Reading mode: Light Dark

Posterization with Mark III?

Go to last post
I can see it - although it does take a brighter monitor than many folks are using - i think it would be noticable in a large print.
Hawkman said:
Maybe I'm blind, ot not looking at the right areass. but those
rafters appear uniformly low to me, and are at very low, but
uniform values of 3-8 from the info pallete.

Gene

--
Gene (aka hawkman) - Walk softly and carry a big lens

Please visit my wildlife galleries at:
http://www.pbase.com/gaocus
http://hawkman.smugmug.com/gallery/1414279

Click to expand...
--
Currenlty shooting w/Nikon gear
 
wupeide said:
Wow - not i can see that this forum is just as bad as the Nikon
forum if not worse!

What is wrong with you people!!??

This man has asked a completely fair question with samples.

And I am so sick of the same nonsense argument regarding exposure.
I mean do any of you actually use your cameras? Have you ever
tried to make a photograph outside the studio? Again I say - what
is wrong with you people?

I had to deal with the same nonsense when I criticized the D70's
horrible noise problems. I just can't understand what is so
difficult about realizing that when you take a photo - much of the
time there will be shadow areas - i.e., areas that are under
exposed - these are important elements of any photo - they need to
look good - especially from a pro camera.

OK enough ranting - but seriously you people need to stop
mindlessly attacking people for reasonable questions!!!!
Click to expand...
Hey, mine was a reasonable question, and so far nobody has answered. I don't have a mk3, but I have seen several posts showing either posterization or banding. I never see it unless I crank my contrast and brightness to 100%.

My images always get slammed, so I suspect I don't have my contrast at the WEB users norm. I just want to know.

I have stopped posting images until I get this figured out.

Ben
 
Unless I'm seeing something different from what the OP is indicating, with a properly calibrated monitor, I think the feature being described would not be visible or at most barely visible. If it's obvious, then the monitor is not set properly.

In other words, I agree with you.
--
http://www.pbase.com/victorengel/

 
Sure - thats true - its not so obvious on most monitors - but if its visible on some monitors that means the problem exists and it will come back to haunt you one day...
Victor Engel said:
Unless I'm seeing something different from what the OP is
indicating, with a properly calibrated monitor, I think the feature
being described would not be visible or at most barely visible. If
it's obvious, then the monitor is not set properly.

In other words, I agree with you.
--
http://www.pbase.com/victorengel/

Click to expand...
--
Currenlty shooting w/Nikon gear
 
Might be a compression issue with the mark III .jpg if you look at the specs avg file size for a full-res image on the MkIII in .jpg is 3.5MB for a 10MP image. Compare that to the 30D's 3.6MB file for an 8MP image ... something doesn't add up ... I ran it by my local Technical Rep for canon but haven't received a response.
 
wupeide said:
Sure - thats true - its not so obvious on most monitors - but if
its visible on some monitors that means the problem exists and it
will come back to haunt you one day...
Click to expand...
The banding displayed here is absolutely not a problem. It is an artifact of the JPEG format. JPEG stores data in only 8 bits per channel per pixel at best. The banding shown here shows bands that are 1 bit apart. There are no values in between. How about an analogy:

Let's represent a gradient across one line of this post from 0 to 9, representing the values with digits. Here is a sample line.

0000001111112222222223333334444444555556666777788889999

Note that you can see where the boundaries are between the levels. Given the format (single digits), I cannot eliminate the banding. We can disguise the banding by adding noise. Then we'd see something like this.

0001010101112212122323233434344454545556565676778898999

Now you can't tell where the boundaries are, but there is more noise.

The fact that the bands are visible simply demonstrates the low noise of the camera. Banding would be concerning only if it was something like this.

00000000000122222222222222444444444444445688888888899999

Here we don't have a smooth gradient. This is NOT the kind of banding you see in the sample, though. You can demonstrate this to yourself by using the eyedropper tool set to point sample and move your cursor around over the image. When the value changes, it changes by only 1 and fairly evenly.

--
http://www.pbase.com/victorengel/

 
Victor Engel said:
Unless I'm seeing something different from what the OP is
indicating, with a properly calibrated monitor, I think the feature
being described would not be visible or at most barely visible. If
it's obvious, then the monitor is not set properly.
Click to expand...
I agree, as I did need to raise the levels to see the issue. Given the fact the diplay properties are variable, however, and that editing may occur, a optimally-designed converter should add noise at tonal levels that need them, but no more than need to dither away posterization, when stepping down from the working bit depth to the final output.

--
John

 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top