gdanmitchell
Offline
Upload & Sell: Off
|
 Re: Diffraction on A7RV in 26MP Mode | |
tsdevine wrote:
gdanmitchell wrote:
Diffraction is not a function of pixel density.
Diffraction is an optical phenomenon that affects all lenses equally at a given aperture.
The so-called “diffraction-limited aperture” varies depending on pixel density. But that’s not a difference in the amount of diffraction — it is a difference in what the camera/lens can record.
A camera and lens with 26MP full frame sensor and the same lens on a camera with a 60MP full frame sensor generate exactly the same amount of diffraction. (Folks who don’t think so need to review how digital sampling works.)
The diffraction-limited aperture on the 26MP example is smaller than on the 60MP sensor. I don’t have the exact numbers at hand (but someone is free to figure them out for me), however using fake apertures for teh purpose of the explanation, the optimal diffraction-limited aperture on the 26MP system might be f/8 while it might be f/4 on the 60M system. (Again, I’m pretty sure those numbers are not accurate — but the concept is.)
Does that men that there is more diffraction on the 60MP system? No, not at all. If you made photographs with that lens on the two camera at the 26MP system’s diffraction-limited aperture they would show the same amount of diffraction. If you repeated this at the diffraction-limited aperture of the 60MP system, the 60MP image would continue to get sharper, while the 26MP example would not. (It isn’t quite that simple, but that’s the general idea.)
tsdevine wrote:
Diffraction is a function of pixel density....so....if you do APS-C, pixel density doesn't change. If you do just the smaller RAW file, then again....that was taken from the same density of pixels to start with, so diffraction already had its impact.
Just my understanding.
While this is good info, to me it's still semantics in the context of the OP, although I realize you were replying to someone else. In my post above, I was treating diffraction as a just a law of physics....an immutable fact (as you note, it's a function of aperture). So back to the OP's question... I don't think it matters whether he shoots it at native resolution, or APS-C mode, or a small or medium RAW. All of those options are derived from a sensor with a certain pixel density (it's a constant). (The OP was talking about using a single camera in different modes). So they will all be impacted by the inherent diffraction of the aperture any lens is shot at correct? Because the pixel density does not change in any of those scenarios, and if aperture is fixed then really it does nothing. Really the only thing you can control to minimize this...is shooting at wider apertures (and losing depth of field) or buying a camera with a lower pixel density that is less affected by what apertures you like to shoot at. But that is kind of cutting off your nose to spite your face, since you are choosing to lose resolving capability to avoid losing resolving capability. So that's why most people say just shoot your camera at the aperture you need to get the shot (even stopped down past where diffraction will be noticeable on the camera you are shooting). The only real way to "cheat" and have your cake and eat it too is to shoot at wider apertures and focus bracket/stack. But there an be challenges with doing that in all situations and it has its own quirks to deal with.
I guess there is an argument that if you shoot in APS-C on his camera, he'll need to back up in order to match the framing (although that changes perspective) and then he might be able to shoot at a wider aperture. But he's already down to 26 MP (from 60 MP) at that pont and now shooting from a further distance, which means small detail is even smaller. So that doesn't really seem like a win. So it still just makes sense to shoot it at native 60 MP and use more advanced sharpening techniques to try to minimize impact.
But I could very well be wrong on that. I don't have a degree in optics or physics, etc. I'm sure someone will point out where I went awry in the above.
Indeed, the situation is a bit different when comparing different formats such as FF and APS-C.
I’d just add that it does matter if he shoots in APS-C mode, since the unchanging among of diffraction blur at a given aperture accounts for a larger percentage of the frame width. The simple answer there is to make the same kinds of aperture decisions that you would make when using an APS-C camera — e.g. the effect of diffraction blur becomes significant sooner as you stop down on the smaller format. For example, to generalize, if you would be happy using f/16 on a full frame sensor camera, you’d be happy using f/11 on APS-C. If you feel that f/22 is pushing diffraction blur too far on FF, then you’d probably feel the same way about f/16 on APS-C.
And I don’t have a degree in optics of physics either. Mine are in… music. (Well, OK, music theory and composition with an emphasis on electroacoustic music.)
|
