Lens reviews update: test data for the Nikon 58mm f/1.4G

Posted by Chris Roubis on 13th, Nov, 2013

DxOMark has recently reviewed Nikon’s AF-S Nikkor 58mm f/1.4G, a high-end (and very expensive) standard prime for full frame SLRs. As a taster for our upcoming review we’ve added the test data to our lens widget; as usual you can compare it to similar lenses, including the Nikon and Sigma 50mm F1.4s. Click through for more details and analysis, and a link to DxOMark’s own review of the Nikon 58mm F1.4.

Also this week, DxOMark has published its sensor review for the Sony A7R – a 36MP mirrorless camera with a full frame sensor – and a number of lenses including Panasonic’s latest 14-140mm superzoom for Micro Four Thirds. Click here for a round-up of DxOMark’s recent reviews, including the Sony A7R, and Panasonic Lumix G Vario 14-140mm F3.5-5.6 ASPH Power OIS.

Click here for our article ‘In Pictures – Nikon’s large and pricey AF-S 58mm F1.4G’

Nikon AF-S Nikkor 58mm f/1.4G lens test data

Here we’re showing DxOmark’s lens test data for the 58mm f/1.4 on both the full frame D800 and the DX format D7100, along with a quick summary of the main findings. We’re also showing a quick comparison to the existing AF-S Nikkor 50mm f/1.4G. After some real-world shooting with the 58mm, though, it’s become pretty clear to us that this doesn’t quite tell the full story. We’ll look at this in more depth in our upcoming full review.

Click on any of the images or links below to open our interactive lens widget, and explore the data further

1) Tested on Nikon D800

On the D800, sharpness isn’t especially high wide open, but this is entirely to be expected from a fast prime. Central sharpness increases rapidly on stopping down, but the edges lag behind significantly. This likely reflects curvature of field as much as anything else (these tests use a flat chart focused for the highest central sharpness). The edges continue to sharpen up at smaller apertures, and by F11 come close to matching the centre.

In all other respects the 58m performs extremely well. Lateral chromatic aberration is very low, and unlikely to be problematic in normal use, even without correction. Vignetting is unusually low for a fast prime: just 1.3 stops wide open, dropping to 0.7 stops at F2, and with a relatively gentle falloff in illumination into the corners (which should make it visually unobtrusive). There’s a little barrel distortion, but its simple profile means it should be easy to correct in software when necessary.

2) Tested on Nikon D7000

It’s very much the same story on the DX format D7000 as on full frame. Sharpness isn’t great wide open, but it improves dramatically on stopping down – by F4 the centre of the frame is as sharp as it’s going to get. The corners again lag behind, but sharpen up very well by F8. Chromatic aberration is pretty low, and unlikely to be anything to worry about in normal shooting. As usual for a full frame lens on DX, vignetting and distortion are very low indeed.

3) Compared to the AF-S Nikkor 50mm f/1.4G

The studio tests don’t give a clear edge to the 58mm f/1.4 over its much-less-expensive 50mm f/1.4 stablemate. Central sharpness is higher, but on the other hand the edges are softer. Chromatic aberration and vignetting are both lower on the 58mm, while distortion is about the same.

Overall, from these test results alone it’s not totally straightforward to see why the 58mm f/1.4 costs so much compared to the 50mm. But after shooting a little with the 58mm, it’s clear that they don’t tell the entire story about the lens. Stay tuned for our upcoming review to get a fuller picture into what it offers.

Full test results on DxOMark (and other recent reviews)

Our lens test data is produced in collaboration with DxOMark. Click the links below to read DxOMark’s own review of the Sigma 120-300mm F2.8 DG OS HSM, or see other recent reviews on the DxOMark website.

Comments

“Nikon invokes spirit of ‘Noct'” as the quote went.

Was the noct this bad?

EPIC FAIL IS EPIC !!

another 1700.00 plastic fantastic. It’s so special because nothing is in focus at f1.4

the Canon L 50mm has a very low rating on photozone.de, still by far my favorite standard lens when shooting Canon. The Nikkor 58mm will conquer the hearts of many, the only reason to get this type of lens is to shoot wide open or close to wide open. Lack of resolution isn’t the problem, but CA, astigmatism and especially coma can ruin a shot. Time for MTF reviewers to update their methodology, center sharpness at f11 for this type of lens is as meaningful as the achievable top speed of a sports car in reverse. Price, about $500 too expensive

Minolta Rokkor 58mm f/1.4 @ Ebay: 49 usd.
Adapter: 10 usd
Rokkor on my NEX: priceless.

On one hand, people want “lovely soft portrait lens”. On the other hand, people want 36MP pixel count. Why using 9 pixels to define a soft circle when you can using 4 pixel to render a sharp circle?

