MicrostockGroup Sponsors


Author Topic: Should I stop using CPL for stock?  (Read 6216 times)

0 Members and 1 Guest are viewing this topic.

« on: January 14, 2009, 14:54 »
0
I got lots of rejections of photos shot with circular polarizer. Inspectors claim that it was over filtered or had wrong WB.


« Reply #1 on: January 14, 2009, 15:03 »
0
Can't you tell from looking at the images? It should be obvious enough what the issue is if rejects are as common as you suggest.

There's no problem in using a polarizer for stock provided it is used correctly.

Xalanx

« Reply #2 on: January 14, 2009, 15:13 »
0
Rejections have nothing to do with polarizers. There's no such thing as "polarizer generated rejection".
Landscapes + blue sky / water / reflections means you need a polarizer.

« Reply #3 on: January 14, 2009, 15:57 »
0
Can't you tell from looking at the images? It should be obvious enough what the issue is if rejects are as common as you suggest.

There's no problem in using a polarizer for stock provided it is used correctly.


I think it's just wrong description for rejection. They should say "we do not need this type of images" or something like that.

First one rejected by SS:

Second one rejected by StockXpert:

« Reply #4 on: January 14, 2009, 16:14 »
0
That's not necessarily an issue with the polarizer although it will probably make it worse. It looks to me as if you're trying to shoot a white building in bright sunshine and the camera is exposing for the building leaving the sky underexposed.

A couple of years back such images would often be accepted but to be honest they rarely sold anyway __ the black unnatural-looking skies hugely reduce the commercial potential of the image.


« Reply #5 on: January 14, 2009, 17:45 »
0
Hm , I would first search for some landscapes on IS sorted by downloads before stating that unnatural skies are reducing image potential cause the results will
show just the opposite.




« Reply #6 on: January 14, 2009, 18:09 »
0
Hm , I would first search for some landscapes on IS sorted by downloads before stating that unnatural skies are reducing image potential cause the results will
show just the opposite.

Obviously it is dependent on the subject matter and the images in question aren't landscapes. From my own experience near-black skies appear to have a negative impact on sales of architectural images.

The OP asked a question and I've taken the time to advise on the issue from my experience. If you've made it your job here to nit-pick and argue about irrelevant detail then I won't bother in future.

WarrenPrice

« Reply #7 on: January 14, 2009, 19:02 »
0
Hm , I would first search for some landscapes on IS sorted by downloads before stating that unnatural skies are reducing image potential cause the results will
show just the opposite.

Obviously it is dependent on the subject matter and the images in question aren't landscapes. From my own experience near-black skies appear to have a negative impact on sales of architectural images.

The OP asked a question and I've taken the time to advise on the issue from my experience. If you've made it your job here to nit-pick and argue about irrelevant detail then I won't bother in future.

Maybe it was the tone of you "advice."  As soon as I read your response, I felt that the OP was being insulted.  Sounded like "aren't you smart enough to evaluate your own pictures?"

Much as this most recent post sounds as if you are being imposed upon to share your time and most valuable experience.   

It is easy to be misunderstood in the two dimensional format.  I'm sure there was no intent on your part to come across as a bit rude. 


« Reply #8 on: January 14, 2009, 19:48 »
0
melastmohican,

I saw the first image in 123RF.  I don't know if it's valid to make a judgement, as it seems to me that 123RF resaves our uploads with a higher degree of compression, but when zooming to aprox. 100% I can see some banding noise in the sky and reflections.  This can be already present in the original, but sometimes I also observe them being intensified with edition, so maybe this is what they are complaining about, not the polarization.

Regards,
Adelaide

« Reply #9 on: January 14, 2009, 20:07 »
0
melastmohican,

I saw the first image in 123RF.  I don't know if it's valid to make a judgement, as it seems to me that 123RF resaves our uploads with a higher degree of compression, but when zooming to aprox. 100% I can see some banding noise in the sky and reflections.  This can be already present in the original, but sometimes I also observe them being intensified with edition, so maybe this is what they are complaining about, not the polarization.

