[swanlefitte]

Would any build in image processing setting of the K3 come close to the results from the CCD?

I think you mean this.
Get "that CCD look" with the K-3 / K-3II and Lightroom - PentaxForums.com

[TonyW]

If there is a preference for the 'look' of one sensor CCD vs CMOS so called SOOC that I can accept, but still the caveat remains that this is a function of the editing applied to the raw data by a particular editing application. Raw data is just that, raw in nature exactly as acquired by the acquisition device.


There is no secret sauce added to the raw data by the manufacturer other than the recipe they may embedd in the data file that allows there own software to apply specific algorithms. This proprietary data is embedded in the Tags and Fields section of the raw image container and is only usually interpreted by the manufacturers raw editing software in the case of proprietary raw such as PEF, NEF, CRW etc. DNG may also contain manufacturer proprietary information but this will not really be understood by most third party editors.


Debate has raged for some time on CCD v CMOS and there are pros and cons to both systems. For example I understand that CCD have noise advantage over CMOS. But the issue of colour rendering is not one that I have seen proved objectively.

With respect all the images posted so far appear to show small rendering differences (to be expected even between similar sensors from the same manufacturer). As a quick example the original gif altered with a touch of Hue and Sat only to Yellow, Blue and Red - not an attempt at an exact match but if this can be done in post it can certainly be done in a camera profile or within the raw editor environment and saved either as default or a preset.

Similarly the landscape scene presented can be fixed either in the making of a profile or even a simple one button Match Colour in PS.

IMHO it is all about an individual or companies desire to produce a pleasing/artistic/acceptable first image from raw data. That raw data looks horrible dark, low contrast prior to normalisation and demosaicing and then tweak to please. The image here shows a normal scene capture typically JPEG representation then to the right a raw data dump of the combined RGBG channels, a normailisedview with white and black points selected and finally a demosaiced view prior to WB, TRC and user preferences applied. This particular image from a DNG Validation dump

[BigMackCam]

If there is a preference for the 'look' of one sensor CCD vs CMOS so called SOOC that I can accept, but still the caveat remains that this is a function of the editing applied to the raw data by a particular editing application. Raw data is just that, raw in nature exactly as acquired by the acquisition device.
...
With respect all the images posted so far appear to show small rendering differences (to be expected even between similar sensors from the same manufacturer). As a quick example the original gif altered with a touch of Hue and Sat only to Yellow, Blue and Red - not an attempt at an exact match but if this can be done in post it can certainly be done in a camera profile or within the raw editor environment and saved either as default or a preset.
...
IMHO it is all about an individual or companies desire to produce a pleasing/artistic/acceptable first image from raw data.

Absolutely, Tony... we agree

I guess my point is, with more than one popular raw conversion software (I already mentioned darktable and RawTherapee, but I recall my experiences with Lightroom - when using the photo's embedded matrix - were similar too), the files from the GX-10 / K10D look better before adjustment than those from the K-3. Using the software's camera-specifc / enhanced profiles helps.

I completely accept that the photographer is free to make adjustments and save those as presets, or create custom profiles, which can then be reapplied as necessary. This is the approach I've used in the past to make my K-3 files look much, much closer to those from the GX-10.

In summary, I'm not saying the GX-10 produces better raw data than the K-3. Only that in the various raw converters I've used, the GX-10 files require fewer and smaller adjustments in order to achieve a pleasing result.

Of course, this is only my opinion

[Dartmoor Dave]

There is no secret sauce added to the raw data by the manufacturer other than the recipe they may embedd in the data file that allows there own software to apply specific algorithms. This proprietary data is embedded in the Tags and Fields section of the raw image container and is only usually interpreted by the manufacturers raw editing software in the case of proprietary raw such as PEF, NEF, CRW etc. DNG may also contain manufacturer proprietary information but this will not really be understood by most third party editors.

The DNG format has a generic Adobe colour profile and tone curve, but it also allows for an embedded profile that provides alternative values to plot colours and tone curve from the raw data. Any raw converter that works with DNG should be able to use that embedded profile as an alternative to the generic profile. What I called the "secret sauce" is the values in the manufacturer's own profile that the camera embeds into the DNG.

When you use .dcp camera profiles at the raw conversion stage, you replace the embedded profile with a different set of values for plotting the raw data within the colour space. Because of this, custom .dcp profiles can make photographs taken with CCD and CMOS sensors appear much more similar than the manufacturer embedded profiles tend to make them look.

But I've never managed to create .dcp profiles that can make a straight raw to jpeg conversion from my K-S1 look exactly the same as one from my GX-10. This makes me tend to believe that there is something different about the sensor hardware technologies that can't be completely overcome at the raw conversion stage.

