Cooke, Zeiss, Lumatech, Nikon, Lomo and Sigma lenses compared
Intro
The border between still lenses and cinema glass is slowly fading. There is a growing amount of cinema versions of stills lenses using the same glass, partly due to the developments of mainstream large format cinema cameras.
We decided to test an array of lenses from different eras and manufacturers, including the popular 50mm Sigma ART prime and the renowned Cooke Speed Panchro.
Finally, we also included an old Russian anamorphic lens to see how the character compares to the spherical lenses.
We ended up testing the following list of lenses:
- Cooke Speed Panchro 50mm T2.2 Series II
- Opitca Elite Mk II 50mm T1.3
- Zeiss B Super Speed 50mm T1.4
- Lumatech Super 35 Illumina 50mm T1.3
- Lomo Square Front Anamorphic 50mm T2.5
- Nikon AI-S 50mm F1.4
- Sigma DG HSM Art 50mm F1.4
The test setup - Miles Lembke, Jesper Wahlqvist, Minco van der Weide
Differences
There are a few factors that differentiate cinema glass from stills lenses. The main difference often isn't the glass itself, but the housing and how the lenses perform during a shoot. In general cinema glass offers the following benefits over stills lenses:
- T-stops instead of F-stops for more reliable exposure
- Manual and de-clicked aperture
- PL-mount for solid locking
- Uniform front diameter and length
- Follow focus gears
One of the main problems when using non-modified stills lenses is that lens series often have different lengths, filter sizes and focus ring positions. This means that with every lens change both the follow focus and matte-box needs to be adjusted, which is quite impractical on a fast-pace filmset.
Changing the PL mount to an EF mount - Jesper Wahlqvist
The same consistency is found in the way the speed of the lens is measured. The T-stop rating measures how much light reaches the sensor, while F-stops are mathematically determined. They don't take light loss due to different coatings and internal reflections into account. this often results in a different T-stop with equally rated F-stop lenses.
In terms of optical performance the difference of cinema glass is smaller, generally the only differentiating factors are:
- Less focus breathing
- Parfocal (when using zooms)
However, most of these problems can be overcome when choosing the right stills lenses, using custom focus gears and unified lens fronts.
One subjective but definitely noticeable difference is the fact that cinema glass, especially older handmade ones, are often made with a certain look in mind. Most modern lenses are made using precise electronic machinery. You could say this generally speaking results in optically superior performance but a lack of character.
Putting focus marks on the Arri FF5 - Jesper Wahlqvist
The test setup
We decided to shoot an outdoor test recreating a real-world shoot. We have experience with indoor lens tests, but the difference between lenses becomes more apparent when there are more things happening in the background.
The test was shot in front of our office at a cloudy day. The only light modification is a black flag at the shadow side of my face to increate the contrast.
During the recording we performed a rack focus from infinity to a face shot, and then a focus to the colour- and focus chart at the end.
The test equipment
We decided to shoot the test on the Arri Alexa Mini, mainly because the ALEV-III is the most used sensor in high-end productions. The footage was recorded in ProRes 4444 with the LOG-C profile. Apart from that we used the following gear:
- Arri MMB-2
- Arri FF5
- Sachtler Video 18
- X-Rite ColorChecker Classic
- Carl Zeiss test chart for Cinematography Lenses
- SmallHD 502 bright
- Atomos Sumo19
To be able to directly compare the lenses we all used them at T2.8. To keep lighting consistent we used the Atomos Sumo19 to match the light coming traugh the sensor of all lenses with a vectorscope.
Correcting the footage in the grading basement - Miles Lembke
Results
The difference between lenses can be very subtile. During comparison you usually look for the following characteristics that together create the "look" of a lens:
- Bokeh character
- Sharpness
- Flaring
- Colour cast
- Chromatic- and spherical aberration
- Contrast and bleeding of highlights into shadows
During the grading we decided to only do luma adjustments to match the brightness and contrast of the thots and left the individual colours as they were. Since the LOMO lens has a 2x squeeze we cropped the desqueezed footage to match the 16:9 frame of the spherical lenses
We noticed that the biggest differentiating factor between the spherical lenses was the age of the glass. Most modern lenses looked very comparable in terms of colour and sharpness, while the older glass had a lot more character and colour difference.
The on-set experience of the lenses varied slightly. The biggest problem was the fact that we couldn't set focus marks on the Sigma lens, since the stills version doesn't have hard stops. This made focus pulling very challenging, this is visible in the focus rack of the Sigma lens. This problem doesn't apply in the cine version of this lens (Sigma CINE 50mm T1.5) that shares the same optical design.
Racking focus on the Lomo lens showed quite severe focus breathing, this is often seen in anamorphic lenses.
Verdict
The best lens is the one that suits the look and feel of the production the best, so there can't be a winner in a lens comparison. But there were definitely noticeable differences. We noticed that the Cooke had a green/yellow cast with quite a lot focus fall of towards the edges. while the Nikon lens tended to look quite magenta.
The first thing that's often looked at when comparing lenses is the sharpness, but we don't think it's the most important factor to look at for cinema use. For the average viewer, the variations in terms of colour and bleeding are much more noticeable than the difference in sharpness. Furthermore, a slightly soft look can often help create a more filmic look, hence the popular use of diffusion filters.
Even though the difference in sharpness can be seen wide open, the performance of lenses becomes near-identical when being stopped down to T5.6 or more.
At the end of the day we had some lenses we thought looked good, but it's best to look at the footage and decide for yourself what lens you like best for your projects.
A big thanks again to Falsett Rental in Stockholm for providing the equipment that made this comparison possible.
Download the files
Since YouTube compresses the footage quite severely we uploaded a 4k render. Click on the link below to download the files on Mediafire.
You are free to use, modify and publish it under a Creative Commons Attribution licence, make sure to mention Filmbron and provide a link to our website (www.filmkompaniet.com).
Let us know your thoughts on the differences in the comments below, and what lenses you would like to see tested!