Jump to content

Failed submissions - non-discrimination


Recommended Posts

18 minutes ago, MDM said:

 

are you shooting raw? I presume you are. If so, things have moved on a lot over the years and it is possible to do a lot of things now on the raw image that used to require working in Photoshop before. I should say I am still a bigtime Photoshop user but I do as much as I can on my raw images before I open them in Photoshop. It is generally preferable to work on the raw image especially if working on highlight and shadow detail. Also raw developers have improved massively - ACR and Lightroom use the same raw converter engine but there are a lot of features of Lightroom that are not available with an ACR workflow - really worth investigating I think. 

 

 

Yes , I use Rawtherapee to process my photos and I finish the last settings with photoshop before saving the file as jpg:  RAW-> png-> jpg. I am more comfortable with PS to get the rendering I want. Plus, I have some filters that I use regularly , but they have more interest for my microstock activities than for Alamy images. 

Link to comment
Share on other sites

I have never paid attention to black/white point. I watch the histogram and make adjustments until it and the image look good. Maybe back when I first started I tried looking at the numbers and I felt that was less successful than how I am presently doing it.  That includes images with limited tonal range.

it must be working, because I’ve never had a failure for that. CA, dust spots, soft, yes. 

Betty

Link to comment
Share on other sites

45 minutes ago, Betty LaRue said:

 I watch the histogram and make adjustments until it and the image look good.

 

Can you elaborate on "I watch the histogram" ? What does the histogram look like when it looks good (keeping Alamy's requirements in mind)?

 

My QC failures have always been for the same reasons as yours (except for CA, never had one of those).

Link to comment
Share on other sites

With my Fuji, I have the histogram showing as I look through the viewfinder. I like my images exposed to the right, just short of clipping or touching the right side. If I see clipping, then I dial exposure comp down or change another setting like increasing shutter speed.

 

That gives me a well-exposed image without blowing the highlights. Especially important if I want any semblance of blue skies.

 

Fuji also has WYSIWYG. What you see is what you get. Even without checking the histogram, the exposure I see through the viewfinder is what I get when I press the shutter. 

On my X-T2, I can change a dial while looking through the viewfinder and see the exposure change as I move a dial. Even so, I still want a glance at the histogram.

Betty

Link to comment
Share on other sites

3 hours ago, Betty LaRue said:

With my Fuji, I have the histogram showing as I look through the viewfinder. I like my images exposed to the right, just short of clipping or touching the right side. If I see clipping, then I dial exposure comp down or change another setting like increasing shutter speed.

 

That gives me a well-exposed image without blowing the highlights. Especially important if I want any semblance of blue skies.

 

Fuji also has WYSIWYG. What you see is what you get. Even without checking the histogram, the exposure I see through the viewfinder is what I get when I press the shutter. 

On my X-T2, I can change a dial while looking through the viewfinder and see the exposure change as I move a dial. Even so, I still want a glance at the histogram.

Betty

 

However, the camera histogram is based on the in-camera JPEG and is not necessarily a good indicator of what is clipped, as the raw image will generally have a lot more latitude for bringing back overexposed highlights. As I don't shoot JPEGs, I don't bother with the camera histogram at all but I do use the Lightroom (ACR same thing) histogram extensively. 

Link to comment
Share on other sites

On 02/12/2017 at 20:31, John Mitchell said:

 

Can you elaborate on "I watch the histogram" ? What does the histogram look like when it looks good (keeping Alamy's requirements in mind)?

 

My QC failures have always been for the same reasons as yours (except for CA, never had one of those).

 

The histogram depends on the tonal range in the scene itself as well as the exposure. Assuming the scene has a wide enough tonal range, the histogram should stretch the whole way from black to white but should not show clipping, so should just reach without passing the end points. If the tonal range in the scene is narrow (mist for example) then the histogram will never reach the end points.There are many scenes where the tonal range is too wide so clipping will occur. This is why dynamic range is such an important technical feature of a camera sensor as the bigger the dynamic range, the greater the capacity to capture the extremes in a scene. 

 

A mass of pixels at the end points is likely to indicate clipping. In LR/ACR, clipped pixels can be visualised by turning on the clipping indicators which show up as red and blue in the image. Holding down the option or alt key while dragging a slider also gives a a visual clue about clipping.

 

In the case of highlights, the idea is to drag the Whites slider back to the point where clipping has just stopped (no more red) and then pull back the highlights using the Highlights slider. It is also possible to drag on the different parts of the histogram to achieve the same thing rather than use the sliders. 

 

I don't know what features Capture One has to indicate clipping but it presumably has something similar.

 

If one finds one's images look correct on a monitor in terms of brightness but the histogram is always short of reaching the white end, then the monitor is probably turned up too bright. Hardware calibration is always recommended (I recall you got a calibrator a while back).

 

 

Link to comment
Share on other sites

Just to add a tips about calibration. For MacOSX users, in the last versions of  the system the 'Expert Mode' is hidden when you want to calibrate your monitor. 

When you open the Monitor setup window, hold on the OPTION Key when you click on 'Calibrate...'. You will enter in 'Expert Mode' to setup you monitor calibration. 