[TonyW]

.....
In summary, I'm not saying the GX-10 produces better raw data than the K-3. Only that in the various raw converters I've used, the GX-10 files require fewer and smaller adjustments in order to achieve a pleasing result.

Of course, this is only my opinion

I can only guess that the raw converters standard adjustment is closer to your requirements/liking with the GX-10 and maybe the K-3 standard presentation is lacking in those raw converters you have tried? I may be wrong here as I have not handled GX-10 raw data, however I really cannot think of any reason why you could not make a profile/s that copied the GX-10 look under different lighting conditions closer to your requirements.

The DNG format has a generic Adobe colour profile and tone curve, but it also allows for an embedded profile that provides alternative values to plot colours and tone curve from the raw data. Any raw converter that works with DNG should be able to use that embedded profile as an alternative to the generic profile. What I called the "secret sauce" is the values in the manufacturer's own profile that the camera embeds into the DNG.

Yes that is correct to a point, Adobe does automatically use its own standard profile TRC and it is also true that the container which is DNG (TIFF/EP standards) can contain a ton of other proprietary manufacturers information on how to render the file.


But I believe that the assumption that any raw converter working with DNG could use the embedded profile as an alternative is wrong. The information embedded into the DNG may be placed in a Tag/Field reserved for that purpose. At its simplest it may be just a code to allow the manufacturers converter to do a specific job but not a way of describing to other software how to do it. A more complex method would be the embedded data that is the sidecar file. The problem is that there is no standard for encoding a series of instructions so that others may use. Simple enough to prove make a few adjustments using several different raw converters. Examine the resulting sidecar file instruction set and try and find another raw converter that can read and apply those instructions. The only application that can apply the instructions correctly will be the software that wrote the instructions to any other converter it will read as gobbledygook.

When you use .dcp camera profiles at the raw conversion stage, you replace the embedded profile with a different set of values for plotting the raw data within the colour space. Because of this, custom .dcp profiles can make photographs taken with CCD and CMOS sensors appear much more similar than the manufacturer embedded profiles tend to make them look.

Not sure I am following this correctly but a raw file does not have any embedded profile and I have also seen it argued that it does not even have a colour space (not sure I support that view completely)! Your .dcp or .icc profile is just another set of instructions on how to interpret/render your raw data. Therefore good custom profiles should be able to match any desired rendering but it really does depend on how you produce your profile and if you are going for accurate or a copy of a certain look.

But I've never managed to create .dcp profiles that can make a straight raw to jpeg conversion from my K-S1 look exactly the same as one from my GX-10. This makes me tend to believe that there is something different about the sensor hardware technologies that can't be completely overcome at the raw conversion stage.

Not having access to either camera I cannot comment other than to say in theory it should be possible to produce a profile that matches the colours, hue and sat and density of another at least close enough for most. You may have to use something a little more advanced than the X-Rite offering for example Lumariver Profile designer Lumariver

[BigMackCam]

I can only guess that the raw converters standard adjustment is closer to your requirements/liking with the GX-10 and maybe the K-3 standard presentation is lacking in those raw converters you have tried? I may be wrong here as I have not handled GX-10 raw data, however I really cannot think of any reason why you could not make a profile/s that copied the GX-10 look under different lighting conditions closer to your requirements.

As I mentioned, I've tried darktable, RawTherapee and Lightroom. Whilst there are variations to the rendering from each, the same general differences in hue, saturation and luminosity between GX-10 and K-3 files can be seen when converting using the embedded matrix. In all three of those converters, using the embedded matrix from the raw files, I like the GX-10 output "as is". Using the embedded matrix for the K-3, the files have (to a greater or lesser extent depending on the raw converter) the differences I've previously described.

Also as mentioned, I've created profiles of my own before - both single and dual illuminant - that go some way to replicating the GX-10 look with my K-3 files. So I have no problem recreating a similar (if not exactly identical) look with the K-3.

Again, my point isn't that the K-3 files are worse, nor that they can't be adjusted to give a similar appearance. My point is, the GX-10's files look great in all of the raw converters I've tried when I use the raw file embedded matrix as the input color profile. When I do the same with the K-3, in those same raw converters, the default starting point requires greater adjustment. Which can be addressed with manual adjustment, or custom presets / profiles. What I'm saying is, even allowing for different raw converters, the embedded matrix (and perhaps the sensor?) in the GX-10 results in files that look better without any work

[BigMackCam]

Not sure I am following this correctly but a raw file does not have any embedded profile and I have also seen it argued that it does not even have a colour space (not sure I support that view completely)! Your .dcp or .icc profile is just another set of instructions on how to interpret/render your raw data. Therefore good custom profiles should be able to match any desired rendering but it really does depend on how you produce your profile and if you are going for accurate or a copy of a certain look.

