F8: last stop for gas

[wparsonsgisnet]

Advertisement (gone after registration)

I have noticed, working with the M8, even when only using the viewing screen on the back (chimping for low resolutoin?), that after closing down to F8, any further reduction in aperture results in less resolution.

 

I have seen this -- almost in passing. That is, during photo shoots or when just wandering around shooting for fun, it is so apparent with the M8 that it is visible without having to take extraordinary steps on the computer screen to see it.

 

I am commenting because I just finished reading Sean's review of the CV 40 f1.4 lens, and I note it in his focus-shift tests. After f8, resolution is reduced at both f11 and f16. So, it's the sensor and not the lenses.

 

I take it that the 1.33 factor comes into account here, causing the circle of confusion to exceed the ability of the sensor to resolve details.

 

Anyone else noticed this?

 

And any of the optical monsters care to explain?

[Morten Grathe]

I can't explane it, but a professor in physics told me, that it is because of difraction. On a full frame-sensor this usually kicks in after f11. The problem starts earlier, the smaller the sensor. In theory the small digicams have difraction even wide open.

Maybe someone in this forum can explain it better than this.

[punktum]

I can't explane it, but a professor in physics told me, that it is because of difraction. On a full frame-sensor this usually kicks in after f11. The problem starts earlier, the smaller the sensor. In theory the small digicams have difraction even wide open.

Maybe someone in this forum can explain it better than this.

 

 

I´m not a physics, but what was told to me when I studdied photography, is thas as smaler the aperture is, the more likely you will get diffraction. I think, as far as I can remember this technical detail stuff it is related to the real diameter of the aperture.

[stunsworth]

It's difraction begining to kick in. The reviews I've seen of Leica lenses have said that they are sharpest at f4 to f5.6.

[jaapv]

You'll find an in-depth article on Sean McHugh's site (Cambridge in Colour) here:

Diffraction Limited Photography: Pixel Size, Aperture and Airy Disks

[jaapv]

Another thing: as the Airy disk on the M8 cannot be much larger than the pixel size on the M8, due to the lack of an AA filter, diffraction effects are visible much sooner thatn on AA filtered camera's (all others ) There the Airy disk size may be up to two pixels in size without losing resolution. This, imo, explains why the M8 at 10 Mp can compete, resolutionwise, with camera's of far higher pixel count. An argument could be made that the resolution of the M8 is similar to a 17Mp AA filtered full-size 135 sensor because of this theoretical consideration.

[rosuna]

Advertisement (gone after registration)

It is the format Bill, the crop factor.

Smaller formats suffer more the effects of diffraction. Well, the size of the Airy disk can be the same, but small formats need much larger enlargements for a given print.

[jaapv]

Which is the same,Ruben, as saying with the same pixel number, the pixel size is larger, so the Airy disk can be larger, so the higher diffraction is not recorded by the sensor.

Edit, come to think of it, it is not quite the same thing; you are saying it is less obtrusive because of smaller magnification (true), I am saying, up to a certain size, it will not be recorded because it is below the diffraction limit of the sensor.

[Photoskeptic]

Using Sean McHugh's formula the M8 should be okay at f11.

[sean_reid]

It's diffraction, as mentioned above (and in the CV 40 article). Most lenses will show higher resolution at F/8 than at F/11 or F/16, even on film or on a camera with a FF sensor. This article may be of interest:

 

Luminous landscape

 

As a rule of thumb, in my own photography, I *do* think of F/8 as being "the last stop for gas" unless I'm willing to give up a bit of res. for DOF (which is sometimes the case).

 

None of this is new, really, we're just noticing more because the M8 makes it easier for us to put our pictures "under a microscope".

 

With respect to the "formula" post, that may be true theoretically but I can say that in practice (with the M8) one normally sees a drop in res. from F/8 to F/11.

 

Then again, photography isn't about resolution any way <G>.

 

Cheers,

 

Sean

[jaapv]

Using Sean McHugh's formula the M8 should be okay at f11.

 

The formula is for AA filtered sensors. In actual practice on the M8 diffraction effects start to be visible (albeit at the insane 100% crop) between 5.6 and 8.

