[BigMackCam]

So I've now looked at the MTF figures for a bunch of the Limited lenses, and - subjectively - this is what I come up with:

DA15 - very good from F4, best centre at F5.6, best across the frame at F8 & F11
DA21 - good centre from F3.2, best centre F4 / F5.6, best across the frame at F5.6 / F8, still good across frame at F11
DA35 F2.8 - great from F2.8, best across frame F4 - F8, still good at F11
DA40 - great from F2.8, best from F4 - F8, still good at F11
DA70 - great from F2.4 - F2.8, best at F4 & 5.6, great at F8

Although this doesn't include appraisal of CA performance (or vignetting, for example), it makes me question whether it's simply possible to say "one or two stops down to get decent performance"; since some of the lenses perform really well - at least, from an MTF perspective - wide open...

[clackers]

diffraction is usually visible at f/8 on crop and f/11 on ff, because that's where they both have the same dof

Incorrect, Osv, as usual with you. (Rolls eyes, again)


As far as the sensor goes, diffraction depends to the first approximation on pixel size, not format size.


Diffraction Limited Photography: Pixel Size, Aperture and Airy Disks


Forum members, do not trust what this man says.

[UncleVanya]

Also one thing that bugs me. With the FA lenses - we get MTF data but it's geared to film era cameras right? So diffraction is likely not a consideration for the MTF program line. I don't know if any lenses would have an MTF pushing the aperture too small for a given sensor's pixel pitch but I suppose it is possible.

As I pointed out earlier - you can get by with more than what the math may suggest since it really only applies to visual observation and unless you print large and stick your nose up against the frame it's hard to see early onset diffraction.

[osv]

So I've now looked at the MTF figures for a bunch of the Limited lenses, and - subjectively - this is what I come up with:
DA15 - very good from F4, best centre at F5.6, best across the frame at F8 & F11...

that's questionable, and here is the proof.

two sets of tests: same lenses on ff vs crop bodies, all shot at f/11... in both tests, the crop cameras have significantly more visible blur at f/11:

Canon EF 100mm f/2.8L IS USM Macro Lens Image Quality

Canon EF 35mm f/1.4L II USM Lens Image Quality

i've posted the link with details multiple times already on this forum: "Typically, the effects of diffraction softening do not even begin to become apparent until f/11 on FF (f/7.1 on APS-C and f/5.6 on mFT -- 4/3) Equivalence

none of that makes one format better than the other; it's just a statement of fact, because when diffraction is equal, so is the dof.

[UncleVanya]

that's questionable, and here is the proof.

two sets of tests: same lenses on ff vs crop bodies, all shot at f/11... in both tests, the crop cameras have significantly more visible blur at f/11:

This ignores the fact that sensor pixel pitch is what causes this. If you have a super resolution FF it will show diffraction sooner than a low res FF. And a low res APSC will not show diffraction as soon as a high res one. So a K100D will be able to use higher f/stops than a K3... .

Don't take my word... read this:
Diffraction Limited Photography: Pixel Size, Aperture and Airy Disks

[6BQ5]

Go to a site like photozone.de or lenstip.com and look up the test charts for the lenses you are interested in. All the data is there. You will see a pattern where peak sharpness is somewhere between f/4 for fast primes to f/8 for long zooms.

[osv]

This ignores the fact that sensor pixel pitch is what causes this.
If you have a super resolution FF it will show diffraction sooner than a low res FF. And a low res APSC will not show diffraction as soon as a high res one. So a K100D will be able to use higher f/stops than a K3... .

so ignore the ff photos! the crop pics looked horrible at f/11, and they show blurring even at f/8... that's all that anyone needs to know, i proved the point with just the crop pics.

i agree that pixel size matters, but it's largely insignificant, when compared to the effects of aperture size on diffraction... and even then it's a trade-off, not a one-sided liability:

"Pixels with half the size will begin to suffer diffraction softening at half the f-ratio. However, since pixels with half the size will also deliver more resolution, diffraction simply eats into the resolution advantage -- it does not put the smaller pixel sensor at a disadvantage (the same with all other effects, such as motion blur, camera shake, and noise). However, for deep DOF photography, diffraction softening is one of the primary issues limiting the advantages to greater pixel counts." Equivalence

for example: "As you can see, small pixels resolve more detail than large ones, even when diffraction-limited. ...the point here is, the 7D will still out-resolve the other 1.6-crop Canon's, even when shooting at diffraction-limited apertures. Don't be afraid of small pixels because of diffraction! They help!" Are small pixels bad for diffraction, part 2: Canon EOS 7D / 10D - 70D Talk Forum: Digital Photography Review

Don't take my word... read this:
Diffraction Limited Photography: Pixel Size, Aperture and Airy Disks

been there done that several times over the years; i see nothing in there that disagrees with anything that i've posted, but their calculator confuses the issue... diffraction is overwhelmingly caused by dof, not pixel size:

"Most, if not all, online DOF calculators (as well as DOF tables) are based on "standard viewing conditions" of an 8x10 inch photo (or any photo displayed with a 12.8 inch -- 325mm -- diagonal) viewed from a distance of 10 inches with 20-20 vision."