I can't comment on PEF raw files, but DNG files from Pentax cameras do have an embedded profile ("embedded matrix") which is the camera manufacturer's mapping of raw data to HSL information. This is the embedded matrix I've been referring to in most of my posts in the thread thus far

When loading a Pentax DNG file into darktable, RawTherapee, Lightroom and other raw converters, one of the options you'll see in that software's input profile settings is "Embedded". This tells the software to use the embedded matrix within the DNG file. Alternatively you can set the input profile to one provided by that software, or one you've created yourself.

[Dartmoor Dave]

Yes that is correct to a point, Adobe does automatically use its own standard profile TRC and it is also true that the container which is DNG (TIFF/EP standards) can contain a ton of other proprietary manufacturers information on how to render the file.But I believe that the assumption that any raw converter working with DNG could use the embedded profile as an alternative is wrong. The information embedded into the DNG may be placed in a Tag/Field reserved for that purpose. At its simplest it may be just a code to allow the manufacturers converter to do a specific job but not a way of describing to other software how to do it. A more complex method would be the embedded data that is the sidecar file. The problem is that there is no standard for encoding a series of instructions so that others may use. Simple enough to prove make a few adjustments using several different raw converters. Examine the resulting sidecar file instruction set and try and find another raw converter that can read and apply those instructions. The only application that can apply the instructions correctly will be the software that wrote the instructions to any other converter it will read as gobbledygook.

This really isn't a debating point. The DNG format is designed to contain an embedded profile. If you choose to use a raw converter that can't apply that profile, that's your decision. But the fact that your chosen raw converter can't apply the embedded profile doesn't mean that it doesn't exist.

[BigMackCam]

It's also worth mentioning that darktable specifically defaults to using the DNG's embedded matrix if one exists...

[TonyW]

I can't comment on PEF raw files, but DNG files from Pentax cameras do have an embedded profile ("embedded matrix") which is the camera manufacturer's mapping of raw data to HSL information. This is the embedded matrix I've been referring to in most of my posts in the thread thus far

When loading a Pentax DNG file into darktable, RawTherapee, Lightroom and other raw converters, one of the options you'll see in that software's input profile settings is "Embedded". This tells the software to use the embedded matrix within the DNG file. Alternatively you can set the input profile to one provided by that software, or one you've created yourself.

I do agree with what you are saying. DNG files do have all the camera information including embedded profile.


What I am suggesting is even with an embedded profile different raw converters will render the image data differently. As an example the same boring image with the embedded profile (645z) in use in LR and Darktable (I do not really use DT that much). In both cases images SOOC without any editing input from me show marked difference

[BigMackCam]

What I am suggesting is even with an embedded profile different raw converters will render the image data differently.

For sure

My extension to that is, from what I can tell across three different raw conversion softwares - darktable, RawTherapee and Lightroom - when using the embedded profile, the GX-10 images look nicer to begin with, requiring less processing to get a pleasing look. Yes, they look slightly different in each raw converter - but the same kinds of differences are clearly visible.

I suspect, though I'm only making an educated guess, that if I'd been stood beside you and taken that same shot with my GX-10, then loaded my photos into Lightroom and darktable using the embedded profile, the greens would have a little more pop, the blue skies would be bluer, the areas of earth would look slightly richer and warmer. Sadly, I don't have a 645Z so I can't test that

[GUB]

A really interesting comparison Mike.
I realise that the B&W tonal range is close to identical. But could the results in the colours suggest that the K-3 had a greater dynamic range as you would expect with sensor progress. I know I would prefer the K-3 to photograph a red flower. The GX-10 would clip the reds far more easily.

[BigMackCam]

A really interesting comparison Mike.
I realise that the B&W tonal range is close to identical. But could the results in the colours suggest that the K-3 had a greater dynamic range as you would expect with sensor progress. I know I would prefer the K-3 to photograph a red flower. The GX-10 would clip the reds far more easily.

According to DxO, the GX-10 / K10D dynamic range measures as 11.6 EV, versus 13.4 for the K-3. That's a significant difference. Having said that, I don't find myself particularly limited by the older sensor in this regard.

For situations and use-cases to which the GX-10 is suited, it's my preferred camera... but outside of those, its limitations become apparent rather quickly. There's no denying the K-3 is technically far superior, and the better choice for more challenging scenarios...

[Owltown]

EDIT: Thread moved from members' articles to Pentax DSLR Discussion per Adam's advice, for greater visibility...