[wparsonsgisnet]

I had an M4 and an M6 classic in my closet. they've been sitting there since the D2 arrived in Feb 04. I used the film bodies 3 times during that period when I needed iso 1600. It was after one of those shoots that I noticed all the fast slide film had disappeared.

 

Recently, my older son and asked for and got custory of the M6 and the DR lens (I can't say how sorry I am that this lens doesn't fit the M8!). He is doing a lot of film photog and is making superb use of this otherwise idle camera.

 

One of the things he has been doing is pinhole photog, including building one, and he's having a ball doing that, too.

 

So, I dug in the pile of toys and hauled out the pinhole body cap from Finney Field Cameras. Mounted in an M-body cap, it has a hole with a .009 inch diameter. It is also rated at f 128 @ 30mm (that works out to .009 inches).

 

I tried it on the M8 -- blahhh. Everything is just blurry. I don't know why I expected to see anything, when I already knew that f8 is the limit. So, the kid gets another toy.

 

Here is an image -- *** with the M8 mounted on the Leica table-top tripod and (long) ball head *** -- and Just For Fun, I appied Neat Image to it.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[Jamie Roberts]

Hey Bill--

 

I think what Sean was trying to tell you is that all (to my knowledge, anyway) 35mm form factor glass begins to lose resolution--film or digital--due to lens diffraction after f8 (and some lens designs after 5.6).

 

That you didn't notice this as much with film probably speaks to your subject matter and to the fact you probably didn't look so carefully in Photoshop

 

This is one of the reasons the serious landscape guys (and studio guys who need a *lot* of DOF) use medium format, not 35mm

 

That the sensor plays a role *as well* is a complicating factor; I think over at LL one of the experts states that shooting the 1ds2 (16mp) @ f16 gives the equivalent resolution of shooting with a 2mp camera!

 

Having said all that, your M8 50 Lux @ f16 is going to be way sharper than a pinhole camera--those shots are actually not bad for something done without glass; the hole is maybe too big for sharpness with the M8.

[wparsonsgisnet]

Hi, Jamie. I get it now that I wasn't looking hard enuf at my film images.

 

I very seldom shot at f16. You gotta be outdoors with bright sun to do this. Also, I *really* like slow film. In my dreams, Panatomix-X is back! I didn't often have Tri-X loaded for outdoors.

 

And, I realize that I never liked my outdoor shots at f16. Theu just looked crummy. The grain magnifier worked ok under the enlarger, the image just sucked.

 

So, no gas anywhere, right?

 

Thanks for hitting me upside the head.

 

 

BTW, I find the M8 has so much apparent resolution that I'm in the market for a wide lens -- because I wan't to do landscape photog for the first time with 35mm. This is quite apparent in some of the panoramas from the Yosemite thread. And, I know I'll stay at f8 or wider. With my choice of either the CV15 or the WATE (if my ship comes in, maybe both), I won't need f8 anyway. I was looking at pix of the CV15: the DOF scale showed 5 and then infinity! And, the distance between them was about 4mm if the pic scale could be determined.

[lars_bergquist]

Bill and everybody,

diffraction occurs in the lens and has nothing to do with the sensor. It is a phenomenon which arises when light waves pass close to an opaque edge, in our case normally the blades of the diaphragm, and are disturbed in their rightful paths by it. How much diffraction you get depends ultimately on what proportion of the light wave-fronts that actually passes close to the blades. This proportion increases as we stop down, i.e. make that round hole smaller *in proportion to the focal length*.

 

If no other optical errors existed, a lens would consequently get sharper as we open it up, and an f:1 lens would be sharper than an f:2 lens. But various aberrations do exist, and most of these a) degrade definition more than diffraction does in a wide opens lens, and they decrease as we close the diaphragm – while at the same time, diffraction increases! In a really well-corrected lens, these standard aberrations (chromatic, spherical, coma, astigmatism) become negligible very soon. Then, the rising curve of diffraction overtakes the sinking curve of aberrations. At that point, the lens is 'diffraction limited' and has its maximum definition – in the plane of best focus! If however a lens is diffraction limited at f:4 and we stop down to f:5.6 or 8, the image as a whole may look sharper because of the increase of the depth of field, even though sharpness does in fact decrease a wee bit in the exact plane of best focus.

