ClarkVision.com

Photograph the Moon with Different Lenses

by Roger N. Clark

All images, text and data on this site are copyrighted.
They may not be used except by written permission from Roger N. Clark.
All rights reserved.
BUT MOON IMAGES ON THIS PAGE MAY BE FREELY COPIED AND USED FOR NON-COMMERCIAL PURPOSES IN THE CONTEXT OF THE TEST ON THIS PAGE, MEANING YOU CAN COPY MANIPULATE AND POST ON A WEB SITE TO SHOW YOUR RESULTS IN COMPARISON TO THOSE SHOWN HERE.

Contents

Introduction
Part 1: The Full Moon Imaged with Various Lenses (in-camera jpegs)
Part 2: Processed Images from Raw Files
Filters versus no Filters
Fixed Focal Length versus Zoom Lenses

Introduction

This page shows results from photographing the Moon with different focal lengths on the same camera. The goal is to show the Moon and the image detail achievable with different lenses. The images on this page were all taken on the same night, November 1, 2009, with a Canon 5D Mark II 21 megapixel digital camera and various high quality lenses. Try imaging the full Moon with your lenses and camera and see if you can match or better these images. These are not intended to be the ultimate that can be achieved. The images are all out of camera jpegs with no sharpening. The only processing applied was a single curves adjustment to improve contrast. The curves adjustment was done in photoshop with input =155 set point on output = 90.

In the first section, each image presented is a 100% resolution crop with no resampling or sharpening done. All focal lengths are real focal lengths, not some equivalent.

In the second section, the raw data were processed to give increased detail. This section illustrates that detail can be extracted and enhanced due to the high signal-to-noise ratios that cameras with large pixels deliver.

All images were made from a carbon fiber tripod with mirror lockup, cable release, and image stabilization (IS) on. The carbon fiber tripod damps vibrations and image stabilization compensates for micro vibrations that impact image quality, especially at longer focal lengths.

Part 1: The Full Moon Imaged with Various Lenses (in-camera jpegs)

Figure 1. The Moon with a Canon 5D Mark II camera and a 70-200 mm f/4 IS L lens at 70 mm. Exposure: 1/800 second, f/4, ISO 200. Plate scale = 18.9 arc-seconds/pixel.

Figure 2. The Moon with a Canon 5D Mark II camera and a 70-200 mm f/4 IS L lens at 200 mm. Exposure: 1/800 second, f/4, ISO 200. Plate scale = 6.6 arc-seconds/pixel.

Figure 3. The Moon with a Canon 5D Mark II camera and a 300 mm f/2.8 IS L lens. Exposure: 1/2000 second, f/2.8, ISO 200. Plate scale = 4.4 arc-seconds/pixel.

Figure 4. The Moon with a Canon 5D Mark II camera and a 500 mm f/4 IS L lens. Exposure: 1/800 second, f/4, ISO 200. Plate scale = 2.6 arc-seconds/pixel.

Figure 5. The Moon with a Canon 5D Mark II camera and a 500 mm f/4 IS L lens plus Canon 1.4x teleconverter giving a focal length of 700 mm at f/5.6. Exposure: 1/400 second, f/5.6, ISO 200. Plate scale = 1.9 arc-seconds/pixel.

Figure 6. The Moon with a Canon 5D Mark II camera and a 500 mm f/4 IS L lens plus Canon 2x teleconverter giving a focal length of 1000 mm at f/8. Exposure: 1/200 second, f/8, ISO 200. Plate scale = 1.3 arc-seconds/pixel. At this focal length two factors are contributing to the softer image: 1) atmospheric turbulence (called seeing), and 2) optical quality of the 500 mm lens plus 2x teleconverter. The image is showing some color fringing around the edge of the Moon due to the high magnification of the lens and teleconverter but still shows substantially more detail than the 500 mm lens with the 1.4x teleconverter. Sharpening would greatly improve the image.

Part 2: Processed Images from Raw Files

Figure 7. The Moon with a Canon 5D Mark II camera and a 500 mm f/4 IS L lens plus Canon 1.4x teleconverter giving a focal length of 700 mm at f/5.6. Exposure: 1/400 second, f/5.6, ISO 200. Plate scale = 1.26 arc-seconds/pixel. The raw file from the camera was processed in Adobe Camera Raw (CS4) then increased in size by 1.5x with cubic spline interpolation. The image detail was sharpened using Richardson-Lucy image restoration employing a 5x5 Gaussian point spread function, 40 iterations and the edge of the lunar disk was done with the same restoration but 12 iterations. Richardson-Lucy image restoration is true sharpening. Compare this image with that in Figure 5.

