I would like to know wich camera system offer that nice color

[Manoli]

I am far more interested in what comes out of a camera than what goes inside it. YMMV.

One day, you may finally understand that what's inside the camera largely governs what comes out of it.

You may well not have a clue as to what's inside the camera, but trotting out moronic platitudes ad nauseam isn't going to further your understanding and contributes to these threads even less.

[Bart_van_der_Wolf]

One day, you may finally understand that what's inside the camera largely governs what comes out of it.

Indeed, miracles do happen occasionally ...

Cheers,
Bart

[synn]

One day, you may finally understand that what's inside the camera largely governs what comes out of it.

You may well not have a clue as to what's inside the camera, but trotting out moronic platitudes ad nauseam isn't going to further your understanding and contributes to these threads even less.

And one day, you will understand that there are many people who make purchase decisions based on what comes out of the camera.

You understand what goes into each of them? Good for you. Now run along and write to each manufacturer and tell them what they can do better. While you change the world one EXPEED at a time, I'll be happy shooting.

Speaking of contributions, I would REALLY like to know what you have contributed in this forum other than snark and pleas for attention though. Humor me.

[eronald]

Indeed, miracles do happen occasionally ...

Cheers,
Bart

Yes, it seems pigs can fly

Edmund

[Bart_van_der_Wolf]

And one day, you will understand that there are many people who make purchase decisions based on what comes out of the camera.

Hi Sandeep,

One cannot see what comes out of the camera, until the data is processed/Raw converted. That involves Bayer CFA demosaicing, White balancing, color mapping or assigning a profile, Tonecurve assignment, etc. etc.

It's all sorts of post-processing that has to be done before one can see the result (on an output medium that's hopefully properly profiled for the viewing conditions). It's the quality of that post-processing that makes a difference, even in cases where there is little difference in the input data. Of course, some input data is better than others, but that requires knowledge of the internal technicalities to discern, and cannot be simply derived from the processed output, too many variables for that.

As you have proven to yourself, a D800 file with an IQ250 profile gives quite pleasing results, perhaps better than the default D800 profile. It's the post-processing that makes the most significant differences in color rendition.

Cheers,
Bart

[hjulenissen]

Color comes from light source, subject, lens filter, lens coating, lens design, IR filter, CFA filters, sensor, color engine of the software, color profile in the software for the camera, post processing adjustments.

In principle, skilled photoshop artists (or software developers) could make an image of a girl into a landscape. While this is impractical due to the excessive time and effort needed, it goes to show that processing is a kind of wildcard here. Even if we limit ourselves to reasonable amounts of time/money spent on editing (as well as limited skill), it is clear that images from different cameras can end up with a wide range of end-results, occasionally perceptually similar results.

A valid question would seem to be "how important can gear be if so much can be done in (e.g.) photoshop?" Does more expensive (or more carefully chosen) gear improve the "color accuracy" of the (unedited) images? Does it improve the "color pleasantness"? Does it mean that for a given photographers needs/taste, she can get what she wants with less time and effort?

For my own humble color needs, I have found that camera profiling and developing editing skills is more important than e.g. whatever the differences in lens coating, lens design, IR filter, CFA filters, sensor etc of my Sony and Canon cameras.

-h

[synn]

Hi Sandeep,

One cannot see what comes out of the camera, until the data is processed/Raw converted. That involves Bayer CFA demosaicing, White balancing, color mapping or assigning a profile, Tonecurve assignment, etc. etc.

It's all sorts of post-processing that has to be done before one can see the result (on an output medium that's hopefully properly profiled for the viewing conditions). It's the quality of that post-processing that makes a difference, even in cases where there is little difference in the input data. Of course, some input data is better than others, but that requires knowledge of the internal technicalities to discern, and cannot be simply derived from the processed output, too many variables for that.

As you have proven to yourself, a D800 file with an IQ250 profile gives quite pleasing results, perhaps better than the default D800 profile. It's the post-processing that makes the most significant differences in color rendition.