 

So, the point where diffraction gets important depends, first on how well corrected against aberrations it is (i.e. how soon it becomes diffraction limited), second on how much you want to enlarge it – and third, the subject! Sean is right however that it is an acceptable rule of thumb that you should not go further than f:8 if extreme resolution is an issue. In practice, f:11 normally gives good results. This BTW holds good for all lenses irrespective of absolute focal length, and absolute format.

 

Some early, very badly corrected lenses had so much aberration that they continued to improve down to f:32 or even 64 (of Adams & Weston fame). On the other hand, the fact that a pinhole optic never gets really sharp is because just about all the light has to pass extremely close to the pinhole's edge.

 

Also, when I worked in a wet darkroom, I sometimes used settings of f:16 and 22 in order to take some of the edge off the grain in large enlargements. The exposure times grew so long however, that I had to switch off the darkroom lamp.

 

If I have made a scientific error in the above, please do correct me – civilly, please.

 

The old man from the Age of F:6.3 Optics

[Jamie Roberts]

Bill and everybody,

diffraction occurs in the lens and has nothing to do with the sensor. {Snipped}

 

Lars--you're right about lens diffraction, but apparently the sensor plays a large role here as well.

 

See this article:

 

Luminous landscape

[lars_bergquist]

--------------

 

BTW, I find the M8 has so much apparent resolution that I'm in the market for a wide lens -- because I wan't to do landscape photog for the first time with 35mm. This is quite apparent in some of the panoramas from the Yosemite thread. And, I know I'll stay at f8 or wider. With my choice of either the CV15 or the WATE (if my ship comes in, maybe both), I won't need f8 anyway. I was looking at pix of the CV15: the DOF scale showed 5 and then infinity! And, the distance between them was about 4mm if the pic scale could be determined.

 

Bill, don't trust the DOF scales on lenses. These are computed for a circle of confusion of 1/30th of a millimeter, which presumes that the enlargement from the negative (or sensor) is no more than 3X! This was way too little already in the late 1930's, but Leitz never dared change the c.o.c. 1/20th was really required for 35 mm, and that is why demanding workers started using the next smaller f-stop when setting the hyperfocal distance, for instance. Today, with the smaller sensor of the M8, we should go to 1/15th. This means that we exactly double the numbers, i.e. we read DOF at f:4 on the scale if we set the lens to f:8.

 

With the 15 mm however, an aperture of 6.3 (halfway between 5.6 and 8) and a setting to 2 m will give you DOF from 1 m to infinity. This is how I use my CV 15.

 

The old man from the Age of Roll Film

[jaapv]

Sorry, Lars, you are wrong about sensor diffraction. That has to do with the relative size of the pixels and the Ayrie disk, which in itself is a result of the aperture. There is a relationship between both kinds of diffraction, the use of sensor filters and the final resolution. In another post I linked to an article by Sean McHugh, which I'm sure you'll find very interesting

[wparsonsgisnet]

Lars, extremely informative. Thanks.

 

I have been choosing hyperfocal distance in the same way you recommend, by setting the infininty mark at the next (larger) indicator on the DOF scale.

 

It is interesting how the M8 brings the physics lab to the house.

[Advertisement]

Hallo,

Look here: Leica M

[lars_bergquist]

Jamie,

it is correct that the properties of the 'sensor' – grain and emulsion depth of film, pixel diameter in a chip – will ultimately limit the resolution of the image. This however is a different matter than diffraction in a strict optical sense, which occurs in the optical system. Diffraction may well occur in the small lenses above the different pixel loci, but as these lenses are not image-forming – or at least, the individual captor cell does not record an image of its own, only the overall intensity of the light it captures – it is irrelevant. Optical diffraction is a continuous variable, while 'sensor resolution cutoff' is just that, an abrupt limit.

 

This is how I understand it. Diffraction happens when light paths are bent by passing close to optical edges; it has already happened when that light hits the film or the chip. Am I dead wrong?

 

P.S. I do know what a diffraction grating is.

 

The old man from the Age of Classical Physics