It is easy to soften with software or out-focus, but no way to sharpen without damaging the photo.

so Nikon , are we paying 1700 for this special soft vignette look ?

and i love how unsharp and buttery smooth the edges are which does give your photos a different look than almost every other nikon lens below 500$.

thanks a bunch for making that cheapo 50 1.8 such a great lens in comparisson ,and for that price.

is there a Xmas promo pack coming? 58mm1.4+ D600 for 1500$? killing 2 duds with a stone !! now that will be something ;-)))

No comment at all about the lens but I do admire the sarcasm. You must have worked for ages on that.

yup, finished 2 cups of coffee, spent an hour under the shower cooling down,,,and now am having an orga$m 😉

Too much information I’m afraid

Ken, I’m surprised you can still type.

Amazing ,isnt it ?;)

I don’t understand this lens at all.

At that price it should be fully corrected with beautiful bokeh.

I get that with the Fuji 35/1.4. (Granted it isn”t FF, but it’s 50mm equiv in APS-C).

i bet most people at nikon dosent understand the new product releases either ;-))

if there is to be a new product which will bring nikon some money ,then it should be packed in a box which says D400, but no,that´ll be listening to it´s customer base..

looks like at nikon , their new motto is, ” i am Nikon , and i am stubborn”

I am Nikon, I am Expensive
?

another mediocre nikon release.. oh oh.. what will nikon shooter do with all the unsharp D800 MP…?

Ha…Nikon fooled us….fork out more money for 2nd gen 58mm 1.4G sharp wide open.

Are you guys incapable of reading the English language? You don’t have to agree with the way DXO scores, but they did clearly state the reason they knocked the score down. That reason is: not being sharp wide open and not being sharp across the frame until f11. How do you people make it through life without a basic reading comprehension ability?

“On the downside, sharpness is low at full-aperture and isn’t uniform across the frame until its stopped to f11. At larger apertures there’s as much as 30% deviation in sharpness from the center to corners.”

“Individually the new Nikkor has the higher sharpness overall but the current 50mm is sharper at full aperture, while both are sharper than the AF-D model.

Stopped down to f2.8 and the new Nikkor is sharper in the centers than either but the outer field, edges and corners are similar to the old AF-D model and someway behind the current 50mm model until f11, where the new 58mm has high sharpness from corner-to-corner.”

I’m guessing a lot of people have a hard time reading the MTF-lines.

Guys – the left most point shows center resolution. The right one shows the extreme corners. Only by stopping down to f.8 does the MTF in corners reach 1500.

Not particularly impressive considering the price point.

I don’t think that the problem is reading the graphs, more of interpretation i.e. how do the graphs translate into practical use of the lens. I find DPR’s ‘real world’ samples far more revealing than these graphs.

f/11? LOL. Resolution of most modern bodies is limited by diffraction long before reaching f/11.

If you can afford the lens then great! Personally, I’ll stick with my trusty 50mm 1.4 and take one step forward or my 85mm 1.4 and take two steps back.

We need a new metric that is not accounted for called ‘rendering’. This is the same metric that enables us to say that this black and thin audio cable sounds a little ‘tinny’, but this more expensive red and white intertwined audio cable with a shiny nylon wrapping sounds ‘full’ and ‘expansive’.

But in truth there are some lens testing methodologies that I wonder can be supplemented, for example the center point should not *always* be considered the ‘reference’ plane of focus, but another set of measurements should be made at mid-frame as the ‘reference’ plane of focus. So tests should not only test center resolution and measure from there, but establish midframe reference point as well.

Indeed, it is well established that our brain “renders” objective data from our senses into subjective impressions that can be far from the truth.

So we need raw numbers from testing, and then come with solid interprations without applying our old nemesis “in my experience”. On top of that, as Roger Cicala has shown us, any meaningfull conclusion HAS to involve several copies of any lens to deal with sample variation. (something review sites choose to neglect for practical reasons, and another pitfal for people’s “own impression”) DxO Mark is on the right track, but they need more parameters that measure important characteristics of lenses. Your suggestion of using mid-frame focussing is a nice idea. Moreover, they should figure out how to put into objective numbers things like overall and local contrast, colour “rendering” (sorry for using that awfull word), bokeh… Their magical final scores are very incomplete, hence almost useless. DPReview’s final scores are not much better…

“We need a new metric that is not accounted for called ‘rendering’.”

Leica users have been making up such metrics for years.

If DxO Mark is correct then buying the 58 f1.4G is not worthy. The 85 f1.4 and even the f1.8 performs much better.

No real life photos? Charts are interesting, but I believe my own eyes more

and i bet your eyes are sharper across the frame than this 58mm softy ;-))

Well, the Sigma 50 “kills” it according to the Dx0 Comparison page.

