Sigma 35mm f1.4 and Zeiss 35mm f1.4 Imatested

Started Apr 9, 2013 | Discussions
Miljenko Devcic
Contributing MemberPosts: 963
Like?
Sigma 35mm f1.4 and Zeiss 35mm f1.4 Imatested
Apr 9, 2013

There has being lots of discussion about new Sigma 35mm f1.4 lens since it's rollout in November 2012. How good it really is and is it worth $900 are frequent questions all over web forums. Those who surf long enough DPR forums still remember my lens tests from few years back where I used accurate, repetitive Imatest tool. Since my main customer, country's best photo magazine has gone to the dogs, I stopped doing lens test on regular basys.However, since I'm still a photographer and still do buy lenses from time to time, I can afford myself to do some lens tests just for fun.
This time I compared Sigma 35mm f1.4 DG HSM A1 to probably it's main competitor, $1800 Zeiss Distagon 35mm f1.4 ZF2. To stir things up little bit further, I added another (close to) 35 mm lenses: $200 Nikkor 35mm f1.8G DX and $280 Micro-Nikkor 40mm f2.8G DX. Not really in a same class but good contenders for those who believe one should not spend a fortune on lenses.
The outcome is one very simple chart but you should keep in mind it takes huge effort to run those tests in order to get really exact data. So far I have uploaded just center resolution chart but in couple of days I'll add corner res and CA charts.

-- hide signature --

Best regards, Miljenko
www.pbase.com/miljenko
www.fotomag.net
www.wikiphoto.net
This is my twin pbase account coegsistent with 'Miljenko'.

Miljenko Devcic
Contributing MemberPosts: 963
Like?
Test Methodology
In reply to Miljenko Devcic, Apr 9, 2013

A great thing happened in a couple of years which I predicted: many reviewers (including DPR stuff) accepted Imatest as their main tool for lens evaluation and testing. However, due to it's great flexibility, Imatest can be used in many different ways and everyone can chose his or her reference points and default settings. After 9 years of using it, I came to my own default settings which don't deviate much from others'. My main departure from what's being used elsewhere is the way of reading test pic raw files: I use Adobe photoshop with the latest version of ACR plugin (currently 7.4) set at 50% sharpening instead of ACR default 25%. This provides two goals: 1) Use of widely accepted photo editor and raw reader, 2) Just the right ammount of sharpening that avoids the use of any additional sharpening except for the lowest resolution lenses.

Nikon D7100 provided almost ideal testbed due to its high pixel density and lack of optical low-pass filter. Nyquist frequency is so high that the lens becomes a main source of resolution cutoff.
Both Nikkors were tested using D7100 autofocus while Zeiss and Sigma were adjusted manually.

SO, HERE IT IS (please select original size pic):

-- hide signature --

Best regards, Miljenko
www.pbase.com/miljenko
www.fotomag.net
www.wikiphoto.net
This is my twin pbase account coegsistent with 'Miljenko'.

Reply   Reply with quote   Complain
Miljenko Devcic
Contributing MemberPosts: 963
Like?
Verdict
In reply to Miljenko Devcic, Apr 9, 2013

As you can easily see, Sigma 35mm f1.4 does incredibly good wide open (from f2.0 to f5.6) while Zeiss does better when stopped down, say from f2.8 to f11. As long as I remember, I didn't test any lens so far that measures so good wide open as Sigma 35 mm. OTOH, Zeiss Distagon measured absolutely highest resolution recorded so far: 3904 LW/PH at f5.6 (see the pic below). Beside Sigma 35mm f1.4 there is another winner here, a tiny and dirt cheap Nikkor 35mm f1.8G. From f2.5 to f10 it resolves over 3000 lines in vertical direction, not much lower that two expensive counterparts. Micro Nikkor 40mm f2.8 might sound sort of dissapointment here, but you must keep in mind we are dealing with extremely sharp contenders here and Micro Nikkor does remarkably well here, flying high over 3000 LW/PH from f4 to f11 with ease.
The verdict here is: get Sigma 35mm f1.4 if you planned to spend $900 or so for a moderately wide fast lens. If studio work and limitless budget is your kind of game, get the Zeiss and if short on budget get Nikkor 35mm f1.8 and you won't regret it. And if you can't afford longer macro lenses and you're able to squeeze macro lighting within an inch of working distance (I'm not!), then get the Forty.

