[normhead]

Having looked into this when Falconeye was still around, he pointed out the best lenses are sharpest at 2.8, and he had a theoretical explanation. Lenses like the 31 are sharpest at ƒ4. Most lenses are sharpest at ƒ5.6. But you can almost predict how sharp a lens is by where it's sharpest. The DFA* 50 1.4 meets his criteria for expecting sharpness well beyond normal. I'd be interesting in knowing if there really is a lens limited only by diffraction somewhere, I've actually never heard of one.

[Mistral75]

Amusing: the Tokina opera 50mm f/1.4 FF is "Made in Vietnam", therefore most probably manufactured by Ricoh Imaging since Kenko Tokina don't have any plant in Vietnam.

[robbiec]

Having looked into this when Falconeye was still around, he pointed out the best lenses are sharpest at 2.8, and he had a theoretical explanation. Lenses like the 31 are sharpest at ƒ4. Most lenses are sharpest at ƒ5.6. But you can almost predict how sharp a lens is by where it's sharpest. The DFA* 50 1.4 meets his criteria for expecting sharpness well beyond normal. I'd be interesting in knowing if there really is a lens limited only by diffraction somewhere, I've actually never heard of one.

I read this and thought immediately of the (D)FA50/2.8 Macro, then I started going back a few pages and noticed that you already brought it up. It is all theoretical though surely? DFA*50 is excellent wide open but you need to be careful with high contrast (water, bright metals) and you really need to nail focus. It's a more flexible lens. The Macro is not glass hampered as such in day to day stuff barring purple fringing but operationally it is a slower lens to use. I must do a controlled test with both at say 45cm and through f/2.8-f/8. It might be of interest to some but would answer a nagging question in my mind just how much difference there actually is. Based on what I know and seen, not much at all.

[sundown]

Gonna be a proud owner soon - come this Friday or next Monday that's my development story and update.

[normhead]

I read this and thought immediately of the (D)FA50/2.8 Macro, then I started going back a few pages and noticed that you already brought it up. It is all theoretical though surely? DFA*50 is excellent wide open but you need to be careful with high contrast (water, bright metals) and you really need to nail focus. It's a more flexible lens. The Macro is not glass hampered as such in day to day stuff barring purple fringing but operationally it is a slower lens to use. I must do a controlled test with both at say 45cm and through f/2.8-f/8. It might be of interest to some but would answer a nagging question in my mind just how much difference there actually is. Based on what I know and seen, not much at all.

The diffraction limit is in part set by the size of the pixels. Diffraction effects sharpness as the diffraction effect approaches 1 pixel. in width, and the smaller the fraction percentage the lens is affected by diffraction. If you've ever seen how a lens gets sharper as the aperture gets smaller you know where this is going. As the aperture gets smaller the effect of diffraction becomes more and more pronounced. Without diffraction a pinhole camera wold be the sharpest you can get. But the pinhole is so small, diffraction covers the whole frame.

Now look at the DFA 50 1.4



The lens is so sharp, diffraction starts to limit it's performance at ƒ4. Remember, smaller aperture means sharper image. When reducing the size of the aperture, the only thing that affects resolution is diffraction. Not only that, at 1.4 the lens is almost as good as it is at ƒ4. Below ƒ4, the increase in resolution from the smaller aperture is bigger than the loss of resolution due to diffraction

For those of us who have been looking at lens charts for years, this is very unusual. Essentially you get very small increases from the smaller aperture until ƒ4, and because of the high resolution of the lens, diffraction becomes a factor sooner than it would with a less accurately designed lens.

The falloff after ƒ4 as the aperture gets smaller is because diffraction is degrading image quality faster than reducing the aperture can increase it. Looking at the fall off on the edges at ƒ8 and beyond, we can probably assume diffraction affects the edges quicker than it affects the centre.

A more typical but also highly rated lens the DA*200 has graph much more representative of lens charts I've seen over the years.



This is the #1 rated Pentax lens in this article....

Top 28 Best Pentax K-Mount Lenses 2020 | ePHOTOzine

But that would seem to be based on it's sharpness at ƒ8. In earlier times it was a huge savings in cost and manufacture to produce edge sharp wide open glass like the DFA 50 1.4, which was why hardly anyone did it. However with modern design and manufacturing techniques the cost has come down so much that a company like Pentax can design and produce them, without staying manual and producing for a large number of mounts.\

Look at the DFA 28-105 at 50mm for comparison. It in it's own right it is a great lens but (you probably won't find that kind of graph with that kind of edge shaprness from anyone else's kit lens) the 50 1.4 is significantly better. Even if you don't take into account CA and transitions.