Cheers,
Bart

Hi Bart,

I don't think I denied any of that and this discussion has strayed very far from what I was originally saying. Anyone with more than a passing interest in photography knows there are many factors involved before final image takes shape. What I was saying though is, I don't need to know who made what CFA for whom to know if one particular camera delivers me the results I am looking for better than another one.

Post processing is a completely different topic to the post I was originally responding to. As I have mentioned previously, that is one aspect of the workflow I pay a lot of attention to as I CAN control it. I CANNOT control the lens coatings Schneider uses or the CFA dyes DALSA applies for Leaf vs Phaseone or whatever. I have zero interest in understanding the technicalities of any of those because me knowing any of this doesn't mean I can change any of those to my advantage. I cannot write to DALSA tomorrow to request a different CFA or whatever. I only need to choose a device that gives me a good SOOC starting point for my post production. How they do it, I really have no interest in knowing.

 But knowing post production techniques does make a difference as I can pick and choose them to my advantage. I CAN choose a different color profile to get a more pleasant output. With my limited time, I choose to focus on the parts that help me create better images.

As I said at the end of my post, your mileage may vary. This means that I am not holding my opinion as a universal truth. End of the day, I feel happy looking at the images I create. I suppose there might be some here who feel happiness looking at a test chart from Imaging-resource or something. Whatever floats their boat.


You're right though, sometimes miracles do happen. I have seen professional chart readers one day realizing that the benchmark-test-measurometer they have in their hand has another capability, which is making pictures. I like it when that happens.  

[ErikKaffehr]

Hi,

Some guys will say that MFDBs offer the ultimate colour. Or some will say hat CCD offers better colour than CMOS.

There are a lot of factors in this equation:

• IR filter affects the light falling on the sensor
• CFA (Color Filter Array) decides what colours the camera detects
• The Color Conversion Matrix translates the camera RGB intensities to XYZ coordinates, WB is also handled in this step
• An input profile converts the XYZ into local working space
• An output profile converts the image to a final RGB file

So there are two physical filters and a lot of simple math. Personally I would guess that the math plays a significant role in the process.

Personally I shoot both Sony cameras with CMOS sensors and a CCD based P45+. I have placed quite a few raw files from the P45+ on my web-site.

You can find them here:

http://echophoto.dnsalias.net/ekr/Articles/Shoots/BernardSamples/ (several clickable directories)

http://echophoto.dnsalias.net/ekr/Articles/MFDJourney/RawImages/Samples2
http://echophoto.dnsalias.net/ekr/Articles/MFDJourney/RawImages/Samples3

These image are comparable shots with P45+ and Sony Alpha 99:
http://echophoto.dnsalias.net/ekr/Articles/MFDJourney/RawImages/MFDB_VS_DSLR/
http://echophoto.dnsalias.net/ekr/Articles/MFDJourney/RawImages/MFDB_VS_DSLR2/

All these samples include raw originals, feel free to download and process with the raw processor of your choice. Please keep in mind that my copyright applies. But you can use all those image freely for any non commercial purpose.

Keep in mind that all MFDBs are not created equal. Many prefer the colours from the DALSA based sensors over the KODAK sensor the P45+ is based on. Leaf is said to have different rendition from Phase One.

This thread on GETDPI.com may also give some insight:
http://www.getdpi.com/forum/medium-format-systems-digital-backs/52173-if-ccd-rendering-can-achieved-cmos-where-examples.html

Be advised, a lot of critique had been directed at my samples, saying that I don't make good use of the P45+, but I guess that those points are not really relevant, as you are mostly interested in colour rendition and I would suggest you try your own raw conversions.

Note: Many of the raw files are in DNG-format, but all have the original file imbedded. The original file can be extracted by Adobe DNG Converter, downloadable here: http://helpx.adobe.com/photoshop/digital-negative.html#downloads



Best regards
Erik

[Pics2]

Erik and Synn - two different worlds 

[ACH DIGITAL]

Hi guys, i'm new here.
I am architectural photographer and i love the color coming from the images of this two guys, that are identical:

http://scottfrances.com/

http://www.joefletcherphoto.com/projects

Does anyone know with wich system or camera/digital back are they shot?
I have used Sony a7r and Canon 5dIII and these images aren't definitely coming from this cameras.
Thanks in advance!

