Aperture question, re: sharpness

a13 said:

Just a couple of musings that I'm posing to myself..

1. Is the absolute sharpest a lens can and will ever be, when it is stopped all the way down to its smallest aperture?

Click to expand...

No. This is actually likely to be the softest a lens can and will be, due to diffraction effects.

a13 said:

Follow-on from above,

2. If so, is that because it's using the very centre of the lens when it's stopped all the way down like this? Meaning that 1) will be true for each and every lens, without exception (well, except maybe ones designed to be sharper on the edges than in the centre...)?

Click to expand...

No.

a13 said:

3. If a FF lens is stopped down to f/22, what is the f-stop on an m43 that would match the same exact PHYSICAL opening diameter of the FF lens @ f/22?

Click to expand...

F/numbers are independent of format. What matters is the focal length.

a13 said:

Just a couple of musings that I'm posing to myself..

1. Is the absolute sharpest a lens can and will ever be, when it is stopped all the way down to its smallest aperture?

Click to expand...

Absolute sharpness does not exist. Max. sharpness could be achieved for every given lens @ different F stop

a13 said:

Follow-on from above,

2. If so, is that because it's using the very centre of the lens when it's stopped all the way down like this? Meaning that 1) will be true for each and every lens, without exception (well, except maybe ones designed to be sharper on the edges than in the centre...)?

Click to expand...

Because of refraction. Stopping down F you reduce the influence of refraction from the closest to the barrel part of front glass of the lens.

a13 said:

Leading question in relation to above,

3. If a FF lens is stopped down to f/22, what is the f-stop on an m43 that would match the same exact PHYSICAL opening diameter of the FF lens @ f/22?

Click to expand...

There is no linear dependency. Very roughly FF lens @ f/22 about 43 lens @ f/16. Again, this is a ballpark assessment.

The sharpest aperture on a lens is most likely 2 or 3 stops down from the maximum. Or, on a lens with an maximum aperture of f 1.4, that might mean that f 2 is actually sharper. Some lenses are made to be used wide open and will exhibit their best sharpness at that point. Closing any lens down as far as possible will not deliver best quality.

a13 said:

Just a couple of musings that I'm posing to myself..

1. Is the absolute sharpest a lens can and will ever be, when it is stopped all the way down to its smallest aperture?

Click to expand...

There's a lot of technical explanation for this which includes a lot of factors, but when the aperture is stopped down all the way you'll have more depth of field/more in focus, not necessarily a sharper image.

a13 said:

Follow-on from above,

2. If so, is that because it's using the very centre of the lens when it's stopped all the way down like this? Meaning that 1) will be true for each and every lens, without exception (well, except maybe ones designed to be sharper on the edges than in the centre...)

Click to expand...

a13 said:

Leading question in relation to above,

3. If a FF lens is stopped down to f/22, what is the f-stop on an m43 that would match the same exact PHYSICAL opening diameter of the FF lens @ f/22?

Click to expand...

The f-stop is a ratio of lens focal length and aperture diameter, focal length divided by aperture diameter equals f-stop. So if you have a 50mm lens at f22, the diameter of the aperture is 2.27mm (50mm lens/2.27mm aperture opening equals f22), that goes for FF and m43. The lens is a lens regardless, the sensor is what captures the lens projection and initiates the "crop-factor." So the 50mm f22 on a m43 camera will LOOK like 100mm f44 if you compare it to a FF point of view, even though the light gathering for the same exposure is still at f22. Hope I didn't confuse. I'm sure many more people will chime in and start arguing about all this because there's so many technicalities that can be brought in.

Slightly simplifying, two physical phenomena affect lens sharpness: aberrations and diffraction. Aberrations are a result of lens design/construction, diffraction is a fundamental physical effect.

Diffraction limits the resolution of a lens in a way that resolution decreases when the f-number increases. If a lens would be aberration free, the highest resolution would be obtained wide open. In fact, the world's highest resolution lenses are around f/0.19, but they're not photographic lenses and cannot be used as such.