Link to comment
Share on other sites

On 12/2/2017 at 14:31, John Mitchell said:

 

Can you elaborate on "I watch the histogram" ? What does the histogram look like when it looks good (keeping Alamy's requirements in mind)?

 

My QC failures have always been for the same reasons as yours (except for CA, never had one of those).

I haven’t had a failure for CA in a long time. I think because LR is so good at removing it. Plus my attention at 100% and the use of the eyedropper when LR doesn’t get it all initially.

Link to comment
Share on other sites

8 hours ago, MDM said:

 

The histogram depends on the tonal range in the scene itself as well as the exposure. Assuming the scene has a wide enough tonal range, the histogram should stretch the whole way from black to white but should not show clipping, so should just reach without passing the end points. If the tonal range in the scene is narrow (mist for example) then the histogram will never reach the end points.There are many scenes where the tonal range is too wide so clipping will occur. This is why dynamic range is such an important technical feature of a camera sensor as the bigger the dynamic range, the greater the capacity to capture the extremes in a scene. 

 

A mass of pixels at the end points is likely to indicate clipping. In LR/ACR, clipped pixels can be visualised by turning on the clipping indicators which show up as red and blue in the image. Holding down the option or alt key while dragging a slider also gives a a visual clue about clipping.

 

In the case of highlights, the idea is to drag the Whites slider back to the point where clipping has just stopped (no more red) and then pull back the highlights using the Highlights slider. It is also possible to drag on the different parts of the histogram to achieve the same thing rather than use the sliders. 

 

I don't know what features Capture One has to indicate clipping but it presumably has something similar.

 

If one finds one's images look correct on a monitor in terms of brightness but the histogram is always short of reaching the white end, then the monitor is probably turned up too bright. Hardware calibration is always recommended (I recall you got a calibrator a while back).

 

 

 

I tend not to pay a lot of attention (perhaps I should) to the in-camera histograms when shooting RAW and do exposure and contrast adjustments in post-processing, including setting the points to Alamy specs. I'll experiment more with using the clipping indicators more as you suggest. However, Images that do not have a full tonal range -- e.g. cloud images, foggy scenes, etc. -- often do not fit into the Alamy mould. You can't really set the points to the values Alamy specifies without destroying the look of the image. What do you do with those? I'm usually too paranoid to submit them even though they are perfectly fine. Would hate to have a QC failure for a faulty-looking histogram. It's a holdover from "Sin Bin" days.

Link to comment
Share on other sites

Mist and fog.

No, not channelling Alain Resnais, but I have a few of these where white probably doesn't go above 220. Alamy doesn't have a problem with them.

Rule of thumb: when you should correct a histo, do. When you shouldn't, don't.

Knowing the difference is one of our old craft skills that's definitely not obsolete.

Link to comment
Share on other sites

24 minutes ago, spacecadet said:

Mist and fog.

No, not channelling Alain Resnais, but I have a few of these where white probably doesn't go above 220. Alamy doesn't have a problem with them.

Rule of thumb: when you should correct a histo, do. When you shouldn't, don't.

Knowing the difference is one of our old craft skills that's definitely not obsolete.

 

I've always figured that was the case, but QC paranoia can be tough to kick. Adjusting the histogram to Alamy specs usually only makes sense with images that have a full tonal range.

 

Guess I need to be braver and step out of (or into) the fog more often. B)

Link to comment
Share on other sites

19 hours ago, Betty LaRue said:

With my Fuji, I have the histogram showing as I look through the viewfinder. I like my images exposed to the right, just short of clipping or touching the right side. If I see clipping, then I dial exposure comp down or change another setting like increasing shutter speed.

 

That gives me a well-exposed image without blowing the highlights. Especially important if I want any semblance of blue skies.

 

Fuji also has WYSIWYG. What you see is what you get. Even without checking the histogram, the exposure I see through the viewfinder is what I get when I press the shutter. 

On my X-T2, I can change a dial while looking through the viewfinder and see the exposure change as I move a dial. Even so, I still want a glance at the histogram.

Betty

 

Betty, I use the EVF histogram all the time.  I am finding that Fuji, like other manufacturers, are going a little overboard with their metering algorithms when it comes to protecting highlights.  When shooting with the X-T2, I found I generally had to dial in +0.6 to get the histogram where I wanted it, to the right. (expose to the right, or "ETTR").  I keep the exposure comp dial set at +.03 on my X-E2 and adjust further if needed.

 

@MDM is correct in stating the histogram is based on the jpeg representation of the image.  This is based on the jpeg settings you have in-camera, as would be detailed in the "Q" menu and the WB value.  If you are using Provia (Std) for your film sim and have the highlights, shadows etc set to "0" you can certainly get pretty far to the right, even creeping up the right side.  I've done a lot of testing as to how far you can go and it actually is quite a good distance up.  This really helps with high contrast images with deep shadows.

