add Sigma camera to diffraction calculator

[nidoba]

Due to the pixel level sharpness of the newer Sigma cameras, perhaps adding one of the cameras, with the 1.5 crop sensor, would be helpful to some. Including an additional analysis of other cameras without AA filters could also prove informative.

Cameras without an anti-aliasing filter could/should be different than the ones currently used in the calculator that have an AA filter, I think?

Thanks, Mike

[xpatUSA]

Due to the pixel level sharpness of the newer Sigma cameras, perhaps adding one of the cameras, with the 1.5 crop sensor, would be helpful to some.

Thanks, Mike

Well, Mike, good question. However, it seems that Sigma is unsupported by CiC calculators in that not even 1.7 crop appears for the older cameras especially those with big fat pixels (9.12um).

As a devout Sigma user, I call it "The Gap" which often appears in 'list of supported cameras' for something where the list goes:

.
.
Samsung
Sony
.
.

[xpatUSA]

Cameras without an anti-aliasing filter could/should be different than the ones currently used in the calculator that have an AA filter, I think?

Thanks, Mike

Perhaps a bit more reading about diffraction - which is basically a lens aperture setting phenomenon - would help you to understand that diffraction is independent of the sensor pixel pitch or it's sharpness. In other words, diffraction effects appear in the image plane only. For example, even substituting toilet paper for the sensor would not affect the amount of diffraction at all.

[DanK]

Perhaps a bit more reading about diffraction - which is basically a lens aperture setting phenomenon - would help you to understand that diffraction is independent of the sensor pixel pitch or it's sharpness. In other words, diffraction effects appear in the image plane only. For example, even substituting toilet paper for the sensor would not affect the amount of diffraction at all.

Yes, but isn't it the case that the point at which resolution is limited by diffraction IS determined by characteristics of the sensor, such as pixel size, the use of an AA filter, etc.? My understanding of this is limited to the tutorial on this site, but that is how it is explained there.

[xpatUSA]

Yes, but isn't it the case that the point at which resolution is limited by diffraction IS determined by characteristics of the sensor, such as pixel size, the use of an AA filter, etc.? My understanding of this is limited to the tutorial on this site, but that is how it is explained there.

Good morning, Dan,

It all depends what is meant by "resolution".

What you said is true if we are talking about the whole camera resolution of the final image.

However, I was just talking about lens resolution in the image plane, like what is is published in lens data. Let's say that a lens has an center MTF of 50% at 30 lp/mm at a setting of f/16. That lens resolution does not change no matter what camera it is mounted on.

Now, talking only about sensor resolution, that same 30 lp/mm will result in a sensor MTF which IS dependent on pixel pitch and, to a lesser extent, the degree of AA filtering which could be zero.

When a diffraction calculator also has sensor parameters as inputs, the two types of resolution above are compared to determine the 'winner', if you will. A sensor with small pixels will be able to resolve lens diffraction caused by the aperture setting earlier than a sensor with large pixels. Thus it can be said that diffraction "sets in" at a lower f-number on a 16MP micro four-thirds than it does on a 12MP full-frame. Indeed, on the 3.4MP Sigma SD9 with it's 9.12um pixels, I can use up about f/16 without considering diffraction. However, the m4/3 Panasonic 12MP GH1 with it's 4.33um pixel pitch can be noticeably blurry at f/16.

The point is that the lens and the sensor performances are separate issues and the one can not affect the other. Unfortunately, we can not use the one without the other and must, of necessity, consider them as a combination which can lead to instances of PO (photographic obfuscation)

[george013]

Yes, but isn't it the case that the point at which resolution is limited by diffraction IS determined by characteristics of the sensor, such as pixel size, the use of an AA filter, etc.? My understanding of this is limited to the tutorial on this site, but that is how it is explained there.

1. Light is diffracted when passing an object.
2. That diffracted light causes that the top of the light-cone diafragm-diameter image-distance is not sharp. It wil never be a point but a circle. It's called the airy disk.
3. The maximum sharpness a sensor can gain is limited either by the sensor-size or by the airy disk. If it's the airy disk, than it's called diffraction limited.

George

[nidoba]

Perhaps a bit more reading about diffraction - which is basically a lens aperture setting phenomenon - would help you to understand that diffraction is independent of the sensor pixel pitch or it's sharpness. In other words, diffraction effects appear in the image plane only. For example, even substituting toilet paper for the sensor would not affect the amount of diffraction at all.

True, but isn't it also true that the point at which diffraction becomes detrimental to the image can be effected by other factors (lens quality, AA filter strength).

[xpatUSA]

True, but isn't it also true that the point at which diffraction becomes detrimental to the image can be [affected] by other factors (lens quality, AA filter strength).

Indeed that is true, because 'the image' itself (assuming that to mean what's on the screen or in the print) results from the combination of everything between the subject and the raw file data or even the raw file converter/processor and the print driver and the printer quality and the paper used.