A general trait in most photographic lenses is that aberrations are the highest wide open, and reduce when stopped down. This is the reason why sharpness is not optimal wide open, then improves when stopping dow, but decreases again at smaller apertures due to the abovementioned diffraction efffects. This is no absolute truth, however. There are some lenses that perform best wide open, and there are lenses that have a significantly worse performance at smaller apertures than the diffraction limit.

Hope this helped.

a13 said:

Just a couple of musings that I'm posing to myself..

1. Is the absolute sharpest a lens can and will ever be, when it is stopped all the way down to its smallest aperture?

Click to expand...

Most M4/3 lenses are at their absolute worst at f/22 - due to aperture diffraction effects. Best quality appears usually somewhere in the range wide open to f/8 for most lenses.

a13 said:

Follow-on from above,

2. If so, is that because it's using the very centre of the lens when it's stopped all the way down like this? Meaning that 1) will be true for each and every lens, without exception (well, except maybe ones designed to be sharper on the edges than in the centre...)?

Click to expand...

Wide open most lens design and build compromises come into play and may have an effect on the image, stopped right down diffraction causes a very obvious drop-off in quality, so at some point in between the two there is a good sweet spot or good spread of apertures. Some M4/3 lenses are so good that they really are at their best at maximum widest open aperture.

In theory the highest resolution possible for a perfect lens always happens at maximum open aperture, but real life intervenes and it usually is some smaller stop where it is best overall. Though some M4/3 lenses do approach perfection....

a13 said:

Leading question in relation to above,

3. If a FF lens is stopped down to f/22, what is the f-stop on an m43 that would match the same exact PHYSICAL opening diameter of the FF lens @ f/22?

Click to expand...

If a 50mm lens at f/22 then 50/22=2.72mm, same for any format be it pocket camera or 8 x 10 inch plate camera. Same at the other end so 50mm lens at f/2 = 25mm aperture size. Divide the actual focal length of the lens by the aperture to get the aperture diameter.

For a nice interactive display of lens performance, look at the lenses at http://www.slrgear.com/reviews/showcat.php/cat/7 (that's for the Oly lenses) and choose ones marked "(Tested)" and click on the Blur Index graph and play with focal length and aperture to see how the quality changes. The lower and flatter the graph the better. Also check out some full frame primes and zooms from Canon/Nikon and see how they compare for quality.

After you have checked a few of those you can see why f/4 may be more preferable to f/22.

Regards.... Guy

Conrad69 said:

Slightly simplifying, two physical phenomena affect lens sharpness: aberrations and diffraction. Aberrations are a result of lens design/construction, diffraction is a fundamental physical effect.

Diffraction limits the resolution of a lens in a way that resolution decreases when the f-number increases. If a lens would be aberration free, the highest resolution would be obtained wide open. In fact, the world's highest resolution lenses are around f/0.19, but they're not photographic lenses and cannot be used as such.

A general trait in most photographic lenses is that aberrations are the highest wide open, and reduce when stopped down. This is the reason why sharpness is not optimal wide open, then improves when stopping dow, but decreases again at smaller apertures due to the abovementioned diffraction efffects. This is no absolute truth, however. There are some lenses that perform best wide open, and there are lenses that have a significantly worse performance at smaller apertures than the diffraction limit.

Hope this helped.

Click to expand...

This is the best

You've gotten the answers. To summarize.