 

I think I read previously that you use LR for processing.  LR has perhaps the best highlight recovery that I've seen from RAW processing software.  I'm now using Capture One Pro 11 and it is greatly improved.  For me it is essential to shoot ETTR to avoid too much shadow noise when I bring up the shadows in post.  That histogram, for me, is a critical tool.

 

Rick

Link to comment
Share on other sites

23 minutes ago, Rick Lewis said:

 

Betty, I use the EVF histogram all the time.  I am finding that Fuji, like other manufacturers, are going a little overboard with their metering algorithms when it comes to protecting highlights.  When shooting with the X-T2, I found I generally had to dial in +0.6 to get the histogram where I wanted it, to the right. (expose to the right, or "ETTR").  I keep the exposure comp dial set at +.03 on my X-E2 and adjust further if needed.

 

@MDM is correct in stating the histogram is based on the jpeg representation of the image.  This is based on the jpeg settings you have in-camera, as would be detailed in the "Q" menu and the WB value.  If you are using Provia (Std) for your film sim and have the highlights, shadows etc set to "0" you can certainly get pretty far to the right, even creeping up the right side.  I've done a lot of testing as to how far you can go and it actually is quite a good distance up.  This really helps with high contrast images with deep shadows.

 

I think I read previously that you use LR for processing.  LR has perhaps the best highlight recovery that I've seen from RAW processing software.  I'm now using Capture One Pro 11 and it is greatly improved.  For me it is essential to shoot ETTR to avoid too much shadow noise when I bring up the shadows in post.  That histogram, for me, is a critical tool.

 

Rick

I believe you are braver than I with the highlights! :D

Link to comment
Share on other sites

15 hours ago, John Mitchell said:

 

I tend not to pay a lot of attention (perhaps I should) to the in-camera histograms when shooting RAW and do exposure and contrast adjustments in post-processing, including setting the points to Alamy specs. I'll experiment more with using the clipping indicators more as you suggest. However, Images that do not have a full tonal range -- e.g. cloud images, foggy scenes, etc. -- often do not fit into the Alamy mould. You can't really set the points to the values Alamy specifies without destroying the look of the image. What do you do with those? I'm usually too paranoid to submit them even though they are perfectly fine. Would hate to have a QC failure for a faulty-looking histogram. It's a holdover from "Sin Bin" days.

 

I'm pretty much the same. I don't tend to submit flat, low contrast images although I do submit misty pics if they have something clear and in focus, most likely in the foreground. I posted an example on Page 1 of this thread in fact. I guess flat foggy pics might sell as weather shots submitted through news. I was a lot more carfeul about what I would submit back in the days of having to upsize to 48MB from my 36MB D700. I had a few early SoLD failures which did induce paranoia and I have a lot of images from 2008-2012 which I never uploaded for that reason -they looked fine at native size but upsizing made them borderline. Dropping the min file size to where it stands now has made things a lot easier. 

Link to comment
Share on other sites

7 hours ago, Jill Morgan said:

I depend on my eye for almost all my PP.  Rarely refer to the histogram, either in camera or in ACR.  Won't say I've never referred to it, but rarely.

 

Jill

 

I liken this to driving without a speedometer. Although not essential, a quick glance at the raw histo can be very helpful. The human eye adjusts very quickly to different light levels and the histo acts as a calibrator. I just had a look at your recent images and a lot of them look fine in terms of white and black.

 

However, I did notice you have made a mistake in your keywording of those amazing rocks you have at the bottom of page 2 and top of page 3. HF8R1W has amazing textures but is unlikely to be ever found with your current keywording. I hope you don't mind me pointing this out - I'm a former professional geologist so my eye goes to rock pictures. You have used the terms "migmatic" and "migmatic gneises" (also a spelling mistake for gneiss here). There is no such word as migmatic. These rocks are migmatites which is the important keyword here if anybody was searching. You could use migmatitic gneisses as an adjective if you wanted but I would definitely have migmatite and migmatites (for some reason the plural is often used when referring to these rocks).

 

 

 

Link to comment
Share on other sites

2 hours ago, MDM said:

 

I liken this to driving without a speedometer. Although not essential, a quick glance at the raw histo can be very helpful. The human eye adjusts very quickly to different light levels and the histo acts as a calibrator. I just had a look at your recent images and a lot of them look fine in terms of white and black.

 

However, I did notice you have made a mistake in your keywording of those amazing rocks you have at the bottom of page 2 and top of page 3. HF8R1W has amazing textures but is unlikely to be ever found with your current keywording. I hope you don't mind me pointing this out - I'm a former professional geologist so my eye goes to rock pictures. You have used the terms "migmatic" and "migmatic gneises" (also a spelling mistake for gneiss here). There is no such word as migmatic. These rocks are migmatites which is the important keyword here if anybody was searching. You could use migmatitic gneisses as an adjective if you wanted but I would definitely have migmatite and migmatites (for some reason the plural is often used when referring to these rocks).

 

 

 

 

Thanks for the keywording tips.  My niece is a geologist in Alberta.

 

I thought my Batmobile images might push the black a bit, as I intentionally darkened the car and the curtain to give it that "dark knight" kind of feel.

 

Jill

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.