Pardon the facetiousness, but I prefer myself to separate things as far as possible in order to understand them better

[george013]

True, but isn't it also true that the point at which diffraction becomes detrimental to the image can be effected by other factors (lens quality, AA filter strength).

In addition.
Don't mix unsharpness due to diffraction and unsharpness due to other matters.
Diffraction is a physical phenomen. You can calculate it. Lens quality you can't calculate. You can show it by doing some experiments.

George

[xpatUSA]

True, but isn't it also true that the point at which diffraction becomes detrimental to the image can be affected by other factors (lens quality, AA filter strength)?

As a separate discussion I'm hoping that, by saying "the point at which diffraction becomes detrimental to the image", we realize that there is no single such point but more like a range of f-numbers over which diffraction becomes increasingly obtrusive, rather than statements like "diffraction sets in at f/8" as if it occurs suddenly at f/8 while we are stopping down . . .

[george013]

As a separate discussion I'm hoping that, by saying "the point at which diffraction becomes detrimental to the image", we realize that there is no single such point but more like a range of f-numbers over which diffraction becomes increasingly obtrusive, rather than statements like "diffraction sets in at f/8" as if it occurs suddenly at f/8 while we are stopping down . . .

It's physics. There is a point. Simply when the airy disk get bigger as the pixel-size. Maybe there are some other definitions, but in essence it's this.
It is also not related to the f-number. The f-number is the ratio between focal-distance and diafragm-diameter while the airy disk is calculated from the ratio image-distance and diafragm-diameter. If you want to use the calculator you should use the, I hate that word, "effective f-number". Then you are using the image-distance.

Focused on infinity with a certain f-number will give you another diffraction-limitation as focused near with the same f-number.

I've no experience with it. Pure theoretical.

George

[xpatUSA]

It's physics. There is a point. Simply when the airy disk gets bigger than the pixel-size [pixel pitch?]. Maybe there are some other definitions, but in essence it's this.

Looks like we're getting technical, George

By "bigger" do you refer to the Airy disk radius or the diameter? By either, do you refer to the first dark ring?

And now for the trap : What is the exact dimension of such a disk but consisting of white light (not monochromatic green at exactly 555nm) coming through a pin-hole of say 1mm diameter and landing on a piece of white paper at say 100mm axial distance?

It is also not related to the f-number. The f-number is the ratio between focal-distance and diafragm-diameter while the airy disk is calculated from the ratio image-distance and diaphragm-diameter. If you want to use the calculator you should use the, I hate that word, "effective f-number". Then you are using the image-distance.

I have my own spreadsheet for such purposes - it accounts for the above variables.

Focused on infinity with a certain f-number will give you [a different] diffraction-limit as [compared to close focusing] with the same f-number.

We know that. My spreadsheet copes by increasing the diffraction at the image plane with increasing magnification.

I've no experience with it. Pure theoretical.

George

[george013]

Looks like we're getting technical, George

By "bigger" do you refer to the Airy disk radius or the diameter? By either, do you refer to the first dark ring?

The diameter, or the radius, or the surface, or the whatever what. As long you use the same unit also for the pixel.
What's so important if the size refers to the center or the first ring or the second. It's totally unimportant for understanding of diffraction limitation.

And now for the trap : What is the exact dimension of such a disk but consisting of white light (not monochromatic green at exactly 555nm) coming through a pin-hole of say 1mm diameter and landing on a piece of white paper at say 100mm axial distance?

Well, if you're curious, Wiki seems to be a good friend. http://en.wikipedia.org/wiki/Airy_disk. You may use any wavelenght you want.

We know that. My spreadsheet copes by increasing the diffraction at the image plane with increasing magnification.

Who else as you? And how does your spreadsheet copes with increasing magnification?
Than you could tell us the difference using a lens with a given f-nummber at infinity and at a close distance. That would be important for macro.

Can we agree that diffraction limitation is 1) not dependent on the diafragm-diameter and 2) not dependent on the f-number?

George

[xpatUSA]

The diameter, or the radius, or the surface, or the whatever what. As long you use the same unit also for the pixel.
What's so important if the size refers to the center or the first ring or the second. It's totally unimportant for understanding of diffraction limitation.

Hogwash.

Well, if you're curious, Wiki seems to be a good friend. http://en.wikipedia.org/wiki/Airy_disk. You may use any wavelenght you want.

Nice side-step. At least you avoided the trap.

Who else as you? And how does your spreadsheet copes with increasing magnification?
Than you could tell us the difference using a lens with a given f-nummber at infinity and at a close distance. That would be important for macro.

Sorry, I'm not playing.

Can we agree that diffraction limitation is 1) not dependent on the diafragm-diameter and 2) not dependent on the f-number?

No. We are doomed to never agree.

[george013]

You forgot to answer the question how you cope with the magnification in your spreadsheet.
George