The 5D3 has banding issues. (pattern noise)

I know that some people have had banding problems with the 5DII, but I've never been able to get any - I feel left out and it's upsetting.

I've tried all sort of things - sunsets - no banding. What am I doing wrong?:o

G

Glenn NK wrote in post #14076278
I know that some people have had banding problems with the 5DII, but I've never been able to get any - I feel left out and it's upsetting.

I've tried all sort of things - sunsets - no banding. What am I doing wrong?:o

G

Gotta put your pixel peepers on

pixelbasher wrote in post #14075058
As above, buy the camera that you need then.... If another manufacturer has a better camera for what you do, then buy it.... If people like you sit here and whinge about how your camera (or worse, the next one that hasn't been released yet) doesn't do blah blah blah, yet you still continue to use it, where's the incentive for canon to do anything about it?? Buy the camera YOU need and be done with it.

But the much, much bigger question is: What is it to you?

Why do you care? Why do you see a need to post? You don't have a camera issue, but you for some strange reason have an issue with people commenting on the pattern noise...

Next thing: What incentive do Canon have to do anything about the issue if people say "it isn't a problem"? Why would a car manufacturer make something about a car engine that regularly runs on only three cylinders if people who comment about it just gets comments from other users: "But I still get where I should on three cylinders"...

Exactly what is your focus? What is your goal? What do you bring to the camera world with your oppinion that people shouldn't discuss the banding issues?

So you mean that if it would cost a lot of money to switch to another camera, people should ignore that, keep quiet and get the hell away from the 5D2? Just because you have an issue with people noting issues with the camera? That is a very clever view. Real clever.

The thing you totally miss here is that you sit and whinge. Just about a different issue. While not realizing it. If it really is a problem to you: Why not switch to a forum where you do not see anyone complain about Canon cameras? Why not switch to Nikon or another brand if you do have an issue with people commenting flaws in Canon cameras? Oops - that would be exactly the same broken logic you bring to this debate...

So how about commenting on cameras or photos and not spending so much time commenting on the users of cameras? And if you don't have a comment about the cameras, just skip to a thread where you feel you do have something of value to say?

Taken a quick look what subforum you are in? Is it a subforum "Canon owners and their lack of exposure skills"? Or is it a subforum about Canon cameras? I do feel you, and a number of other visitors, are a bit confused. Thinking it's the camera owners that should be reviewed and debated...

pwm2 wrote in post #14076574
Just because you have an issue with people noting issues with the camera? That is a very clever view. Real clever.

Typical fanboy behaviour. Nothing new there.

samsen wrote in post #14076206
Banding is a joke.
Has anyone considered that Mk III has a serious issue of breaking into pieces, if you through down from top of Empire State building? Now I call that a serious problem...

And how many hours have you spent with your 5D2 taking photos with high dynamic range before you got to your well-formulated view that banding is a joke?

Ricku wrote in post #14076616
Typical fanboy behaviour. Nothing new there.

That was a great one. Got a new definition of banboy? Fanboys normally talks up their favourite toy and ignores all their flaws.

But then the term fanboy is most often used by people who have a bit of a trouble finding real arguments. It's a standard brush-off when in a loss of words. Another popular use as a semi-rude comment when when intending to troll.

Next thing - I'm not obsessed about the banding issues since I haven't written a single comment about them for more than 2 years - but have used my camera. And in this thread I did start commenting by noting that the 5D3 is to new for us to truly know how the released versions will behave.

Some RAWs from a production 5DIII can be found here : http://www.kleptograph​y.com/dl/5diii/raw/ (not my camera .. from here : http://www.fredmiranda​.com/forum/topic/10955​20)

I have used the beta ACR to push shadows in the ISO 2000+ sample and not discovered banding. Lots of hidden detail but nothing else of note.

Is this just a 100% crop issue?

FotoDMick wrote in post #14102748
Is this just a 100% crop issue?

It's an issue for people who (needed) to push the shadow areas to the limit. Some seem very perturbed by it and will be moving to Nikon; others far less so and live with the camera as it is rather than strive to build mountains from the mole hills in the landscape.