Funny thing about Dx0 is that they have my Pentax DA*55 looking really bad and yet I know for a fact it’s better than my Sigma 50 AND Lenstip MTF’s say it wipes the floor with the Nikon 50 1.8 & 1.4G lenses too. The disparity between results from different sites is confusing to say the least.

DA*55 MTF
http://www.flickr.com/photos/steve-griffin/8114204872/

50’s on Nikon
http://www.flickr.com/photos/steve-griffin/8114187505/

I just shot an important job where I used both the 85/1.4G and the 58/1.4G. They perform pretty much equal in terms of sharpness. Both are really impressive. Maybe DXO had a not so good copy of the 58 because they rated it much lower than the 85. I’m surprised how they rate it so low. Definitely not what I see in my shots.
There are a lot of qualities and aspects that make a lens worth it’s money. People make the mistake to see these reviews as the ultimate answer. They are missing so much.
Nevertheless, most people will be fine with the 50/1.4G considering the price.

Ok, did you use it @ f1.4 or stopped down for your job?

The samples and data suggests that performance wide open is not spectacular (compared to Zeiss Otus).

There’s no question that the lens performs very well @ f2.8, but then many other lenses with similar focal length do… like Micro-Nikkor 55/2.8 AI-S.

According to DxO, this lesn is 28. The Zeiss Planar Macro f2/50 is 30.
Amazing

Indeed:
Sharpness – Nikkor 18Mpix Zeiss 17Mpix
Transmission – Nikkor 1.7 Zeiss 2.3
Vignetting – Nikkor 1.3 Ziess 1.7
CA – Nikkor 2 Zeiss 14 (wow!)

And the winner is… Zeiss. Lol 🙂

So it basically is equal to the 50mm f1.4 G until f5.6 where it get a bit sharper in the middle. That’s TOTALLY worth $1250 premium.

Yeah, cause sharpness is the only characteristic that matters when it comes to comparing lenses, right? Forget about colour rendition and micro-contrast, forget about CA, coma, astigmatism, forget about bokeh characteristics…

But don’t assume that just because its expensive and not that fantastic at edge to edge sharpness that it must be a champ at micro-contrast, color, CA, coma, and astigmatism…because quite frankly the samples I’ve seen do not impress in the other characteristics either…but hey that’s just like my opinion man.

Love mine! Not for test charts though. Rendering is really really sweet.

Great lens！ Much better than my 50 1.8D
LOL

At 7x the cost, it better be.

Why release such an expensive piece of glass, only to have such poor wide performance? Clearly its intended market is those who want wide open apertures.

It’s for portraiture with beautiful bokeh….you missed the point completely.

What flavor was the Kool-Aid, Dave?

What flavor is the hater-aid tkbslc? Dave stated exactly what the lens is- a portrait lens with a focus on bokeh, rendering, etc.

What flavour? Anyone with a basic understanding of lens design and knowing what results to look for. Sorry, you wouldn’t understand.

Which lens is sharper wide open? According to the P-Mix scores, this lens, along with the 60mm is the sharpest of the bunch. Or to put it in numbers:
58mm F/1.4 18
50mm F/1.4 G 16
50mm F/1.4 D 16
50mm F/1.8 G 16
Zeiss 50mm F/2 17
Zeiss 50mm F/1.4 13

What do you mean by “poor wide performance” – compared to what? Besides, there are other characteristics equally important – colour rendition, micro-contrast, flare resistance, bokeh characteristics, etc.

Dave, it’s a standard lens, it shouldn’t be only for portraiture…

hmmm, at 58mm it’s looking to me like a dx portrait lens (with fx abilities). And on the numbers, it seems to me that portraits is all it’s good for. And since the bokeh ain’t so good, obliteration bokeh is the only option, which is typical of a wide-open portrait.

There are a lot better options for portraits.

DXO lens tests are pretty much useless. I am not saying that they are doing it wrong but for a lens of this class, sharpness is not everything.

Agreed. For sharpness, all 50mm lenses are great so Nikon is obviously going for some optical properties that don’t register on the DxO Lens-O-Meter.

Bingo. Dxo doesn’t test things like rendering characteristics, bokeh, micro contrast, color, etc. All things that make a special lens… Special.
The last thing I’d want is a razor sharp portrait lens.

Somehow Zeiss Otus managess to get both smooth bokeh and high resolution @ f1.4. It seems to have much better wide aperture performance overall.

For the Nikkor 58mm results @ f1.4 are truly disappointing. Samples shot wide open exhibit also coma and lower contrast.

Stopped down to f2.8 center sharpness seems to be really high, but I’m not sure it’s good enough compared to 50/1.4G which renders fine pictures with nice bokeh.

Source Article from http://www.dpreview.com/news/2013/11/13/lens-reviews-update-test-data-for-the-nikon-58mm-f-1-4g