Skip to content

Magic Lantern Dual ISO vs RAW Test Video (EOS 5D Mark III)

Yesterday I reported about the latest achievement of the Magic Lantern team: a hack that allows to obtain 14 stops of dynamic range (3 more than usual), using a a sampling method dubbed “Dual ISO”. Dual ISO means that half of the sensor lines are sampled at ISO 100, the other half at ISO 1600.

Michael Anthony (source: planet5d) produced a test video that compares RAW output and Magic Lantern Dual ISO output on a Canon EOS 5D Mark III. Description and setup after the break.

The setup:

A test of the new Magic Lantern feature “Dual ISO” in raw video mode. magiclantern.fm/forum/index.php?topic=7139.0

This is not a scientific test but just rather something to give people an idea of the quality improvements to Raw video using this new feature.

UPDATE: Just added an additional comparison of Dual ISO footage vs Normal Raw “Denoised” vimeo.com/70511941

Testing methods used…

Camera: 5D Mark III (shot 24fps at 1080p)
Dual ISO was set to use 100 and 1600 ISO.
Normal Raw ISO was set to 100.
Footage was shot in a dimly lit room with both a tungsten lamp and dark areas in the frame.
Shot handheld to reveal any moire or aliasing which may increase using Dual ISO feature.
Raw was converted to DNGs using the newest rawtodng.exe.
Postprocessing via Adobe Camera Raw using identical settings for both Normal Raw and Dual ISO shots…
Exposure was brought up a bit to +2.8
Highlights -100,
Shadow +70,
Sharpening 0,
Noise Reduction 0,
All other settings were at 0 or default values
Exported via After Effects to intermediate AVI files (Cineform 444 Filmscan 1)
Edited in Premiere Pro CS6 and rendered to H.264 1080p (10Mbps).

General Observations

This test was designed to be a worst case scenario. Bright light in a dim room room including fabric textures to help reveal moire. The reduction of shadow noise is dramatic to say the least. Without Dual ISO the footage would have been unusable. But it does come at the cost of increased moire/aliasing, something which Alex from Magic Lantern clearly indicates in his technical documents describing the new feature. Clearly, many scenes won’t exhibit the level of moire/aliasing shown in this test, so one just has to be mindful of the content of your scene and weigh that against the need to retain both highlight and shadow information when the dynamic range of the scene if greater than 12 stops. But when you are unable to properly light a scene to even out the exposure, the Dual ISO feature allows you to bring up the shadows in post significantly with very little increase in noise. Regarding any minor resolution decreases that are supposed to accompany this feature, I think that is a moot point… the footage wouldn’t even be usable in my test scenario without the Dual ISO feature (short of lighting the scene properly). I’m sure in many typical usage scenarios, the benefits will far outweigh any disadvantages.

Photobucket
Back To Top

Notice

This website or its third-party tools use cookies, which are necessary to its functioning and required to achieve the purposes illustrated in the cookie policy. If you want to know more or withdraw your consent to all or some of the cookies, please refer to the cookie policy.
By closing this banner you agree to the use of cookies.