Art, I would dare to say that Scott Frances is DSLR and Joe Fletcher is MF Leaf. Joe's work is evidently medium format for the detail, frame and colors. I would say Leaf for the colors.
Hope I'm not wrong.

ACH

[Chris Barrett]

In response to the OP, I think the cameras are nigh irrelevant.  What you're really responding to is the impactfulness of the photographs.  It's all about the abilities of these two shooters and the quality of their retouching.

I shoot the IQ260 and the A7r side by side.  I prefer the Phase One file, but if I switched completely to the Sony, neither you nor my clients could ever tell the difference.

IMO,
CB

[ErikKaffehr]

Hi Chris,

Well explained, sometimes a few words say more than a thousand pictures!

Best regards
Erik

In response to the OP, I think the cameras are nigh irrelevant.  What you're really responding to is the impactfulness of the photographs.  It's all about the abilities of these two shooters and the quality of their retouching.

I shoot the IQ260 and the A7r side by side.  I prefer the Phase One file, but if I switched completely to the Sony, neither you nor my clients could ever tell the difference.

IMO,
CB

[eronald]

Bart,

 I think that Synn is a photographer, and artist at heart. He has good color vision and judgement. When he was looking for a camera system he adopted "the best", ie a Phase/Leaf configuration, which is exceptional in out-of-the box color experience, by means of its completely seamless fusion of hardware/software color calibration, and software and training that mandates a good workflow.

 As his color judgement is good, he perceives  the deficiencies of non-Phase systems. However as an artist he believes that the difference between exceptionally good and mediocre is intrinsic, and in some way arises from the use of high grade materials and methods, and justifies the price. It doesn't seem feasible to him that it might boil down to custom calibration/linearisation issues in the camera, a couple of matrices written in a file, and some very good marketing. This is something we as engineers can visualize, calibration making all the difference, but which is not immediately apparent to an artist. And of course it has not been in the interest of the camera manufacturers to educate the public; in particular, Phase have lived well off the myth that the back CFAs and software are "special".

What a surprise: artists and engineers have different perceptions.


Edmund

Hi Sandeep,

One cannot see what comes out of the camera, until the data is processed/Raw converted. That involves Bayer CFA demosaicing, White balancing, color mapping or assigning a profile, Tonecurve assignment, etc. etc.

It's all sorts of post-processing that has to be done before one can see the result (on an output medium that's hopefully properly profiled for the viewing conditions). It's the quality of that post-processing that makes a difference, even in cases where there is little difference in the input data. Of course, some input data is better than others, but that requires knowledge of the internal technicalities to discern, and cannot be simply derived from the processed output, too many variables for that.

As you have proven to yourself, a D800 file with an IQ250 profile gives quite pleasing results, perhaps better than the default D800 profile. It's the post-processing that makes the most significant differences in color rendition.

Cheers,
Bart

[Ken R]

Bart,

 I think that Synn is a photographer, and artist at heart. He has good color vision and judgement. When he was looking for a camera system he adopted "the best", ie a Phase/Leaf configuration, which is exceptional in out-of-the box color experience, by means of its completely seamless fusion of hardware/software color calibration, and software and training that mandates a good workflow.

 As his color judgement is good, he perceives  the deficiencies of non-Phase systems. However as an artist he believes that the difference between exceptionally good and mediocre is intrinsic, and in some way arises from the use of high grade materials and methods, and justifies the price. It doesn't seem feasible to him that it might boil down to custom calibration/linearisation issues in the camera, a couple of matrices written in a file, and some very good marketing. This is something we as engineers can visualize, calibration making all the difference, but which is not immediately apparent to an artist. And of course it has not been in the interest of the camera manufacturers to educate the public; in particular, Phase have lived well off the myth that the back CFAs and software are "special".

What a surprise: artists and engineers have different perceptions.