-- hide signature --

Best regards, Miljenko
www.pbase.com/miljenko
www.fotomag.net
www.wikiphoto.net
This is my twin pbase account coegsistent with 'Miljenko'.

Reply   Reply with quote   Complain
Astrophotographer 10
Senior MemberPosts: 4,729Gear list
Like?
Re: Verdict
In reply to Miljenko Devcic, Apr 9, 2013

Thanks very much for that. I have been considering both Sigma 35mm F1.4 and Zeiss so its timely.

Greg.

Reply   Reply with quote   Complain
clouseau2
Forum MemberPosts: 93Gear list
Like?
Re: Verdict
In reply to Miljenko Devcic, Apr 9, 2013

Wow great work . I really like this comparision . The Sig and the Zeiss are both high on my wishlist .

 clouseau2's gear list:clouseau2's gear list
Nikon Coolpix P7700 Nikon D200 Nikon D700 Nikon D800E Nikon AF-S Nikkor 16-35mm f/4G ED VR +20 more
Reply   Reply with quote   Complain
Miljenko Devcic
Contributing MemberPosts: 963
Like?
Re: Test Methodology
In reply to Miljenko Devcic, Apr 9, 2013

Please notice resolution irreguralities with Zeiss lens from f1.4 to f2.5. It is NOT a testing flaw or anything dealing with my procedures. Actualy, when those "hills and valleys" appeared on chart, I retested Zeiss lens taking no less than 10 separate measurements every 1/3 of a f/stop up to f5.6 but nothing changed. This is obviously something only Zeiss could explain...

-- hide signature --

Best regards, Miljenko
www.pbase.com/miljenko
www.fotomag.net
www.wikiphoto.net
This is my twin pbase account coegsistent with 'Miljenko'.

Reply   Reply with quote   Complain
Miljenko Devcic
Contributing MemberPosts: 963
Like?
Re: Sigma 35mm f1.4 and Zeiss 35mm f1.4 Imatested
In reply to Miljenko Devcic, Apr 9, 2013

In case somebody could get an impression I am some kind of Imatest evangelist promoting this software, let me explain: I hate it, I hate Mr. Koren's stuff who decided some years ago it should be targeted to manufacturers and professional labs only. With the price of $299 for Studio and $2200 for Master version it became simply to expensive for enthusiasts like myself who can do simple but useful tests everybody can understand and use. There should be kind of Lite version available for about $100 (the price I payed for the v2 licence some 8 years ago) with limited set of features I use in my tests.

-- hide signature --

Best regards, Miljenko
www.pbase.com/miljenko
www.fotomag.net
www.wikiphoto.net
This is my twin pbase account coegsistent with 'Miljenko'.

Reply   Reply with quote   Complain
goosel
Senior MemberPosts: 1,478Gear list
Like?
Did i understand?
In reply to Miljenko Devcic, Apr 9, 2013

.. that you only used results of  center sharpness in your verdict/graph?

btw: thanx for  the effort and sharing your results.

Reply   Reply with quote   Complain
Miljenko Devcic
Contributing MemberPosts: 963
Like?
Re: Did i understand?
In reply to goosel, Apr 9, 2013

Nope, It's based on all the parameters measured both in center and corners but until today not all Excel data have being converted into charts. Give me another day or two to get the whole job done. I was simply to excited to wait for another day to share the findings with you guys!

-- hide signature --

Best regards, Miljenko
www.pbase.com/miljenko
www.fotomag.net
www.wikiphoto.net
This is my twin pbase account coegsistent with 'Miljenko'.

Reply   Reply with quote   Complain
Alchin
Senior MemberPosts: 1,173
Like?
Re: Sigma 35mm f1.4 and Zeiss 35mm f1.4 Imatested
In reply to Miljenko Devcic, Apr 9, 2013

Yes I remember you and your interesting lens tests.  I don't frequent this forum much but by accident came across your post while looking in on the Nikon 85 G;  wondering if I should update my old 85/1.8D

Thanks for sharing your findings. Hope you are doing well.

Cheers,

AC.

