Colors from different Camera Models

[xpatUSA]

We often compare colors from different cameras, often stating a preference or a dislike for one over another. However, it is not always mentioned fully what rendered those colors. Even worse, some might not mention the extensive slider waggling behind those colors ... the "I always" syndrome ...

A pity because, when posting comparisons, it is not always obvious when one member is posting from, say, some version of SPP not at default while another might be posting from, say, RawTherapee at default.

"So what?" some might say, "what's the difference?"

Here's the difference - the same X3F raw converted by four converters at their default settings:

RawTherapee, FastStone Viewer, XnView MP, RawDigger.

Quite frankly it can be said that quibbling over color casts, skin rendition, shadow contrast, blown highlights, etc. without being specific is a waste of webspace and anyway, with modern processing, any one of the above could be made to look like any one of the others.

Yes, long ago I posted a diatribe about the SD14's green cast. But, I could have learned to live with it and dealt with it instead of selling the camera at a loss shortly thereafter.

 

[Ceistinne]

We often compare colors from different cameras, often stating a preference or a dislike for one over another. However, it is not always mentioned fully what rendered those colors. Even worse, some might not mention the extensive slider waggling behind those colors ... the "I always" syndrome ...

A pity because, when posting comparisons, it is not always obvious when one member is posting from, say, some version of SPP not at default while another might be posting from, say, RawTherapee at default.

"So what?" some might say, "what's the difference?"

Here's the difference - the same X3F raw converted by four converters at their default settings:

RawTherapee, FastStone Viewer, XnView MP, RawDigger.

Quite frankly it can be said that quibbling over color casts, skin rendition, shadow contrast, blown highlights, etc. without being specific is a waste of webspace and anyway, with modern processing, any one of the above could be made to look like any one of the others.

Yes, long ago I posted a diatribe about the SD14's green cast. But, I could have learned to live with it and dealt with it instead of selling the camera at a loss shortly thereafter.

Ted,

Other than RT what are the other 3. ?

S

 

[xpatUSA]

We often compare colors from different cameras, often stating a preference or a dislike for one over another. However, it is not always mentioned fully what rendered those colors. Even worse, some might not mention the extensive slider waggling behind those colors ... the "I always" syndrome ...

A pity because, when posting comparisons, it is not always obvious when one member is posting from, say, some version of SPP not at default while another might be posting from, say, RawTherapee at default.

"So what?" some might say, "what's the difference?"

Here's the difference - the same X3F raw converted by four converters at their default settings:

RawTherapee, FastStone Viewer, XnView MP, RawDigger.

Quite frankly it can be said that quibbling over color casts, skin rendition, shadow contrast, blown highlights, etc. without being specific is a waste of webspace and anyway, with modern processing, any one of the above could be made to look like any one of the others.

Yes, long ago I posted a diatribe about the SD14's green cast. But, I could have learned to live with it and dealt with it instead of selling the camera at a loss shortly thereafter.

Ted,

Other than RT what are the other 3. ?

S,

They were listed in the caption under the posted image but you are maybe viewing on something different to me, so:

FastStone Viewer, XnView MP, RawDigger ... see also the filenames in the four panes for hints as tonwhich is which.

My main point is to show the huge possible differences of color and tone from a single raw and and from a single camera at default settings.

As completely opposed to many comparisons posted here where it appears to be assumed that everybody and his dog is using the very latest SPP 6.x.x ...

... and where far smaller differences than the above examples are treated as crimes of the century.

One of the more ludicrous possible comparisons is that of, say, a Quattro DNG opened in say Photoshop/LR's ACR then printed with some unknown Print Driver which is then compared under unknown light with, say, another camera's raw on a monitor.

 

[xpatUSA]

<>

A pity because, when posting comparisons, it is not always obvious when one member is posting from, say, some version of SPP not at default while another might be posting from, say, RawTherapee at default.

"So what?" some might say, "what's the difference?"

Here's the difference - the same X3F raw converted by four converters at their default settings:

RawTherapee, FastStone Viewer, XnView MP, RawDigger.
<>

For those who "always use" SPP, here is the X3F opened in SPP 3.5.2:

It can be seen that RT is the closest rendering in terms of brightness.

Also that the capture was quite under-exposed:

log-log scales.

Since we are talking about color, it would be interesting to compare Hues, maybe SPP versus the others, by extracting HSV hues and showing the difference between them.

 

[xpatUSA]

<>

Since we are talking about color, it would be interesting to compare Hues, maybe SPP versus the others, by extracting HSV hues and showing the difference between them.

When Hue is extracted in the GIMP, the result is a grayscale image where the grayscale values represent Hue on a circular scale where black is 0 degs (0/255) i.e. red then pro rata all the way around to 360 degs (255/255) which is again red. So for example blue = 240 degs = 240/360x255 = 170/255.

Without further ado, here's SPP:

At 157/255 the sky is 222 degs, so not pure blue. At lower left, possible Hue banding in the shadows.

And now RT:

Sky pure blue at about 170/255. Obvious hue banding in the shadows.

So the "colors" are quite different, even though they look similar in an RGB view.

And here's the one subtracted from the other, where 0 = no difference and 255 = maximally different:

Skies and shadows similar (dark gray) lit foliage less so (light gray).

My point remains that, with such large differences from converters at default settings with one camera and one capture, whining about small color bdifferences between camera models and various converters carries little weight as far as I am concerned!