Filters versus no Filters

People often ask if filters, even a UV or skylight filter can harm image quality. Figures 8a and 8b illustrate that the answer is yes, especially on telephoto lenses. The filter used in this test was a 77 mm Hoya HMC Super UV(0), a filter that is very high quality. But the large aperture 300 mm lens (at f/4 is a 75 mm aperture) uses the entire size of the filter and the long focal length magnifies the small departure from flatness. Such large filters, being so thin, are virtually impossible to keep flat. This same filter would produce effectively no degradation on a shorter focal length lens, due to both the smaller aperture and less focal length magnifying the optical imperfections. See Evaluating Filter Quality for more information on this subject

Figure 8a. The moon taken with the same lens on the same camera a couple of minutes apart. Three images with the filter on and 3 off were compared and all showed the same effect shown here. This illustrates even high quality filters can not deliver the optical quality needed on large aperture telephoto lenses. See figure 8B for these two images in an animated gif. These are in-camera generated jpegs with no sharpening done in post processing.

Figure 8b. The two images from Figure 8a in an animated gif showing the image degradation that happened by adding a high quality UV filter.

Fixed Focal Length versus Zoom Lenses

People often choose zoom lenses for their convenience without realizing many zoom lenses are a compromise in image quality. The Canon 100-400 mm L IS lens is very popular with wildlife photographers. There are some indications on the web that there may be some production problems with some copies being better than others. Whether or not that is the case, here are results from my 100-400 versus a 300 mm f/4 L IS lens. In both cases, the moon was imaged on the same tripod with the same camera a few minutes apart using mirror lock-up, cable release (TC-80N3) and IS was off. Exposure was 1/200 second at ISO 200, f/5.6. Neither lens had any filters on the front.

The 300 f/4 L IS with a Canon 1.4x TCII teleconverter produces sharper images than the 100-400 at 400 mm, Figure 10. The 300 f/4 also stands up well with a 2x teleconverter (Figure 11)

Figure 10. The Moon with a Canon 5D Mark II camera and Canon 300 mm f/4 L IS + 1.4x TC versus 100-400 mm at 400 mm. Both at f/5.6, taken a few minutes apart on the same tripod, with mirror lockup. These are in-camera generated jpegs with no sharpening. There were no UV or other filters on either lens. This comparison shows the that 300+1.4x teleconverter produces sharper images than this 100-400 at 400 mm.

Figure 11. The 300 mm f/4 L IS lens still stands up well with a 2X teleconverter. However, on consumer Canon bodies, autofocus is disabled. For this image, I focused manually using live view.

Figure 12. The 300 mm f/4 L IS lens versus a consumer zoom. Clearly, the fixed focal length lens is much sharper than the consumer zoom. Camera body was the same, a Canon 5D Mark II and the images were taken a few minutes apart. Five images at each setting were obtained with manual exposure, mirror lock-up, IS off, using a carbon fiber tripod. The best image for each setting was selected for this presentation, although there was little difference within each set. The images were converted from raw files using the same settings. Note the 300 f/4 L lens at f/4 is significantly sharper than the consumer zoom at f/8, a 2 stop advantage.

The lunar diameter in pixels is related to real focal length and pixel spacing only.

Lunar diameter in pixels = 1900 * focal_length_in_mm / (206265*pixels_spacing_in_microns/1000)

A 500 mm lens on a camera like the Canon 5D Mark II with 6.4 micron pixel spacing should have a lunar diameter of 720 pixels. That value is very close to the diameter of the Moon in Figure 4.

The 1900 is an approximate diameter of the Moon in arc-seconds on November 1 when these images were obtained. The 206265 factor is the number of arc-seconds in one radian.

Plate scale is the angular size of a pixel.

Plate scale in arc-seconds = 206265*(pixels_spacing_in_microns/1000)/focal_length_in_mm

if you know the distance to the subject and the plate scale, you can compute the resolution on the subject. For example, the Moon is about 240,000 miles away, so 1 arc-second is 240,000/206265 = 1.2 miles. Thus the 700 mm lens (Figure 5, above) has a subject scale of 1.2 miles/arc-sec*1.9 arc-sec/pixel = 2.3 miles per pixel. You can apply this concept to all photography. For example, what is the wing span on the bird you photograph? Knowing the distance to the subject, the camera pixel spacing and the lens focal length, you can measure anything. (Note, the equations above apply to small angles only. For angles beyond a few degrees more complex trigonometry should be used.)

Home Galleries Articles Reviews Best Gear Science New About Contact

http://www.clarkvision.com/articles/moon-test1

First published November 2, 2009.
Last updated June 17, 2010