Lately I've been shooting some higher iso's with my 7D's . I'm wondering if those who use this camera body ,use the in camera noise reduction CFn:II-2, or an alternative NR PS plugin for higher iso's?
Thanks for any opinions on this.
Dennis
Lately I've been shooting some higher iso's with my 7D's . I'm wondering if those who use this camera body ,use the in camera noise reduction CFn:II-2, or an alternative NR PS plugin for higher iso's?
Thanks for any opinions on this.
Dennis
I suspect that most folks run NR in Photoshop on almost all of their images, applying it selectively to the BG. That way you can be a little more aggressive with the NR. Neat Image is my current favorite (because it runs in 64-bit mode on the Mac, whereas Noise Ninja is limited to 32-bit mode).
Upcoming Workshops: Bosque del Apache 2019, Ecuador 2020 (details coming soon)
Website - Facebook - 500px
Dennis,
On most cameras any noise reduction is not applied to the raw image, but is to the jpeg. Some newer models (but not Canon DSLRs so far) seem to be modifying the raw data (not good).
All algorithms produce artifacts, so I prefer control of settings on a per image bases so I can limit artifacts. They only way to do that is in post processing. I alwasy leave all noise reduction off on all me DSLRs.
Roger
Thanks for the replys :) that helps.
Dennis
I will do it in post-processing where I can control the affect, degree of strength and make decisions on loss of detail.
I don't know Canon, but in Nikon there are two different noise reductions in the camera -- long exposure, and regular (which is really a high ISO noise reduction).
The regular noise reduction is for JPG only, not RAW, and hopefully anyone worred about noise is already shooting raw.
But the long exposure noise reduction is a whole different thing -- it is a dark frame subtract, and does affect the raw image, and is definitely worth doing in the camera. If Canon has one, and you shoot long exposures (more than a few seconds, the longer the more relevant) it is likely the same.
In a dark frame subtract, a separate exposure is made without opening the shutter, in a similar time frame (and because done at the same time, at the same temperature, etc.). This captures the thermal noise pattern of the sensor, and then this is subtracted from the actual raw exposure, and your raw exposure is the result. Note it means you wait a LONG time between shots for long exposures (Nikon cheats a bit and it is maybe half the exposure length).
If you want to see the impact of this, take a lens-cover-firmly-on long exposure, and then up the levels in Photoshop (or whatever) and you will see the pattern -- often with hot spots on one side or corner. Doing the subtract pretty effectively removes this noise pattern, as it is fairly repeatable.
But again -- only relevant for long exposures, NOT for high ISO regular noise.
Linwood,
Here again post processing would be better. When you take a long exposure, you can make several long exposures, then put the lens cap on and make another, called a dark frame. You can also take several dark frames and average them. Then subtract the average dark frame from all your long exposure images. You can even use dark frames from different days or even months, just that they need to be at the same temperature.
Dark frames also work for high ISO short exposures too. Many cameras have residual pattern noise that becomes apparent at high ISO, even in short exposures. Measure multiple darks and average them to reduce the noise and reveal the low level fixed pattern noise. Again, you can use darks from a different day, week, or month, as long as the temperature is within a couple of degrees of your subject images.
There is also a bias, so one can measure the bias (measured at the fastest exposure time, say 1/4000 second), and the if you made a high ISO image at one exposure time, you can scale dark frames measurements to different exposure times by first subtracting a bias frame.
Once you have bias and dark frames made, it can save you time in the field for real imaging. And because the dark frame is an everage of many dark frames, the resulting image is lower noise than one would get from in-camera dark frame subtraction. So you save both time and get lower noise by doing it in post processing.
Some examples:
http://www.clarkvision.com/articles/...t.photography/
Roger
Dennis,
What was already said about raw versus JPEG is a huge point, I'm pretty sure that none of the higher end Canon or Nikon bodies modify the raw capture - this includes the 7D.
Even if you're shooting JPEG, my take is that you should only use in-camera noise reduction if for some reason you are forced to make final use of the JPEG images straight from the camera. Hard to imagine that unless you're a PJ with tight deadlines.
There is no way that the algorithms and processing spent on in-camera noise reduction can come close to those used in post. The camera just doesn't have the software memory space or processor power. Not to mention that you can't tune the in-camera parameters for each image, or mask where you want to use NR - remember NR will generally reduce detail.
Alan