In fact from 1.4 to ƒ4 the DFA is shaper than the DFA 28-105 is at it's sharpest at ƒ8.

The tighter the design and tolerances of the lens the less it will be affected by a smaller aperture, and diffraction will be the only limit on the resolution of the images produced.

[robbiec]

Don't forget the Macro



Remember folks, this was tested on APSC circa 2014 so K-3 time...Until Norm gave me a lecture / lesson / briefing () my basic understanding of it was that it started to kick in on APSC at f/8 - f11 and on FF from f/11 so as you can see by the chart above it looks like a lens that could be diffraction limited. Or to put it another way, when the big one starts to fall off, take out the Grenade So DFA* from f/1.4 to f/5.6 and Macro from f/5.6 through f/11

[clackers]

DFA*50 is excellent wide open but you need to be careful with high contrast (water, bright metals) and you really need to nail focus. It's a more flexible lens. The Macro is not glass hampered as such in day to day stuff barring purple fringing

?

I have the DFA 50 Macro, it purple fringes in water, bright metals, etc - and that's at f2.8!

It will have way more chromatic aberration than the DFA*50, which has very little, possibly better than any autofocus fifty of any brand.

[robbiec]

?

I have the DFA 50 Macro, it purple fringes in water, bright metals, etc - and that's at f2.8!

It will have way more chromatic aberration than the DFA*50, which has very little, possibly better than any autofocus fifty of any brand.

Day to day stuff for the macro is close in and stopped down. I can make all my lenses misbehave, but I try not too

[TwoUptons]

Amusing: the Tokina opera 50mm f/1.4 FF is "Made in Vietnam", therefore most probably manufactured by Ricoh Imaging since Kenko Tokina don't have any plant in Vietnam.

Does that mean it's really a Tamron after all?

kidding, kidding...

-Eric

[clackers]

Does that mean it's really a Tamron after all?

kidding, kidding...

-Eric

Yes, Tamron do have a factory in Hanoi ... the conspiracy lives on!

[TwoUptons]

Yes, Tamron do have a factory in Hanoi ... the conspiracy lives on!

Every really good ridiculous conspiracy needs a nugget of truth...

-Eric

[Class A]

Remember folks, this was tested on APSC circa 2014 so K-3 time...

Not sure which graphs you are comparing but note that you cannot compare graphs produced with different cameras for a number of reasons. If graphs are produced by the same tester, using the same camera, using the same test parameters then the graphs are comparable, otherwise they are not.

N.B., not everything that normhead wrote is correct but I won't debate technical matters in a thread devoted to a lens.

[robbiec]

Not sure which graphs you are comparing but note that you cannot compare graphs produced with different cameras for a number of reasons. If graphs are produced by the same tester, using the same camera, using the same test parameters then the graphs are comparable, otherwise they are not.

N.B., not everything that normhead wrote is correct but I won't debate technical matters in a thread devoted to a lens.

Same test house, different camera so different format (DFA*50 on K-1ii / DFA50 Macro on K-5iis). It was more an Illustration mainly - it would match against the DA200/2.8 though - correction - 200 was tested in 2010 so K-7, different sensor again. I must get myself an IMATEST setup.

[RobA_Oz]

N.B., not everything that normhead wrote is correct but I won't debate technical matters in a thread devoted to a lens.

Given that this thread was about the development of the D FA*50, I wouldn’t feel constrained by it, at this stage.

Having said that, Norm’s comment about diffraction starting to affect the lens’s performance from f4 had me thinking about the conventional view, which is that you generally don’t need to worry about it until at f16 and above. My recollection of the basis for that was, as Norm says, that diffraction effects will fill the aperture at f16, so, if I’m correct, they’ll also affect the peripheral rays at any aperture setting, but increasingly so as the f-stop increases. So, given that the edge resolution of the D FA*50 starts to noticeably deteriorate firstly at f4, while the centre holds up, this seems consistent with that conclusion.

So, was that what you were contesting, or have I missed something?

Either way, the wide open performance of the D FA*50 is phenomenal, and even on its own suggests that it’s up there with the likes of Leitz or Zeiss lenses, which, if I remember correctly, were designed to perform like that, and previously unchallenged in that respect.

[angerdan]

Amusing: the Tokina opera 50mm f/1.4 FF is "Made in Vietnam", therefore most probably manufactured by Ricoh Imaging since Kenko Tokina don't have any plant in Vietnam.

Interesting, thx for sharing