Reply   Reply with quote   Complain
HSway
Senior MemberPosts: 2,742
Like?
you forgot
In reply to Miljenko Devcic, Apr 9, 2013

one interesting piece – Samyang 35/1.4. Unlike on Canon cameras the Samyang work well on Nikon.  Here it is at f8  next to the Zeiss. I found it a while ago. it’s from verybiglebowski on 5d ll, full res samples.

well behaving bokeh Samyang   Zeiss

But the Zeiss' bokeh will no doubt win in demanding scenes for bokeh-cream lovers (the Nikkor 35/1.4 seems a tad better also). But I like it for 35mm in this respect.

There would be more interest if the test is on fx I reckon. And I noticed and mentioned before a similar observation on another chart and close-up crops. The Sigma at small apertures vs large apertures difference when compared to Samyang 35/1.4 – both lenses showed similar preference for small vs large apertures regarding the resolution curve as your couple does. Ultimately, lenses can perform differently at longer distances.

Thanks for the work,

-- hide signature --
Reply   Reply with quote   Complain
RSchussel
Regular MemberPosts: 220Gear list
Like?
Re: you forgot
In reply to HSway, Apr 9, 2013

Thank you for the analyses.Any chance of doing the Nikon 35mm F1.4G

I think what is forgotten is that you should be buying a F1.4 lens to shoot wide open and/or low light.

Thats why the Sigma is such a wonderful lens. It looks sharp compared to most other F1.4 lens wide open. I dont usually extoll the virtues of a lens but if you need a fast 35mm lens Sigma is the way to go.

 RSchussel's gear list:RSchussel's gear list
Nikon D800 Nikon AF-S Nikkor 70-200mm f/2.8G ED VR Nikon AF-S Nikkor 14-24mm f/2.8G ED Nikon AF-S Nikkor 24-70mm f/2.8G ED Nikon AF-S Nikkor 24mm f/1.4G ED +4 more
Reply   Reply with quote   Complain
Reivilos
Regular MemberPosts: 262Gear list
Like?
Re: Sigma 35mm f1.4 and Zeiss 35mm f1.4 Imatested
In reply to Miljenko Devcic, Apr 9, 2013

Thanks for the tests and write-up.

 Reivilos's gear list:Reivilos's gear list
Nikon D300S Nikon AF-S DX Nikkor 16-85mm f/3.5-5.6G ED VR Nikon AF-S DX Nikkor 35mm f/1.8G Nikon AF-S DX Nikkor 10-24mm f/3-5-4.5G ED Nikon AF-S DX Nikkor 17-55mm f/2.8G ED-IF +6 more
Reply   Reply with quote   Complain
Miljenko Devcic
Contributing MemberPosts: 963
Like?
Re: you forgot
In reply to HSway, Apr 9, 2013

Started recently business cooperation with Samyang importer here so there is a chance I might be testing some of their lenses in the near future.

Besides, I have some other Zeiss lens at hand like 21 mm and 100 mm so I might be doing them too...

-- hide signature --

Best regards, Miljenko
www.pbase.com/miljenko
www.fotomag.net
www.wikiphoto.net
This is my twin pbase account coegsistent with 'Miljenko'.

Reply   Reply with quote   Complain
tom43
Forum MemberPosts: 62Gear list
Like?
Re: you forgot
In reply to Miljenko Devcic, Apr 10, 2013

www.Photozone.de has tested the Sigma, Nikon and Samyang 35/1.4 and Zeiss 35/2.0 lenses on Nikon FX D3X demonstrating the highest sharpness for the Sigma even at higher apertures (based on Imatest).They also documentated a weaker performance for the Sigma with DX sensors. Maybe we should be careful to transfer these findings with the D7100 to the full-frame field?

Miljenko, thanks for these data. Any attempst to proove performance with full-frame?

Regarding Zeiss: many people would be happy to have data on the newer models, e.g. the 15/2.8 and the 135/2.0 Apo.

 tom43's gear list:tom43's gear list
Nikon D40 Nikon D3100 Nikon D810 Nikon AF-S Nikkor 70-200mm f/2.8G ED VR II Nikon AF-S Nikkor 24mm f/1.4G ED +8 more
Reply   Reply with quote   Complain
Miljenko Devcic
Contributing MemberPosts: 963
Like?
Re: you forgot
In reply to tom43, Apr 10, 2013

We should always be careful with interpreting test data between the sensor formats. When comparing similar pixel count, APS-C size sensor packs photo sites at higher density and challenges the lens with higher resolution requirement. OTOH, full frame sensor's edge photo sites are approached at higher angle which can add to chromatic aberrations, fringing and vignetting.