 

[xpatUSA]

<>

Since we are talking about color, it would be interesting to compare Hues, maybe SPP versus the others, by extracting HSV hues and showing the difference between them.

When Hue is extracted in the GIMP, the result is a grayscale image where the grayscale values represent Hue on a circular scale where black is 0 degs (0/255) i.e. red then pro rata all the way around to 360 degs (255/255) which is again red. So for example blue = 240 degs = 240/360x255 = 170/255.

For those who are uncomfortable with interpreting grayscale values as Hues, here are SPP and RT in glorious Hue "colors":

SPP:

RT:

Astounding differences, especially in the shadows !!

These Hues are shown with Saturation and Value set to light gray, then recomposed to RGB.

How do you like your foliage Sir? Green or Orange?

I'll be sticking with SPP, I reckon ...

... but my point remains that, with such large differences from converters at default settings with one camera and one capture, whining about small color-differences between camera models and various converters carries little weight as far as I am concerned!

 

[SandyF]

Basically, we've known forever that some RAW converters handle the X3F RAW really poorly. We've also known (at least I'm convinced) that there are differences in the "color" output from different model Sigma cameras. Some cameras seem to be usually generally "spot on" to desired color; other cameras' output regularly need some "color" adjustment to be satisfactory to the user.

Some cameras are "easier" to adjust to get the desired color than other cameras. To me, that's the big take-away point to a "color" discussion. For example: for me SD15 and other Sigma cameras are easier to adjust than the SD14. I find many Merrill photos need a slight touch more magenta, but it's not difficult to adjust a series of Merrill photos taken same time/same place. I can often adjust to a consistent result by adding M5 on the colorwheel/colorblock.

What made the SD14 output difficult for me was that my two different physical cameras could vary the output greatly from a photo taken and then another taken same time/same place. My original SD14 was more variable than the second one, so I also think there is some variation between physical cameras (the sensors?)

 

[xpatUSA]

Basically, we've known forever that some RAW converters handle the X3F RAW really poorly. We've also known (at least I'm convinced) that there are differences in the "color" output from different model Sigma cameras. Some cameras seem to be usually generally "spot on" to desired color; other cameras' output regularly need some "color" adjustment to be satisfactory to the user.

Some cameras are "easier" to adjust to get the desired color than other cameras. To me, that's the big take-away point to a "color" discussion. For example: for me SD15 and other Sigma cameras are easier to adjust than the SD14. I find many Merrill photos need a slight touch more magenta, but it's not difficult to adjust a series of Merrill photos taken same time/same place. I can often adjust to a consistent result by adding M5 on the colorwheel/colorblock.

What made the SD14 output difficult for me was that my two different physical cameras could vary the output greatly from a photo taken and then another taken same time/same place.

Dare I ask if that variation was with same settings and the same framing? If so, that would be really bad.

My original SD14 was more variable than the second one, so I also think there is some variation between physical cameras (the sensors?).

 

[SandyF]

....

What made the SD14 output difficult for me was that my two different physical cameras could vary the output greatly from a photo taken and then another taken same time/same place.

Dare I ask if that variation was with same settings and the same framing? If so, that would be really bad.

My original SD14 was more variable than the second one, so I also think there is some variation between physical cameras (the sensors?).

My first SD14 (probably one of the first purchased in US) had some peculiarities. Sigma Japan replaced it. Second one was much better, but still had a tendency to be greenish, and I would edit color. I still have that camera, but I haven't used it in years. Where I would get especially frustrated was then matching output of SD14 with simultaneously used dp1 and dp2. I've always had a habit of carrying around multiple cameras

Added: I can match up my DP Merrills pretty easily, ditto Quattros dp3Q and SDQ, and even Quattros to Merrills and to the 2 big Canons I have. I gave Pentax cams to son-in-law after being dissatisfied with them.

 

[SandyF]

Some of my earliest SD14 photos are here

https://pbase.com/sandyfleischman/sigma_sd14_2007_gallery&page=all

Scroll down to

#3213 Cannot be too bad.... Laurence made a print of the Chihuly glass which was hung at a PMA trade show... Of course after his processing, not mine LOL. I honestly don't know if this was first or second camera. The very low SDIM#s are probably first camera.

https://pbase.com/sandyfleischman/image/83056463

Added: I then have to # SDIM6133 on flickr 2012. Seems to be when I switched over to other Sigma cameras for regular usage.

 

[xpatUSA]

....

What made the SD14 output difficult for me was that my two different physical cameras could vary the output greatly from a photo taken and then another taken same time/same place.

Dare I ask if that variation was with same settings and the same framing? If so, that would be really bad.

My original SD14 was more variable than the second one, so I also think there is some variation between physical cameras (the sensors?).

My first SD14 (probably one of the first purchased in US) had some peculiarities. Sigma Japan replaced it.

So your first SD14 got variations with the same scene, time, settings and framing, it seems.

Second one was much better, but still had a tendency to be greenish, ...

As did all three of mine.

... and I would edit color. I still have that camera, but I haven't used it in years. Where I would get especially frustrated was then matching output of SD14 with simultaneously used [DP1] and [DP2].

Yes, that's why God invented input profiles or sidecar files. A long time ago, I made one for the SD14 - but got tired of it and used Custom WB instead.

I've always had a habit of carrying around multiple cameras

Added: I can match up my DP Merrills pretty easily, ditto Quattros dp3Q and SDQ, and even Quattros to Merrills and to the 2 big Canons I have. I gave Pentax cams to son-in-law after being dissatisfied with them.