p.3 #1 · Sony A7R IV Photographic Dynamic Range (PDR)
cgarcia wrote:
No... because there is more than lens aberrations.
Yes, usually the lenses are aberration limited before becoming diffraction limited, but even a perfect lens with zero aberrations will be diffraction limited.
So the sensor doesn't neccesarily needs to outresolve the lens to get rid of aliasing: at worst, at a given aperture, it only needs to have pixels small enough to cause the lens to become diffraction limited if it was not already limited by its aberrations (as likely was if not somewhat stopped down).
Totally removing the aliasing by the means of diffraction requires really small pixels (see recent Jim's article). But diffraction is increasingly filtering out the aliasing each time the pixels get smaller. And it is not merely getting scaled down. That's what I pointed out a few posts above. A sensor with twice the pixels gets the same aliasing "safety" a full f-stop before. If we felt safe shooting a fabric at F11 in the A7R3, the A7R4 will get us just the same feeling at F9.5 (not just the same artifacts smaller, but less artifacts compared to the A7R3 also at F9.5).
I was including ALL lens conditions in what I stated; diffraction included.
It is immaterial if a lens is diffraction limited at f11 or f8 if you are shooting at f5.6, for example. At that point, if a lens exceeds the sensor, there is the potential for moire if the subject is right.
Note that many people are paying close attention to the diffraction point on these lens systems to avoid this explicitly to maximize the performance of the system as a whole, not just the sensor (by buying a newer, higher density sensor), but by shooting at an aperture that improves the chance to pass the optimal lens performance to the sensor as well.
The (in my opinion) much larger question is exactly how poorly are the corners of many of these lenses compared to the centers, especially when considering that you cannot stop down nearly as far as previously while still maintaining performance that isn't diffraction-limited for the sensor? The most modern Sony zooms like the 24-105 and others appear to have remarkable performance close to wide open to support this, but many (most?) do not.
If you can't stop down past f5.6 without being diffraction-limited, I think many will find (if they analyze this carefully) that many of the older Sony lenses won't be so good with the 62MP camera. To maintain even performance across the frame (when that is desired, as it often is for landscape work) you may find you have to go into diffraction-limited range a bit to keep things looking even throughout with many of the older E-mount lens designs.
I can't believe I just used the term "older E-Mount lens designs"... It's been about 7 years for the Sony FF E-mount, but it's a lifetime ago.
p.3 #3 · Sony A7R IV Photographic Dynamic Range (PDR)
mjm6 wrote:
I was including ALL lens conditions in what I stated; diffraction included.
It is immaterial if a lens is diffraction limited at f11 or f8 if you are shooting at f5.6, for example. At that point, if a lens exceeds the sensor, there is the potential for moire if the subject is right.
Note that many people are paying close attention to the diffraction point on these lens systems to avoid this explicitly to maximize the performance of the system as a whole, not just the sensor (by buying a newer, higher density sensor), but by shooting at an aperture that improves the chance to pass the optimal lens performance to the sensor as well.
The (in my opinion) much larger question is exactly how poorly are the corners of many of these lenses compared to the centers, especially when considering that you cannot stop down nearly as far as previously while still maintaining performance that isn't diffraction-limited for the sensor? The most modern Sony zooms like the 24-105 and others appear to have remarkable performance close to wide open to support this, but many (most?) do not.
If you can't stop down past f5.6 without being diffraction-limited, I think many will find (if they analyze this carefully) that many of the older Sony lenses won't be so good with the 62MP camera. To maintain even performance across the frame (when that is desired, as it often is for landscape work) you may find you have to go into diffraction-limited range a bit to keep things looking even throughout with many of the older E-mount lens designs.
I can't believe I just used the term "older E-Mount lens designs"... It's been about 7 years for the Sony FF E-mount, but it's a lifetime ago.
It seems that you are more worried about lens performance (which is legit) and myself about aliasing, which was the initial talk. Maybe I see things from my personal use case: I mostly shoot architecture and landscape and use relatively small apertures (and very few portraits don't caring a lot about resolution). The new high res sensors start to enter in an aliasing free territory at my mostly used apertures (the A7R4 is not yet there, but is a solid step forward). Yes, existing lenses may now throw away the added resolution (err... I mean: they can't go beyond the nature limits) specially where they already were "throwing" it by themselves (the corners?). But personally I don't care because I need the DOF and already had enough resolution and acceptable corners. I just want to get rid of the aliasing.
In the end, the aliasing debate is all about education. Every engineer knows that a proper digitization requires oversampling and using an AA-filter (for a very good reason the A9 has it). Ideally our cameras should have twice the pixels than wanted and a suitable AA filter, to get lowly-aliased images with the originally intended resolution (just as audio CD does). But marketing departaments are listening to people crying for AA-less cameras, so we'll have to wait for gigapixel sensors free of aliasing with wide open lenses (also to shoot landscapes at barely 8K of effective resolution; either using F8 in full frame or F4 in four thirds... this is gear-independent: it is the physics limit for a single exposure with certain DOF). Most people don't care about the aliasing (the false resolution/invented sharpening becomes visible only when happens to generate patterns) but some of us are crying for more pixels, just because the patterns are sometimes visible and can't be removed by any algorithm.
At the end, the anti-AA-filter, we-want-even-more-resolution troop will get their new pixels for just the same resolution with many lenses/many photographic scenarios; and we, the pro-AA-filter troop will use them to defeat aliasing, so we'll win ... at the cost of storage space.
... at the cost of storage space.