I don't have a constant access to FF camera (D800e would be ideal) in order to repeat these tests on FF; someone else would have to do that

-- hide signature --

Best regards, Miljenko
www.pbase.com/miljenko
www.fotomag.net
www.wikiphoto.net
This is my twin pbase account coegsistent with 'Miljenko'.

Reply   Reply with quote   Complain
Dan_168
Veteran MemberPosts: 4,049
Like?
Re: Sigma 35mm f1.4 and Zeiss 35mm f1.4 Imatested
In reply to Miljenko Devcic, Apr 10, 2013

Thanks for the review, I also own quite a few 35mm including the new Sigma 35 and the Zeiss 35 1.4, from my experience of them on D3 and D800E, one thing I didn't see you mention and I want to bring it up is at F1.4, the Zeiss has very strong color fringing while the Sigma has a lot less, so I actually avoid shooting the Zeiss at 1.4 in some situation when I think this particular issue will show up in the picture, and with the Sigma I don't really have to worry about that, and I agree with you I still saw the Zeiss perform better than Sigma at my "landscape aperture"  like F8-F11 at the edge, the Zeiss perform so well there put my 35L to shame, so I am still keeping both for different applications.

Reply   Reply with quote   Complain
RodluvanII
Senior MemberPosts: 1,504Gear list
Like?
When is sharp simply 'sharp enough'?
In reply to Miljenko Devcic, Apr 10, 2013

Do I understand you correctly that these test measure centre sharpness? If not, where in the frame are they made?

Do you have any thoughts on some of the other many and very important aspects of the lenses?

 RodluvanII's gear list:RodluvanII's gear list
Nikon D3X Nikon AF-S Nikkor 24-70mm f/2.8G ED Nikon AF Nikkor 180mm f/2.8D ED-IF Carl Zeiss Biogon T* 2,8/28 ZM Carl Zeiss C Biogon T* 2,8/35 ZM +8 more
Reply   Reply with quote   Complain
HSway
Senior MemberPosts: 2,742
Like?
Re: you forgot
In reply to RSchussel, Apr 10, 2013

Absolutely, it shows in this test. In practice the lenses, as with anything, are used by their strength and properties. 35mil lens will always be judged and looked through the eye of a landscaper and for similar uses. By design, the Sigma looks like exemplary example of its kind imo, an ideal design realization especially in respect of resolution.

I have found the comparison after a google search yesterday on the canon forum here on the site. It was done by Verybiglebowski who comes also to this forum – he owns 14-24 I remember. It looks he is more a canon user though and that he owns canon 24/1.4 ll, I guess it’s unlikely he has the 35/1.4g. We have had no conversation.

-- hide signature --
Reply   Reply with quote   Complain
HSway
Senior MemberPosts: 2,742
Like?
Re: you forgot
In reply to tom43, Apr 10, 2013

tom43 wrote:

www.Photozone.de has tested the Sigma, Nikon and Samyang 35/1.4 and Zeiss 35/2.0 lenses on Nikon FX D3X demonstrating the highest sharpness for the Sigma even at higher apertures (based on Imatest).They also documentated a weaker performance for the Sigma with DX sensors. Maybe we should be careful to transfer these findings with the D7100 to the full-frame field?

Miljenko, thanks for these data. Any attempst to proove performance with full-frame?

Regarding Zeiss: many people would be happy to have data on the newer models, e.g. the 15/2.8 and the 135/2.0 Apo.

FWIW, photozone doesn’t show the curve I am mentioning in dx nor in the fx. Compared to their Samyang data I don't see any superior results except the large aperture, when you mention that.

Btw, one would expect dx showing steeper curve from the peak resolution down towards f11 and f16. I was looking at some comparisons some time ago and this showed far less on the charts. Then I realized it’s the peak sharpness that in fact is reduced compared to fx causing relatively flatter dx curve than I would think. It’s more apparent when adding concrete numbers into the perspective.

-- hide signature --
Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads