Crop Factor, Low Light and Aperture with m4/3 lenses? Part 2

Started Jun 16, 2014 | Discussions thread
Anders W Forum Pro • Posts: 21,468
Re: Reply to Just another Canon shooter

Just another Canon shooter wrote:

knickerhawk wrote:

Let me repeat that the St.Dev. cannot tell you anything at all about the character of the noise or even if it can be qualified as noise.

Nonsense. The Std. Dev., especially when broken out in LAB tells you a lot about the character of the noise for what is ideally a uniform patch with 0 deviation. We do not disagree that the number tells you nothing about the spatial distribution of the noise, but spatial distribution is only one element of noise.

You still do not get it. St. Dev. tells you absolutely nothing about the character (statistical distribution) of the noise. I can easily generate an absolutely noise free image (by human standards) that will have the same mean and St.Dev.

No you can't for a target patch assumed to be uniform with regard to color and luminance as is the case here. Without noise, the standard deviation would be zero.

Going back to noise, if you know the type of the noise, then yes, St.Dev. tells you about its strength.

Under the assumption that the target patch is uniform with regard to color and luminance, the standard deviations simply tell us the magnitude of the noise. And the standard deviations computed by knickerhawk shows that you were wrong in your original suspicion that ACR/LR tipped the scale in favor of the E-M1.

and my own two eyes confirm that the character of the noise (the look) is different between these two cameras and the two converters but that's hardly a surprise and certainly not evidence that Adobe is somehow inappropriately tipping the scales in favor of the Oly (and only the Oly) with respect to noise reduction. Your speculation on that point is meritless.

I did not say "inappropriately". And you did not answer the question, which was very simple. Does the lower right square look more NR'd than the rest or no? As simple as that.

I answered your question in detail. It's just that my answer wasn't the admission you were looking for. To repeat, there is less luminance variation in the LR Oly patch but more color variation relative to the DXO Oly patch. The spatial distribution and color "pattern" you keep referring to is very similar between the two patches. This conclusion is supported by the Std. Dev. numbers and examination of the individual channels for the two patches in LAB. Your characterization that LR "clearly...applies more NR (or whatever you want to call it to the color to the Oly than DXO" actually has things turned around. The character of the noise is more luminance influenced in the DXO rendering and more color influenced in the LR rendering. The opposite of what you claimed. Unfortunately, your refusal to utilize readily available objective information to enhance your initial subjective impression (which is apparently weighted strongly toward better discrimination of luminance differences than to color differences) has biased your observational skills.

You miss the point again.

No, you do. If you were right, the standard deviations wouldn't be those reported by knickerhawk.

The lower right crop has what is usually described as "blotchy" noise - areas with too much correlation between the pixels.

I asked you already at this point whether you had used ImageJ to check the power spectrum. You dodged the question.

You are fixated on the color values only.

knickerhawk did a thorough analysis of the color as well as luminance values.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +28 more
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow