New 58/1.4 has Focus Shift?

[glenerrolrd]

Diglloyd has obtained a new 58/1.4AF Nikkor and has been critical of its focus shift . Having followed lloyd s tests pretty closely I have no doubt that the lens he has in hand exhibits some level of focus shift ...but my question is "so what ?"

Evidently its not all that unusual in fast lenses and comes from a common aberration . Every lens design has trade offs (even the uber Zeiss ) and the Nikon lens designers accepted some focus shift to produce the new lens .

Personally when I have a lens that has focus shift ..I try to set it up to slightly front focus . The older 35/1.4 M asph had this problem ..but set it for dead on at F2 and with a little thought you could avoid any real impact .

I don t generally shoot at F1.4 unless I need to or am seeking some very limited DOF . If I did of course the lens can be set for perfection wide open .

 

[Bob]

Diglloyd has obtained a new 58/1.4AF Nikkor and has been critical of its focus shift . Having followed lloyd s tests pretty closely I have no doubt that the lens he has in hand exhibits some level of focus shift ...but my question is "so what ?"

Evidently its not all that unusual in fast lenses and comes from a common aberration . Every lens design has trade offs (even the uber Zeiss ) and the Nikon lens designers accepted some focus shift to produce the new lens .

Personally when I have a lens that has focus shift ..I try to set it up to slightly front focus . The older 35/1.4 M asph had this problem ..but set it for dead on at F2 and with a little thought you could avoid any real impact .

I don t generally shoot at F1.4 unless I need to or am seeking some very limited DOF . If I did of course the lens can be set for perfection wide open .

I expect to receive mine on Wednesday. I will let you know. Also remember that there has not yet been a camera body lens correction update for this lens.

Actually, I don't think I have seen a lens that didn't have some degree of focus shift, especially any lens over f/2.8
-bob

 

[Shac]

Roger - excuse my ignorance but how do you "set up" a manual focus lens to overcome focus shift?
Thanks in advance
David

 

[glenerrolrd]

Roger - excuse my ignorance but how do you "set up" a manual focus lens to overcome focus shift?
Thanks in advance
David

The 58/1.4 is an AF lens and can be "calibrated " thru the menu on the D800/D4 etc . With an M lens ...its done thru the RF coupling . But you are right if you have a manual lens (like the older 58/1.2 Ai Noct Nikkor) ...you will always be viewing wide open (and focusing in most cases ).

 

[Bob]

We always used to get to know our lenses and their quirks.
It became second nature where to focus depending on the lens and aperture.

The 58/1.4 is an AF lens and can be "calibrated " thru the menu on the D800/D4 etc . With an M lens ...its done thru the RF coupling . But you are right if you have a manual lens (like the older 58/1.2 Ai Noct Nikkor) ...you will always be viewing wide open (and focusing in most cases ).

 

[Shac]

Roger and Bob - thank you

 

[Hulyss Bowman]

I do not like this lens at all. This is my opinion but for that price and the samples posted around the web, so far, nothing impress me. The actual 50 f1.4 AFS-G do the same, the SIGMA do better. Bokeh wise, it is close, very. This lens do not have any magic like the Nikkor 50mm f/1.2. It is a cold technical lens.

Ok, you might have some color fringing with those lenses but you HAVE color fringing with this new "58" too (I spotted some with the Otus as well).

For video, I would not use it at all over the f1.2, period.

 

[ptomsu]

I think a 1.4 lens (or faster) is not for everybody, so who does not like a bit more care and work should not get these lenses.

On the other hand they will allow you some stunning photos if used the right way. From what I have seen of DL's samples I am happy and would say this is a winner. Maybe too expensive, but this is another story.

 

[jduncan]

I think a 1.4 lens (or faster) is not for everybody, so who does not like a bit more care and work should not get these lenses.

On the other hand they will allow you some stunning photos if used the right way. From what I have seen of DL's samples I am happy and would say this is a winner. Maybe too expensive, but this is another story.

Let's wait and see. I am not impressed at this moment, but web examples are difficult to gauge. it seems like colors could be especial. If it proves to be a technical feat I will be satisfied. It could be sharpness that match the D800E, or distortion or something similar.

Best regards,

J. Duncan

 

[glenerrolrd]

I do not like this lens at all. This is my opinion but for that price and the samples posted around the web, so far, nothing impress me. The actual 50 f1.4 AFS-G do the same, the SIGMA do better. Bokeh wise, it is close, very. This lens do not have any magic like the Nikkor 50mm f/1.2. It is a cold technical lens.

Ok, you might have some color fringing with those lenses but you HAVE color fringing with this new "58" too (I spotted some with the Otus as well).

For video, I would not use it at all over the f1.2, period.

