" />

CANON 5D MARK II - LONG EXPOSURE NOISE TESTS (PART 2)

[LINK TO PART 1]

Well, after taking a weekend trip to the Washington coast, rebuilding my computer Sunday night, and trying to recover some files from a dying microdrive (I only lost two) I am finally able to post part 2 of the long exposure noise tests!

Here are the basic specs from the last post:
- Ambient air temperature was about 70 degrees F
- ISO 100
- f8 for 15 minutes
- Processed with DPP 3.5
- DPP noise reduction and sharpening OFF (sharpening will exaggerate all noise and the type we’re looking at here is no exception)

In this test I’m looking at the difference in a 15 minute exposure when the in-camera long exposure noise reduction (LENR) is ON or OFF. The first crop of each section is with LENR
OFF and the second crop with LENR ON. My comments are at the bottom.

Again, keep in mind that there was no sharpening and no attempt to extract the most detail from the images so evaluate only on how the image holds together over the longer exposures.

Here is the full image - slightly different this time but it makes no difference for the tests.

Crop 1

15 minutes - Long exposure noise reduction OFF

15 minutes - Long exposure noise reduction ON

Crop 2

15 minutes - Long exposure noise reduction OFF

15 minutes - Long exposure noise reduction ON

Crop 3

15 minutes - Long exposure noise reduction OFF

15 minutes - Long exposure noise reduction ON

From these three crops it’s clear that besides the minor (and if you don’t think it’s minor look at the images below before jumping to conclusions) speckling and minor differences in exposure the images are virtually identical. In my opinion these results are amazing.

Below are three other variations of how one might judge the noise in an image.

Extras

15 minutes - Long exposure noise reduction OFF - Bicubic Smooter to 50% size

15 minutes - Long exposure noise reduction ON - Bicubic Smooter to 50% size

15 minutes - Long exposure noise reduction OFF - Screenshot of “View -> Print Size” in Photoshop CS3

It is clear from the additional views including both the down-sampled image and the screenshot of the ‘Print Size’ view that the tiny speckles are only visible if the viewer has their eye against the monitor and really knows what they’re looking for. Needless to say the results are very impressive and have far surpassed by expectations.

More soon…but probably not these “studio” type tests, it’s a little boring…

Brian

Leave a Reply


Comment moderation is in use. Please do not submit your comment twice -- it will appear shortly.