealt with physicists, but mostly I'm past believing everything physicists say
AmbientMike wrote:
Honestly the average physics degree holder doesn't seem to know much about optics. Ive talked to physicists, I've dealt with physicists, but mostly I'm past believing everything physicists say
I suspect the average physics degree holder knows more about optics than the average photographer but I have no interest in arguing with you about who is dumber.
SGinNorcal wrote:
I agree. But its too bad as it really is an interesting topic to me if we focused on the physics of it and why lenses have diffraction at different apertures. Once you think of light as a wave form, it gets easier to understand or even visualize with water flow. But that's no fun when we could be calling each other names like a bunch of little boys.
Ok, I'll stop responding to the negativity. I would be very interested to learn more about diffraction and why some lenses are more affected than others.
Maybe a new thread? As this one was nonsense to start with.
Gerry
SGinNorcal wrote:
I suspect the average physics degree holder knows more about optics than the average photographer but I have no interest in arguing with you about who is dumber.
If anyone seriously thinks I dont know anything about optics, that's pretty stupid, but feel free to read the 1st several paragraphs and explain how it relates to sharpness on camera lenses, at all. Certainly not intuitive. Might be a good place to begin, or might be too confusing, i certainly don't claim to be able to explain this to you
I'm probably going to regret this, but at the link supplied in the previous post, particularly relevant information is found in the section labeled "Diffraction-limited imaging."
Anyone can read that whole section — it isn't too long — but here's a crucial fact:
"Thus, the larger the aperture of the lens compared to the wavelength, the finer the resolution of an imaging system. This is one reason astronomical telescopes require large objectives, and why microscope objectives require a large numerical aperture (large aperture diameter compared to working distance) in order to obtain the highest possible resolution." [emphasis added]
To anticipate misinterpretation, this does not mean that one must always choose the very largest aperture, nor that the largest aperture will always be the sharpest one on a given lens for reasons relating to other aspects of lens design and construction. In addition, digital sensors are limited in their ability to record image detail by the nature of digital sampling — in short, higher photo site density, all else being equal, will let the sensor take advantage of higher resolution lenses and apertures.
Also to anticipate misinterpretation, this does not mean that there are never reasons to choose a smaller aperture that will produce an image that is affected more by diffraction blurring. In this thread, I and others have been crystal clear about this. If the only concern is maximum resolution in the center of the frame, then using an excellent lens at its diffraction-limited aperture could make sense. But other photographic concerns, such as a desire for thinner DOF or thicker DOF (and various exposure-related issues) may lead a photographer to choose other apertures than the diffraction-limited aperture at the cost of some loss of sharpness.
Adding to what Dan shared, this was also in that very same link:
>>as light travels through slits and boundaries, secondary point light sources are created near or along these obstacles<<
And everybody can understand that "secondary point light sources" equate to multi-point focus error at visible at the focus plane, which in turn equates to --wait for it-- a loss of resolution at the focus plane; or "fuzz" if you prefer. Oh, and for our purposes as photographers, the "focus plane" happens to be the camera's sensor...
And to be clear f8 on an astronomical telescope is a much larger physical opening than f8 on a camera lens. This is the reason larger formats that have larger physical lens openings will have a higher number f stop at which diffraction starts reducing sharpness. I think Ansel Adams was part of the 'f64 club', though I think the name was rather loosely used. But the point is that f64 might be a sensible shooting aperture on an 8x10 view camera and not suffer from diffraction much, whereas f64 would produce a very soft image if it were available on APS-c
gdanmitchell wrote:
I'm probably going to regret this, but at the link supplied in the previous post, particularly relevant information is found in the section labeled "Diffraction-limited imaging."
Anyone can read that whole section — it isn't too long — but here's a crucial fact:
"Thus, the larger the aperture of the lens compared to the wavelength, the finer the resolution of an imaging system. This is one reason astronomical telescopes require large objectives, and why microscope objectives require a large numerical aperture (large aperture diameter compared to working distance) in order to obtain the highest possible resolution." [emphasis added]
To anticipate misinterpretation, this does not mean that one must always choose the very largest aperture, nor that the largest aperture will always be the sharpest one on a given lens for reasons relating to other aspects of lens design and construction. In addition, digital sensors are limited in their ability to record image detail by the nature of digital sampling — in short, higher photo site density, all else being equal, will let the sensor take advantage of higher resolution lenses and apertures.