1. No, a lens is NOT going to be at its sharpest stopped all the way down to, say, f/22. At some point most lenses become subject to diffraction. I shoot past f/11 only when I I'm already at 1/4000th sec and ISO 200 or 100 and have no other choice. Ansel Adams and the other members of Group f/64 took that name because they had to go to f/64 to get decent depth of field on their huge cameras; but things have changed a lot.
2. Most lenses have a "sweet spot" — an aperture (and if its a zoom lens, a focal length) that provides the best optical performance. Where this sweet spot is varies from lens to lens. DxO or the manufacturer's own statistics might help you find it for your lenses. The common rule of thumb is that for most lenses, the sweet spot on a prime lens is not wide open, but stopped down one or two stops. So if you've got an f2.8 lens, the sweet spot might be f4.5 or f5.6. But it varies.
3. A good lens's sweet spot won't be dramatically better than the other apertures, that is, buy good lenses and you can more or less stop worrying about this problem.

Now if you want sharp pictures, the best thing you can do doesn't necessarily cost any money at all: Understand how to light your photos well. It's kind of absurd to say that "good light trumps good glass" — you'd always prefer to have both — but it feels true nonetheless. There are other things that can help your photos get sharper: tripod, appropriately faster shutter speeds and others. But better lighting is just about the best way to improve the sharpness of your photos, especially if better lighting allows you to reduce your camera's ISO to its optimal setting and/or increase your shutter speed.

Final point: Sharpness is just about the least interesting aspect of a good photo.

Will

a13 said:

Just a couple of musings that I'm posing to myself..

1. Is the absolute sharpest a lens can and will ever be, when it is stopped all the way down to its smallest aperture?

Click to expand...

NO, sharnes increases when you go down and then again decreases , fx. pany 20 1.7 according to


from 1.7 to 5.6 increases 5.6 to 16 decreases

use this site and play around with Blur index table of tested lenses like that for pany 20 1.7 , with slider change aperture and graph shows you sharpness of lens at that aperture.

a13 said:

Follow-on from above,

2. If so, is that because it's using the very centre of the lens when it's stopped all the way down like this? Meaning that 1) will be true for each and every lens, without exception (well, except maybe ones designed to be sharper on the edges than in the centre...)?

Click to expand...

NO, diffraction comes into play

a13 said:

Leading question in relation to above,

3. If a FF lens is stopped down to f/22, what is the f-stop on an m43 that would match the same exact PHYSICAL opening diameter of the FF lens @ f/22?

Click to expand...

it should be something smaller in diameter as we need 4 times less light thru lens to have the same aperture as on FF cameras as we have 4 times smaller sensor area to illuminate, but how much smaller I have no idea and probably its not simply linear. There are some folks who says that f2.0 has the same opening dimensions , no matter it,s FF camera or Nokia smartphone with 2.0 lens, but it is hard to believe.

Ok, thanks for all the answers. To re-iterate, the question about FF @ f/22 vs m43 was purely about physical diameter of the hole, not light equivalence or any such conversion.. I was thinking only about the distance in mm from each end of the aperture blades.

reinish said:

reinish said:

3. If a FF lens is stopped down to f/22, what is the f-stop on an m43 that would match the same exact PHYSICAL opening diameter of the FF lens @ f/22?

Click to expand...

it should be something smaller in diameter as we need 4 times less light thru lens to have the same aperture as on FF cameras as we have 4 times smaller sensor area to illuminate, but how much smaller I have no idea and probably its not simply linear. There are some folks who says that f2.0 has the same opening dimensions , no matter it,s FF camera or Nokia smartphone with 2.0 lens, but it is hard to believe.

Click to expand...

Now that is a confusing answer!

To go back to the OP question we need to ask, which lens on the FF?

If it were to be a 50mm lens then f/22 aperture is 50/22=2.3mm (rounded) and if a 50mm lens on M4/3 50/22=2.3mm (rounded).

But to get same framing on M4/3 as on FF then on M4/3 we need to use a 25mm lens, so to get the f/number needed to match the physical size (begs the question, why?) then 25/2.3=f/11 (rounded to nearest aperture).

If a more likely and usable f/2 on 50mm then aperture diameter is 50/2=25mm, to get the same physical aperture (again, why?) and same framing on M4/3 we need 25mm lens and now 25/25=f/1.0 (not sure that we have one of those).

