AF 50mm f/1.4 Purple Fringing

[ronb]

I bought a new AF 50mm f/1.4 lens to go with my D300 and took some test shots the other day. It is quite soft at

1.4, though I can see where the shallow DOF would be nice for portraits and some other shots. There is

pronounced (obnoxious) purple fringing at 1.4, however; I thought this was supposed to decrease (or is that for

lateral chromatic aberration) at smaller apertures/stopping down, but it goes away at f/2.8.

So the questions I have are these: is a large amount of purple fringing normal, and is everyone's copy of this lens

soft at f/1.4? Does anybody have experience with PF the the AF 50mm f/1.8? Thanks.<div></div>

[Two23]

The 50mm f1.8 is just as bad. I'm dumping ALL my single focal lenses and have been replacing them with modern f2.8 pro lenses. The older lenses don't have ED elements or the modern digital friendly coatings. I have had it with these over hyped dinosaurs screwing up my shots.

 

 

Kent in SD

[Jim_Tardio]

That's precisely why Nikon needs to update their prime lens line. However, you're likely to get some CA in a shot like that even with modern zooms.

[Matt Laur]

Yup. It's interesting shooting with my Nikon 50/1.8 or 60/2.8 - and noticing that problem - and then mounting up my Sigma 30/1.4... clean as a whistle under similar circumstances. As Kent mentions, my good Nikon zoom glass doesn't have any such trouble. But the older-style primes do need some help. In the meantime, Sigma got some of my money!

[michael_bradtke]

Kent

 

ED glass is not going to fix everything. My 180 f/2.8 has CA my 80-200 f/2.8 has CA.Both of these lenses have ED

glass in them. The 70-200 I have used will show CA.

I can not think of a lens that Nikon makes that will not show fringing in a shot like the one you have posted. The

contrast range is to high and the sensor is blooming.

 

Lens coatings are not going to fix CA. They will help with ghosting and flair.

 

The only way to build a lens that will show no CA is to have it focus all the wave lengths of light to the same point on

the film plane.

 

Lens design is a set of compromises. You want super speed you loose some sharpness and contrast wide open.

 

You want fast with no CA cough up the money and buy a Noct 58 f/1.2. BTW its a design from 1977 and has no ED

glass in it. If you would like to read about its development go here

 

http://imaging.nikon.com/products/imaging/technology/nikkor/n16_e.htm

 

There is one on Ebay right now. The starting bid is only $2600.00

 

Part of being a good photographer is knowing the limitations of your gear.

[Matt Laur]

Michael: wouldn't you say it's more a question of degree? Sure, all (reasonably obtainable!) lenses will do this to some degree. But it's a lot more correctable in post as seen from some lenses, compared to others. The very marginal bit of this I see when shooting outdoors in contrasty settings with my 70-200/2.8 cleans up with a twitch of the mouse in Capture NX. But it's a lot harder to chase around, as seen coming from some other lenses. ED isn't a magic bullet... but it sure helps.

[dangin]

my Nikon 50mm f/1.4 exhibits the exact same symptoms. f/1.4 is soft and usable only under certain circumstances. but realistically, one shouldn't be shooting that wide open all of the time any way. :)

[ronb]

I meant to say I thought this (lateral CA) was supposed to get worse on stopping down, but the purple I'm seeing in this shot is less by f/2 and is gone by f/2.8.

By the way, I just tested it against my Series E f/1.8 50mm lens and the old one is much sharper and shows considerably better contrast at f/2. There is really no comparison... even my wife could see it clearly. So I may try the 50mm AF f/1.8 just to compare sharpness and contrast w/the old lens.

[arthuryeo]

First, you are shooting the leaves against the sky. This is a very high contrast situation. It will cause CA on most lenses without ED elements.

 

Second, since you're using a D300 with the eXpeed chip, why don't you shoot JPG if you intend to shoot such high contrast images. The CA will be cleaned out by the firmware.

 

Third, you can also clean it out in CaptureNX if you are shooting in NEF.

[oskar_ojala]

Like Arthur said, the CA reduction of the D300 takes care of much of the problem. What's left is usually too little to be of a major consequence.

 

If you Ser. E 50/1.8 is much better at f2, then something is probably wrong.

 

Current pro zooms with their exotic optical formulas do a wonderful job, but they are not without limitations too, e.g. the 24-70/2.8 has quite significant field curvature and the 70-200/2.8 has limitations to corner performance on FX. It's just about picking one's compromises, although in all fairness there are a number of primes that Nikon could improve. I believe that the Zeiss 50/1.4 is better performing that the Nikkor equivalent, even though the Zeiss isn't an exotic design.

[Peter_in_PA]

You are creating an absolute worst-case scenario, it's almost as if you were trying to get this as an effect. You say it's

gone by f2.8? Great. problem solved. Almost nothing is in focus at f1.4 anyway, you only use that for pretty specialized

photos.

[ronb]

Thanks for all the input.

I know it's a worse-case scenario; I'm not likely to take a shot like this to frame on my wall. I was interested in whether the D300 would remove this CA or fringing (the shots were in JPEG) and it did not. And I was interested in whether the old lens had as bad a CA - it's about the same or less - more magenta than purple.

And I know the 1.4 is said to be soft wide open, but compared to the Series E lens at f/2 it was surprisingly bad. It'll probably go back to the store.

[dan_brown4]

This doesn't make any sense. I've never noticed a chromatic aberration problem in the 50/1.4. Looking at the test results

(photozone.de), the Nikkor 50/1.4 easily betters the Signma 30/1.4 in both sharpness and lesser CA's. It is also better than

most f2.8 zooms. The posted pictures is obvisouly blown highlights againts underexposed leaves. What good is a

photograph exposed in that manner? Then why conduct such a test.

[michael_bradtke]

Matt

 

Yes it is a question of degree. But most of the problem with this shot is the shot itself.

[dan_brown4]

On secondthought, this test is not brutal enough. Take the hood off and put a cheap uV filter on there. But first, smudge

the filter with your fingers. Then, get a couple of those painfully bright flashlights, and shine them diagonally into the front

element, making the internal barrel of the lens glow. Now, recompose with the noon sun in the frame, blowing the

highlights while underexposing everything else. Then we'll be able to tell just how bad this lens really is ;~)

[ronb]

The highlights were not overexposed according to the histogram. I know it doesn't make any sense; that's why I posted the shot. The question is did I get a bad sample or is this typical? I'm going to take the same shot in raw to see if there's any difference.<div></div>

[huyzer]

<p>Hi Ron B,<br /> <br /> I have this problem also with my lens (50mm 1.4 AFD) on the S5 Pro. I thought it might've been my copy of the lens. But thanks to your thread, I feel at ease ... well, you know what I mean. I thought it might have been a digital thing. I looked at some of my film shots, and it exhibited this too, but to a lesser degree. The transition seems smoother too, and not as "out there". I'm guessing the microlenses on the sensor exacerbates the problem.<br /> <br /> Take care,<br /> Huy<br>

P.S. I JUST found out that Nikon came out with an AF-S version with a 9-blade diaphram blade, and it's rounded. SWEET! I hope it doesn't exhibit this purple fringing!</p>