Also to anticipate misinterpretation, this does not mean that there are never reasons to choose a smaller aperture that will produce an image that is affected more by diffraction blurring. In this thread, I and others have been crystal clear about this. If the only concern is maximum resolution in the center of the frame, then using an excellent lens at its diffraction-limited aperture could make sense. But other photographic concerns, such as a desire for thinner DOF or thicker DOF (and various exposure-related issues) may lead a photographer to choose other apertures than the diffraction-limited aperture.
You directly quoted diffraction page. Then spent the next paragraph completely contradicting that quote saying aperture doesn't matter
*sigh*
And nobody has been able to link the intro and following paragraphs to anything remotely resembling linking diffraction to lens resolution . You had to scroll down the page to find something mentioning imaging then you did a 180 completely contradicting it in the next paragraph. Which of course, unbelievably, almost certainly doesn't keep you from thinking you're educating me
gdanmitchell wrote:
I'm probably going to regret this, but at the link supplied in the previous post, particularly relevant information is found in the section labeled "Diffraction-limited imaging."
Anyone can read that whole section — it isn't too long — but here's a crucial fact:
"Thus, the larger the aperture of the lens compared to the wavelength, the finer the resolution of an imaging system. This is one reason astronomical telescopes require large objectives, and why microscope objectives require a large numerical aperture (large aperture diameter compared to working distance) in order to obtain the highest possible resolution." [emphasis added]
To anticipate misinterpretation, this does not mean that one must always choose the very largest aperture, nor that the largest aperture will always be the sharpest one on a given lens for reasons relating to other aspects of lens design and construction. In addition, digital sensors are limited in their ability to record image detail by the nature of digital sampling — in short, higher photo site density, all else being equal, will let the sensor take advantage of higher resolution lenses and apertures.
Also to anticipate misinterpretation, this does not mean that there are never reasons to choose a smaller aperture that will produce an image that is affected more by diffraction blurring. In this thread, I and others have been crystal clear about this. If the only concern is maximum resolution in the center of the frame, then using an excellent lens at its diffraction-limited aperture could make sense. But other photographic concerns, such as a desire for thinner DOF or thicker DOF (and various exposure-related issues) may lead a photographer to choose other apertures than the diffraction-limited aperture.
AmbientMike wrote:
If anyone seriously thinks I dont know anything about optics, that's pretty stupid, but feel free to read the 1st several paragraphs and explain how it relates to sharpness on camera lenses, at all. Certainly not intuitive. Might be a good place to begin, or might be too confusing, i certainly don't claim to be able to explain this to you
Its seems you don't want to play nice. Anyway, that's a general description, not necessarily for cameras. But waveforms are waveforms and they don't bend around corners and maintain their shape. Diffraction in light is similar to turbulence in air or vortices in water and the disruption of light is why the image is softened after passing through an aperture. Two evening ago I was at the beach at sunset. There was a rock in the path of the sun slicing it in half with the sun appearing to burn into the rock/bend around it. With the sun as light, the rock as the aperture, and my eyeball as a sensor, a very practical display of diffraction. I would be curious about aperture blade design and what has been done to reduce diffraction in state of the art designs. It seems that more blades/more circular aperture perform better. I wonder of edge conditioning or thickness of the aperture have much impact. People like sunstars and that seems to come from lower blade counts. I wonder if a "good" sunstar lens also has a low diffraction limit.
AmbientMike wrote:
You have not figured out the question (although you are close.)
Yet you insist on lecturing me about the solution.
Enough of the enigmatic insult.
Look at your beloved Adaptall lens tests. Look at any other group of well regarded lens tests. Such as Lens Tip, or Optical Limits.
You will see that in almost every case, excepting a few optical exotica like your quoted 58 1.2, overall resolution reaches a peak and then starts to decline. On aps-c sensors it's about f/8, on full frame about f/11.
What in your view is causing that?