Regards,
Adelaide

If noise is a problem it should be be stated in rejection reason not WB or exposure. I guess when I am shooting with ISO 100 noise should not be an issue. Of course I can make mistake in post processing so I would like to get warning about it from reviewer.
« Last Edit: January 14, 2009, 20:10 by melastmohican »

« Reply #10 on: January 14, 2009, 20:37 »
0
I don't see a problem with WB, but "overfiltering" means, as I understand, problems with excessive edition. What I see is not noise as we normally see it, but banding, which can be present in the original, but not an issue, and then be made more noticeable after the edition process.

I found a better technical explanation here:
Quote
In 8-bit image formats, such as JPEG, each colour number is specified as being in the range from 0 to 255. If you apply something like a curve to such a pixel the curve might say "darken the colour by 43.41%. If the original colour happens to be, say, 133, then the new colour will be 133 x 43.41 = 57.7353. But only whole numbers are allowed so the new colour will be rounded to 58.
In a series of such operations the rounding errors can accumulate, resulting in what is called posterization or banding. This looks like streaks of colour instead of smooth transitions and is often seen in sky. Banding is simply the most obvious manifestation of the problem; accumulated rounding errors affect everything.


Regards,
Adelaide

« Reply #11 on: January 14, 2009, 22:04 »
0
What I see is not noise as we normally see it, but banding, which can be present in the original, but not an issue, and then be made more noticeable after the edition process.

That has always been my nightmare with dark blue skies, especially if you add a soft light gradient on it to make the top more dark or you do for instance a Gaussian blur on the sky of 30px up. I figured the banding was rounding errors but that can't be the whole story since you have it with 16bit images TIFF/BMP as well as with 8bit. Actually that dreaded banding thing was one of the reasons I went to raw and 16 bit, but it didn't work for me. Mystery...

I started to see this banding when I worked on my high contrast LCD monitor. It's less visible on CRT monitors. To speak in favor of IS reviewers, they've always spotted it, and the other ones didn't.

Mohican: on topic and at first sight, the shot has blown out highlights on the facade and that's why it might have been rejected, not because of the sky. That's very easy to correct. Do a second development of the raw with exposure minus 2, then put this image in a layer under the original and selectively erase there around 30%. One of the many advantages of raw!

Not that it matters much. I found out the hard way that shots of not well known historical architecture landmarks like churches don't sell well. The girl with headset at the contrary...  ::)
« Last Edit: January 14, 2009, 22:09 by FlemishDreams »

« Reply #12 on: January 16, 2009, 07:00 »
0
There are several problems connected do this "CPL issue" - its not necessarilly true that dark skies have negative impact on sales. Most great landscapes and architecture shots are done with CPL, same with flowing water etc. Nothing to do with "unnatural" look - those oversaturated and overcontrasted pics widely present at all microstock sites are "natural"? HDRs or IRs are "natural"? And those shadeless studio shots or swapped (sandwich) skies at many landscapes are "natural"? Dont think so...

The problem is that some editors simply hate CPL and so reject, Im suspicious some of those even dont know how CPL works and think you "overfiltered" that in postprocess. You must realize there are many folks amongst microstock reviewers often hired from contributors - many of those do only studio work or even illustartions and have NO idea about anything like HDR, landscape filters or real world print. Some of stated "artifacts, noise or banding" issues are also result of non-calibrated and/or cheap super-overcontrasted LCD monitors.

Another issue is your real shot - shooting white building and blue sky with spot metering on white results in seriously underexposed areas in the sky because its far beyond the capabilities of sensor dynamic range. That means there is real noise even at iso 100. If you push levels too much in postprocess or push too much on noise reduction it results in typical blocky artifacts and banding. Its not necessarilly always this way but its typical technical problem.


 

Related Topics

  Subject / Started by Replies Last post
20 Replies
9718 Views
Last post December 11, 2009, 15:25
by RacePhoto
34 Replies
17440 Views
Last post August 28, 2010, 16:05
by FD
2 Replies
6031 Views
Last post February 02, 2011, 07:36
by Niakris
4 Replies
4152 Views
Last post February 05, 2011, 13:36
by pancaketom
2 Replies
4145 Views
Last post September 04, 2014, 13:36
by scottbraut

Sponsors

Mega Bundle of 5,900+ Professional Lightroom Presets

Microstock Poll Results

Sponsors