[normhead]

Almost all Pentax lenses are sharpest at ƒ5.6. The three amigos are sharpest at ƒ4. Diffraction starts to set in both on FF and APS-c at ƒ8 but the drop-off at ƒ8 is not so bad. By ƒ11 it's more noticeable, but it's not until ƒ32 you start to see the coke bottle lens effect. You might want to look over the photozone charts, they are presented in an understandable format, unfortunately they haven't done any Pentax testing on anything newer than a K-5. But you get a nice feel for how a lens produces with good distortion and CA numbers.

Pentax K Lens Tests
Click through the list of K mount lenses, by the time you're done you will have seen some very different philosophies in lens design, from the centre sharp 77, which excels at the centre but not so much on the edges, to the 70 2.4 sharp across the frame, and not much difference from ƒ2.4 to ƒ 11.I default to ƒ5.6 on most lenses, unless I know I want a lot of depth of field when I go to ƒ8 or ƒ11 or 16. If I want a nice creamy background I use a faster lens and a wider aperture. I've never found a lens where the basic approach is much different.

Some ƒ4 lenses are very good wide pen ut are still better at ƒ5.6. I would never hesitate to use my DA*605250 wide open if I needed a faster shutter, Same with he DA* 300 ƒ4. The DA*200 is almost identical MTF, from ƒ4 to ƒ8. the numbers are engaging and different lenses can have quite different characteristics.

[osv]

since it appears that no one on this forum has ever tested their lenses for diffraction losses, i'll put up this 21mp crop f/4 vs f/11 comparison: Canon EF 35mm f/1.4L II USM Lens Image Quality

one crop camera, one lens, two different apertures.

notice how much the contrast has dropped at f/11... contrast(or microcontrast if you will) is an integral part of the perception of sharpness.

[BigMackCam]

that's questionable, and here is the proof.

two sets of tests: same lenses on ff vs crop bodies, all shot at f/11... in both tests, the crop cameras have significantly more visible blur at f/11:

Canon EF 100mm f/2.8L IS USM Macro Lens Image Quality

Canon EF 35mm f/1.4L II USM Lens Image Quality

i've posted the link with details multiple times already on this forum: "Typically, the effects of diffraction softening do not even begin to become apparent until f/11 on FF (f/7.1 on APS-C and f/5.6 on mFT -- 4/3) Equivalence

none of that makes one format better than the other; it's just a statement of fact, because when diffraction is equal, so is the dof.

My subjective analysis was based on the MTF figures at Photozone for each lens... And yes, F11 shows to be worse than F5.6 on those lenses but still very good on several of them.

[pinholecam]

Its usually 2 stops from wide open.

But I find that its kind of pointless since we use what we gotta use right?
As in the aperture to give the DOF needed, factoring the ability to hand hold at that setting and that ISO remains reasonable.
Also, the real use factors like 30s auto camera exposure limit if w/o a IR remote to use B mode.

[BigMackCam]

Its usually 2 stops from wide open.

But I find that its kind of pointless since we use what we gotta use right?
As in the aperture to give the DOF needed, factoring the ability to hand hold at that setting and that ISO remains reasonable.
Also, the real use factors like 30s auto camera exposure limit if w/o a IR remote to use B mode.

I do get that, and I have no problem using a lens wide open if needed. I was merely wondering if the optimal apertures for each lens had been summarised. No matter, I can pull the info together from various reviews.

[kh1234567890]

So I've now looked at the MTF figures for a bunch of the Limited lenses, and - subjectively - this is what I come up with:

DA15 - very good from F4, best centre at F5.6, best across the frame at F8 & F11
DA21 - good centre from F3.2, best centre F4 / F5.6, best across the frame at F5.6 / F8, still good across frame at F11
DA35 F2.8 - great from F2.8, best across frame F4 - F8, still good at F11
DA40 - great from F2.8, best from F4 - F8, still good at F11
DA70 - great from F2.4 - F2.8, best at F4 & 5.6, great at F8

Spot on from my experience, except maybe the DA15 - I would not call its f4 corner performance 'very good'. More like 'better than the end of a beer bottle' ...

[BigMackCam]

Spot on from my experience, except maybe the DA15 - I would not call its f4 corner performance 'very good'. More like 'better than the end of a beer bottle' ...

I've never shot mine wide open, so it's good to get a view on that! I'll have to give it a try, just for laughs...

[UncleVanya]

I've never shot mine wide open, so it's good to get a view on that! I'll have to give it a try, just for laughs...

I have and it doesn't suck depending on what your subject is.