SGinNorcal wrote:
Its seems you don't want to play nice. Anyway, that's a general description, not necessarily for cameras. But waveforms are waveforms and they don't bend around corners and maintain their shape. Diffraction in light is similar to turbulence in air or vortices in water and the disruption of light is why the image is softened after passing through an aperture. Two evening ago I was at the beach at sunset. There was a rock in the path of the sun slicing it in half with the sun appearing to burn into the rock/bend around it. With the sun as light, the rock as the aperture, and my eyeball as a sensor, a very practical display of diffraction. I would be curious about aperture blade design and what has been done to reduce diffraction in state of the art designs. It seems that more blades/more circular aperture perform better. I wonder of edge conditioning or thickness of the aperture have much impact. People like sunstars and that seems to come from lower blade counts. I wonder if a "good" sunstar lens also has a low diffraction limit. ...Show more →
I don't think there is much that can be done in terms of blade design that can change the diffraction limit. Thinking about the physics, I'm guessing a purely circular aperture of exactly the same area as a hexagonal opening might do slightly better as a circular shape will maximise the distance between blades. Though if you overlay a hexagon over and equivalent area circle some parts of the hexagon will be inside the circle (worse for diffraction) and some parts will be outside the circle (better for diffraction). Thinner blades might also do slightly better, but blades these days are likely about as thin as can be. As diffraction has been known about for centuries, if there were a way around it through design I am guessing it would have been discoverred. It is not a particularly complex topic in comparison to other elements of lens design.
Well there are curved blades, more blades, less blades. Then that becomes about bokeh shape along with diffraction. It seems there are few "free lunches" in lens design, one improvement might yield another problem.
SGinNorcal wrote:
I would be curious about aperture blade design and what has been done to reduce diffraction in state of the art designs. It seems that more blades/more circular aperture perform better. I wonder of edge conditioning or thickness of the aperture have much impact. People like sunstars and that seems to come from lower blade counts. I wonder if a "good" sunstar lens also has a low diffraction limit.
The shape of the aperture blades does not change the diffraction limit (i.e. the ability of the system to resolve detail); but circular apertures (compared to hexagonal apertures) do minimize the appearance of diffraction-related artifacts.
But to the previous commenter's point, circular blades -- and more blades in general -- are usually more expensive to implement, so the lens designer is choosing them reasons of aesthetic effect wide open (or close to it); that the blades might also appear to improve sharpness near the diffraction-limited f-number is an ancillary benefit.
As there is wide agreement that round irises are a very good thing, why would you not want to closely emulate the undisturbed light path of the lens when wide open, at smaller apertures? Most know that there are solid benefits to using aperture settings a little off wide open, easier focusing, more lens contrast, greater detail retention in bokeh., less artefacting, etc.
14-15-16 seems to be the sweet spot range in iris blade counts, though Zeiss added an 18-blader to its longer Supreme Primes. As the manufacture and fab of iris mechs have evolved to the point that high blade units are fitted to very affordable lenses, this is very close to a free lunch, IMO. The greatest benefit is to bokeh quality, where it can be profound.
Many others are finally moving in the right direction too, many new entrants from China (e.g. DZOfilm, Sirui, and I see even today a crowd called Brightin Star with a 50/1 with a 15-blade unit - $269). So it is no longer esoteric nor is it prohibitive on the basis of cost, however the industry moves with the alacrity of a walker traversing a field of deep mud.
philip_pj wrote:
14-15-16 seems to be the sweet spot range in iris blade counts, though Zeiss added an 18-blader to its longer Supreme Primes. As the manufacture and fab of iris mechs have evolved to the point that high blade units are fitted to very affordable lenses, this is very close to a free lunch, IMO. The greatest benefit is to bokeh quality, where it can be profound.
.
I think the downside to high blade count might be mechanical complexity and the motors to drive that complexity since the iris is no longer driven direct by the aperture ring/manual input. But I get the positive aspects of a more perfectly round iris.
I think that the lenses with extremely large numbers of blades and designs that round the aperture face a basic problem. Yes, they can improve the quality of the OOF components of an image, especially in photographs at very large apertures and with lots of OOF background, but…
- image quality is already quite good with simpler designs, so the significance of the difference is less than we might imagine.
- the difference is even less meaningful for other sorts of photographs that don’t rely on huge apertures.
- given costs and complexity and the likelihood that these designs appear in relatively large lenses, the market for them will remain small and specialized.
That doesn’t change the facts about what they might accomplish, but it does put it in perspective.
[Edited to fix some really dumb typos… and then, a few other things since I was here.]
