Before we fully delve into this topic, remember that each type of lens has its own unique conditions to sample and compare. We can prove this by looking at the most traditional way to test a lens. Historically, a user would place a camera/lens in a fixed spot and aim it at a stationary object on a wall, such as an eye pattern chart like you’d find in an optometrist’s office. However, this method brings with it inherent flaws, since each lens type is different. For example, ideal test conditions for a wide-angle lens are not the best for a long lens. The same is true for a box lens or ENG lens, especially with respect to focal length.

 

The ideal method for lens resolution testing encompasses utilizing a 16 x 9 multiburst test chart to establish the modulation transfer function (MTF) of the lens and camera system. In essence, the MTF is the contrast at a given spatial frequency (f) relative to contrast at low frequencies.  As photographer Norman Koren says, “MTF is a function of the distance from the image center, the aperture (f-stop), the spectrum of the light, the focal length (zoom), and even the focusing distance.”

 

Prior to performing these tests, it’s important to make sure the test charts are in absolute alignment - both horizontally and vertically - in order to establish the MTF profile of a lens. The first assessment should be to the luminance (Y) video signal at the picture’s center, where MTF will be the highest. Next, adjust the gain of a connected waveform monitor to place the 50 TV lines per picture height (TVL/ph) in alignment with the 100 IRE scale. This becomes the 100% spatial contrast reference.

 

Repeat the process for each of the four additional spatial frequency bursts placed in the center of the image. Data can then be compiled to account for the 200 TVL/ph to 800 TVL/ph scale and relative curves plotted accordingly.

 

The classification of color reproduction can be broken into four elements; spectral response of an HD lens, spectral response of the camera imagers, beam splitting optics in the camera and linear matrix which vary between camera manufacturers.  If the camera variable is removed, only the spectral response of the lens plays into effect.

 

It is important that a lens/camera meet the Society of Motion Picture and Television Engineers (SMPTE) 274M & 296M standards for colorimetry, which define the two parallel data channels of luminance and chrominance. Once the settings of a camera are fixed at the optimal position, you will notice subtle variations in the colors. This is a combination of the camera manufacturer and the optics of the lens, resulting in subtle tolerances in the spectral response of three factors; lenses, prisms and image sensors.

 

After choosing a camera, every lens should be tested concurrently using the same camera while establishing a proper white balance for setup. This will equalize any differences in RGB transmittance. Only after this is performed will any measurements of color become valid.

 

Testing should be performed in a well lit studio setting with color correction zeroed out and the camera’s linear matrix switched in. As color is a very subjective value, it is important to correctly balance color with a proper gray scale chart before taking any measurements.

 

Additionally, it can be a crucial step to test a camera lens setup in real life scenarios. These settings take into account a wider spectrum of colors, including highly color saturated objects, medium saturated objects and pastels. Even specific props and set materials can be used by a production team to assure the colors on set will look as authentic as possible. It can be very helpful to a production team to test a camera/lens setup on a calibrated monitor.

 

Because of the subjective nature of color, it is highly recommended that when choosing a camera/lens system for a production, that key members of the production team be involved in the selection process. 

 

Don’t forget to come back next week for our third installment of comparing HD lenses. You can also learn more by visiting www.fujinon.com.

Selecting a lens can be a tedious undertaking, with so many different specs and features to compare. Excluding price (a big consideration due to the cost of HD lenses), there are a few essential factors when purchasing a lens: sensitivity, contrast, resolution, color reproduction and geometrical distortion. Given how every single imperfection can be seen in high def, taking all these issues into account is critical to acquiring the best images possible.

 

Since these factors can get quite complicated, we’ll break them up into three posts. Today, we will highlight sensitivity and geometrical distortion. Subsequent posts will examine resolution, color reproduction, and contrast.

 

Lens sensitivity is critical because it creates a quieter picture with improved depth of field. A lens’ F rating only tells a user about the size of the iris opening, not how much light hits the sensor. Instead of focusing solely on the F rating, be sure to look at the T rating, which highlights the image transmission. T stop is essentially the F stop combined with the lens transmission losses. You can lose up to 1/3 stop due to transmission loss, so this is a much more accurate representation of a lens’ true characteristics.

When comparing sensitivity between two or more lenses, it is important to use a camera with fixed level lighting and fixed camera gain. Using those settings, record a video of a gray scale test chart with the F stop opened as wide as possible. This will determine the maximum relative aperture performance of each lens.

 

 

 

 

The second factor we’ll look at in today’s post is geometrical distortion. To some degree, this occurs in every lens, although today’s lenses do a much better job reducing this distortion than their predecessors from five years back. Due to the internal mechanics and dynamics of a lens, the wider the lens angle, the more difficult it is to reduce geometrical distortion, such as “pincushion” and “barrel” effects.

 

The best way to determine a lens’ propensity for geometrical distortion is to use the lens while shooting a studio set showcasing material with intersecting lines. Doors, windows, desks and even picture frames are perfect examples of content with orthogonality that will create geometrical distortion. A difference as small as 1% between lenses will have a big effect on the picture, even to a relatively untrained eye.

 

Check back here next week to learn about resolution and color reproduction. Or visit www.fujinon.com.