Perhaps (or probably not) the 5d III files will be clean enough to satisfy. If they are not good enough, there are other manufacturers who will happily take people's money.

These production RAWs look very clean indeed - the ISO 1600+ files need very little work to be commercially viable.

bunyarra wrote in post #14102714
Some RAWs from a production 5DIII can be found here : http://www.kleptograph​y.com/dl/5diii/raw/ (not my camera .. from here : http://www.fredmiranda​.com/forum/topic/10955​20)

I have used the beta ACR to push shadows in the ISO 2000+ sample and not discovered banding. Lots of hidden detail but nothing else of note.

Did you look at the low ISO shots (e.g. 100, 200)? The banding complaints are typically in regards to the low ISO files.

Sample 07: The chroma noise makes the banding visible with just a +0.5 push of the image. Turn on color noise reduction to 25, and you can get up to +1.0 push before it's really noticeable in the tree trunks/leaves. Which makes me wonder what's the main color channel we're looking at there, because the black camera bag doesn't get noticeably pattern noisy until after a +2.5 exposure push. Since there are less red and blue pixels on a bayer CFA, those channels tend to suffer pattern noise more than the green channels.

I can do about a +2 push on sample 12 with 25 on the color noise reduction and find the dark background acceptible. Adding in some luminence noise reduction eliminates most of the banding look to the noise.

Poe wrote in post #14103682
Did you look at the low ISO shots (e.g. 100, 200)? The banding complaints are typically in regards to the low ISO files.

Which is because it's only at very low ISO that the sensor have captured a great amount of dynamic range. Making it sad to have the beautiful captured shadow details destroyed by pattern noise much stronger than the captured features.

At higher ISO, the noise level goes up so at high enough ISO, the pattern noise will no longer be several stops stronger than captured features.

Poe wrote in post #14103682
Did you look at the low ISO shots (e.g. 100, 200)? The banding complaints are typically in regards to the low ISO files.

Sample 07: The chroma noise makes the banding visible with just a +0.5 push of the image. Turn on color noise reduction to 25, and you can get up to +1.0 push before it's really noticeable in the tree trunks/leaves. Which makes me wonder what's the main color channel we're looking at there, because the black camera bag doesn't get noticeably pattern noisy until after a +2.5 exposure push. Since there are less red and blue pixels on a bayer CFA, those channels tend to suffer pattern noise more than the green channels.

I can do about a +2 push on sample 12 with 25 on the color noise reduction and find the dark background acceptible. Adding in some luminence noise reduction eliminates most of the banding look to the noise.

Just for reference, that image was shot as a reflection in the window - accounts for some of the strange horizontal bands that confused me for a bit

Pushing shadows and exposure, to +0.5, I can't see anything at all. Push it 2-3 stops and I do see noise and some patterns but nothing really significant enough that would not go when, as you say, you apply any noise reduction.

Genuinely interested to know what people consider unacceptable here when distressing pixels like this.

bunyarra wrote in post #14104192
Just for reference, that image was shot as a reflection in the window - accounts for some of the strange horizontal bands that confused me for a bit

Pushing shadows and exposure, to +0.5, I can't see anything at all. Push it 2-3 stops and I do see noise and some patterns but nothing really significant enough that would not go when, as you say, you apply any noise reduction.

Genuinely interested to know what people consider unacceptable here when distressing pixels like this.

I don't see any of the "banding" you're talking about due to this being a reflective surface.

bunyarra wrote in post #14104192
Genuinely interested to know what people consider unacceptable here when distressing pixels like this.

That's the million dollar question, none of us know what that agenda might be. I don't think they've discouraged a single person from the 5DIII, serious buyers of this class of camera know if you can't get the shot with the Canon, far more often than not you won't be able to get it with the competition.

Well, no. To eliminate shadow banding in the image, you simply let that shadow print to black. If you wanted detail in that shadow but have to print it to black to avoid banding, that means the image has less dynamic range than you desired.

It can't have "captured a great amount of dynamic range" and still have banding in the shadows--that's the definition of the low-end limit of dynamic range: The point that you have to print the shadows to black because of artifacts in the underexposed portions of the image defines the bottom end of dynamic range.