I should know this week as my copy is scheduled to arrive on monday or tuesday . We have both the 50/1.4G and the 58/1.2 NOCT to compare . Diglloyd has done a writeup on the bokeh and his examples look excellent . This lens was designed to be an update of the NOCT 58/1.2 with special consideration given to the "character" .

I expect it will be similar to when Leica updated the Noctilux (1.0 to 0.95) they corrected some of the aberrations to improve overall image quality (which in turn changes the character of the lens ) .

 

[Guy Mancuso]

I actually like what I am seeing. I am going to get one just have to wait a bit.

 

[Jan Brittenson]

Not much in the way of focus shift that I can see. Maybe a small amount going from f/1.4 to f/2, but nothing dramatic.

I'd expect Nikon to update the L firmware for it.

 

[D&A]

Just a couple of brief thoughts. Roger is correct. Focus shift in a fast lens such as a f1.4 can be minimized by optimizing accurate focus at f2, thereby exhibiting very slight front focus at f1.4 and a very small amount if that when shooting at f2.8. The increasing depth of field at smaller apertures covers any focusing discrepancy.

As for this new 50 f1.4 G, I like what I see, but as it appears additional optical corrections were made for this lens to enhance certain performance characteristics, other optical attributes and anomalies that made the 58mm f1.2 so desirable inevitably had to change. I think careful consideration was given to this but it's impossible to have it both ways. The example of the older Leica 50mm f1 Noct vs. the current 0.5 Noct is one such example. There will be fans of one more than the other.

Having been impressed with the Sigma 50mm f1.4, especially at it's price point, it would be both interesting and fun to see some comparative images made with the Sigma vs. the new Nikon although I'm certain significant differences will be readily evident.

Dave (D&A)

 

[johnnygoesdigital]

Imo, trying to overcome a discrepancy by stopping down and increasing DOF, negates the purpose of a 1.4 lens The 58mm is not a very interesting FL to me either. Sigma has really surprised me with the 35mm 1.4, so their 50mm should be interesting. Add the USB dock and Sigma has the nod...so to speak.

 

[Jan Brittenson]

The shift is small enough that the regions of critical focus on the D800E overlap at f/1.4 and f/2, so it can be tuned for consistent critical focus at any aperture. However, with a new lens database this should be a non-issue since the camera will be able to adjust for it automatically. There's no way on earth to get critical focus at f/1.4 or even f/2 with this lens at 3-5'. At f/1.4 it's about 1/4", which means if you or the subject exhales they go out of focus. The only way to manage it is with AF-C locked onto an eye to track these fractional distance changes. (But then again, f/1.4 isn't terribly useful at such short distances so the discussion is academical.) Since for the shift to be relevant you need the consistent focus of AF-C, the AF system can apply the L firmware data to adjust for it, making it disappear completely.

 

[johnnygoesdigital]

I'd like to see how color fringing behaves at 1.4. My 85mm/1.4 is loaded with it wide open, but I try to not shoot towards bright, high contrast areas.

 

[Jan Brittenson]

I'd like to see how color fringing behaves at 1.4.

Didn't try it at 1.4, but this was shot at f/5.6. No profile corrections applied, or lens fixes as the LR doesn't yet have any for this lens. The lens is very well behaved - usually spherochromatism shows up in images like this (because the sky isn't in good focus and so is prone to bleed).

http://www.rockgarden.net/download/nik58/D8E_4726-full.jpg

D800E, ISO 100, handheld, f/5.6 1/1000s.

One thing clear from this is that it easily pushes the D800E to aliasing in high-contrast details. I noticed this in some other images as well.

This is one freaking sharp lens.

 

[Swissblad]

Impressive, thanks for sharing Jan.

 

[Ben Rubinstein]

Focus shift is a very real issue if you shoot both wide open and stopped down. As a wedding shooter who varies from wide open for more arty stuff and f2.8 for almost everything else, the focus shift on my 50L drives me crazy.

 

[Jan Brittenson]

There is 'focus shift', but not enough to be a problem. It mainly manifests itself as DoF building disproportionately behind the plane. Since you don't really lose critical focus on even the D800E (although you might on future cameras) it's not really accurate to characterize it as a focus shift.

Here's a demonstration. Shot from 8 feet, which is basically the furthest distance where the center AF point comfortably fits inside the left side of the target - the focusing part. It's probably also a reasonable working distance for this lens. It was aimed with live view, then racked out and refocused using PDAF AF-C (since that's what I use). It's AF tuned for the 0 line.

No lens profile was used for this. The settings for the first shot was copied to the others, so they use exactly the same processing. Even though it still has some glowiness to it at f/1.7 it still manages to drive the D800E to aliasing. Which nicely punctures the myth that a high-resolving lens needs to be high contrast or highly corrected for every aberration.

Note the spherochromatism. (No lens profile or correction for it was applied.)

f/1.4

f/1.7

f/2

f/2.8

f/4