Despite having been released in 2006, the Pentax K10D and its cousin, the Samsung GX-10, remain popular cameras even now in 2019. A major reason for this popularity is the claim by fans that the CCD sensor produces more pleasing colour and tones when compared to later CMOS sensor Pentax models.

I decided to perform a quick test to demonstrate the difference in colour and tone rendering between my Samsung GX-10 and Pentax K-3, as I felt it may be of interest to other members.

This was not a scientifically-controlled test, though I did attempt to make the conditions close to identical for both cameras.

I placed my X-Rite ColorChecker chart on my dining room table. Facing towards the window. It was a dull and very cloudy afternoon (the sky was completely greyish-white). I also had the overhead room light turned on, but this was not shining directly onto the target, being some way behind and above it.

Both GX-10 and K-3 were set to their sRGB colour space, "Natural" profiles, with DNG selected as the file format, ISO set to 100, and AWB white balance.

Using the same SMC Pentax-DA 35mm f/2.4 lens, I took almost identical shots of the ColorChecker chart with each camera, differing only slightly in angle and distance since these were taken hand-held.

I imported the resulting files into Darktable 2.4.4. For each photo, I:

- disabled all default processing options (base curve etc.) that might affect colour, shadows and highlights, black and white points, contrast, tone curve and so on
- set input color profile to "embedded matrix"
- set demosaic to AMaZE
- cropped and rotated to select the same area of the chart
- set white balance by spot selecting the same mid grey tone on the chart

Both shots were slightly under-exposed, so I adjusted them such that the luminosity of the mid-grey tones matched. (using the color picker tool for measurement). I then re-performed white balance adjustment, just in case.

Both shots were exported with output color profile set to sRGB (web-safe).

See below the results of the test, displayed as an animated GIF... please excuse the compression artefacts :


The difference in colour and tone is clear, with the GX-10 photo looking brighter, warmer and richer. Indeed, the GX-10 photo requires little further processing. The K-3 photo, by comparison, looks flatter... lacking in saturation; almost pastel-like. It requires more post-processing to achieve a pleasing result.

Some believe that the GX-10 / K10D sensor has a yellow cast, but it's not that simple. White balance has been normalised for both shots, and we can see that the white-grey-black squares match very closely, with no obvious colour tint. It's true that most of the colours have a warmer hue, but saturation and luminosity are also quite different. For example, the GX-10's reds are warmer, brighter and more saturated, while the blues are more saturated yet darker. Clearly there's a lot more to the differences than a simple colour tint.

Is the CCD sensor responsible for this difference in colour and tone? In the past, I've said that I prefer the "look" of photos from cameras with CCD sensors, but it's difficult to say conclusively whether the sensor alone should be credited for the rendering. I believe the sensor is partly responsible, but feel the camera's colour matrix is also a significant factor.

Whatever the reasons, I continue to prefer the colours and tones from my GX-10 raw files. I'm in no way unhappy with the K-3's files, though. Quite the contrary... They respond well to post-processing, and it's possible to achieve similarly appealing results. It just takes a little more work

As a final note, I'm unaware if the color matrices for the GX-10 and and K10D are the same, or tweaked slightly by each manufacturer. I don't own a K10D, so can't compare. However, it seems reasonable to assume they are the same, or at least quite similar.

I hope this is of interest to some of you

Many thanks for proving what I've suspected, ever since I started using a K3. I shoot jpgs 95% of the time, and use the "vibrant" setting just to get what is (to my eye) normal color saturation. Not looking for Kodachrome-style results, just trying to see the natural brilliance present in the original. And the "natural" setting doesn't get it ... results appear to be washed-out, trending towards pastels. A greater tonal range may indeed be available with the K3, but it requires boosting the sensor's color output to realize it.

[BigMackCam]

Many thanks for proving what I've suspected, ever since I started using a K3. I shoot jpgs 95% of the time, and use the "vibrant" setting just to get what is (to my eye) normal color saturation. Not looking for Kodachrome-style results, just trying to see the natural brilliance present in the original. And the "natural" setting doesn't get it ... results appear to be washed-out, trending towards pastels. A greater tonal range may indeed be available with the K3, but it requires boosting the sensor's color output to realize it.

It's very noticeable in straight-from-camera JPEGs, for sure. The "Natural" setting is, I think (but I'm only guessing), intended to provide a relatively neutral starting point for further adjustment in post-processing. I can't imagine folks using that setting for a finished image. As you say, it tends to look rather washed-out and pastel-like. But the K-3 provides a reasonable amount of adjustment for each of the profiles, so it's possible to adjust hue shift, saturation and contrast. These adjustments can be used to create very pleasing JPEGs... yet they don't provide sufficient adjustment to replicate the same kind of colour and tone from earlier cameras such as the K10D...