Regards..... Guy

To reiterate, the f# is a ratio between the focal length and the aperture opening. So, a 50mm lens at f/2 is 50/2 = 25mm in diameter. A 100mm lens at f/4 is 100/4 = 25mm in diameter. A 25mm lens at f/1 is 25/1 = 25mm in diameter. It doesn't matter if the lens is designed for full frame, aps, m4/3, or any other format. Divide the focal length by the f# and you get the diameter of the entrance.

a13 said:

Ok, thanks for all the answers. To re-iterate, the question about FF @ f/22 vs m43 was purely about physical diameter of the hole, not light equivalence or any such conversion.. I was thinking only about the distance in mm from each end of the aperture blades.

Click to expand...

Actually, f/22 is not the physical size of the hole, but the size of the hole as seen through the front end of the lens. You can usually ignore that distinction.

No!

Every lens have a different characteristic, but usually would be sharpest at 1-2 stop down from its widest aperture.

As a suggestion, why not test your lens? Put the camera on a tripod, 2 second shutter delay, and photograph something with a lot of detail, like a page from a news paper. Take shots from wide open (say, f2.0) to f22. I think you will find that the lens is most sharp around f 3.5 to f 8.0 on MFT. Wide open won't be too bad. But you will start to notice some degradation by f11, more by f16 and f22 will be pretty bad. A MFT 25mm lens at f22 has a lens opening of only about 1.1 mm. This is almost like a pin hole camera. This, as others have pointed out, is caused by diffraction, which results from the light scattering as it passes through a narrow opening. What you get at f22, is greater depth of field. But I doubt the extra DOF at f22 over, say, f11, is worth the loss of resolution.

Another consideration is image size. A small Web image or a small print may look fine shot at f22.

If you find yourself trying to get a slow shutter speed in bright light, and you're at your camera's max shutter speed, lowest ISO, and f22, the real solution is to have a 3 to 5 stop neutral density filter and get away from f22.

I had a Raynox 250 on a few shots I was doing yesterday and the only way to get any DOF at all was to stop it all the way down to f/22. I tried at 11 but didn't seem any better than f/5; probab;y should have tried 16 or something.

a13 said:

I had a Raynox 250 on a few shots I was doing yesterday and the only way to get any DOF at all was to stop it all the way down to f/22. I tried at 11 but didn't seem any better than f/5; probab;y should have tried 16 or something.

Click to expand...

When extremely close as in macro then the depth of field shrinks like crazy. Only then you may use f/16 or f/22 to try and make the image look a bit better but all at the expense of loss of resolution due to diffraction. The compromise works OK then.

The other technique to gain depth of field for macro is to take a bunch of shots at the best aperture, say f/4 or f/5.6 or whatever is best for that particular lens, but all the while shifting the focus through the subject. Then later use image stacking or focus stacking software to combine all the sharp bits of the set of images to make it look like there's great depth of field. For example here's one commercial program that sprung from the microscopy world.

Regards.... Guy

Guy Parsons said:

a13 said:

I had a Raynox 250 on a few shots I was doing yesterday and the only way to get any DOF at all was to stop it all the way down to f/22. I tried at 11 but didn't seem any better than f/5; probab;y should have tried 16 or something.

Click to expand...

When extremely close as in macro then the depth of field shrinks like crazy. Only then you may use f/16 or f/22 to try and make the image look a bit better but all at the expense of loss of resolution due to diffraction. The compromise works OK then.

The other technique to gain depth of field for macro is to take a bunch of shots at the best aperture, say f/4 or f/5.6 or whatever is best for that particular lens, but all the while shifting the focus through the subject. Then later use image stacking or focus stacking software to combine all the sharp bits of the set of images to make it look like there's great depth of field. For example here's one commercial program that sprung from the microscopy world.

Regards.... Guy

Click to expand...

Here's an example of image stacking using 19 frames