Edmund

You should start a company that provides that service for any Digital camera owner. One sends in the camera, you do the custom calibration, solve the linearisation issues and write a few files and voilá, great MFDB like color in any DSLR (or mirrorless) camera. Im sure you can make good money. 

[uaiomex]

Wow, that's big Chris!

If you'd print to 36" on the short side from both cameras, could you say the same?

Thanks in advanced.

Eduardo

I shoot the IQ260 and the A7r side by side.  I prefer the Phase One file, but if I switched completely to the Sony, neither you nor my clients could ever tell the difference.

IMO,
CB

[eronald]

great MFDB like color in any DSLR (or mirrorless) camera. Im sure you can make good money.  

"Any dSLR" is a tall order. But many, yes. Camera calibration often does work, if one knows how to do it. Of course it would be easier to do in factory.
As for money, I don't think so. I guess most people who want color badly enough to care just get a DB where *decent* calibration is part of the package you pay for.
And the few photographers who *really need* color to earn their living, eg. to do product photography or art repro are usually themselves already technical experts , and have a very precise workflow with on-site calibration in the location lighting.

BTW, I was a consultant for company that makes a well known consumer camera calibration device, and before they marketed this device we tried to push a more sophisticated in-factory camera calibration method to dSLR makers, and were met with a very tepid response. I contacted the brands myself  at their request so I saw it firsthand. The existing color quality is perceived as good enough, especially within the context of workflows with uncalibrated monitors.

Edmund

[araucaria]

I personally own a D800, and also several medium format film cameras as I don't have the money to buy any digital back (until recently, because now the p45 seems to get cheaper every minute).

So my opinion regarding medium format digital backs is completely based on the material I have gathered on the internet.

My personal opinion, and I believe Synn has pointed out something similar on his leaf puchase thread, is that the d800 (for example) has this sort of color bleeding (I've made this name up to describe it somehow). I see the effects of this very clearly in some situations.
For example,the dynamic range seems huge, but there is no way of using it without Photoshopping/masking many layers, and even then it depends a lot on your skills. If you try to raise the shadows or lower the highlights you will immediately get this Tonemapping sort of look (bad HDR look).
Try to compare this to one of Eriks samples,http://echophoto.dnsalias.net/ekr/Articles/Shoots/BernardSamples/Distagon_50/ When you raise or lower the shadows it seems to only affect the shadows, if you do the same on the D800 you will get halos around every detail, and the color quality goes down the drain dramaticly. There seems to be some sort of inertia spanning over larger areas (dozens of pixels) which gives this sort of tonemapping look. I'm sorry I don't have any technical term for this, but I think you all know what I'm talking about. Synn's samples of nightscapes are another example where color seems to bleed, or just being ugly. Of course if you go all crazy with this with the samples provided by erik you will get that problem too, but there is a big difference to the d800.

Another scenario is in skin colors. I guess you are familiar with the way c41 films render skins (portra 160 for example). I'm talking about this slightly over exposed skin look. Well, try to do the same on a d800, there will be no detail, the skin seems like one big blub that only get's sort of details when it goes into the left side of the histogram. I'm not talking about recovering highlights (which is also pretty terrible, but that's where film wins over both digital backs and cmos), but about this strange behaviour where the subtilities of lighter skin can't be represented by the D800, but both C41 film and some samples I've seen of digital backs have no problem to represent it.

I would love to know if there is some way of fixing this because I can't think of a reason to explain this "inertia" (I do have very good lenses, it's not an aberration or contrast problem)

[eronald]

I would love to know if there is some way of fixing this because I can't think of a reason to explain this "inertia" (I do have very good lenses, it's not an aberration or contrast problem)

A consequence of the local-delta encoding of the files?
I don't think it's the sensor per se, more the way in which it is being employed in the camera (firmware).

Edmund

[ErikKaffehr]

Hi,

I am using LR as raw converter, even for the P45+. Now LR5 has some issues on P45+, in my view mostly in the demosaicing department: http://www.luminous-landscape.com/forum/index.php?topic=94812.0

I have only Sonys to compare with, but what I see is that I can pull incredible shadow detail out of the Sony images using Lightroom's shadows control, without obvious artefacts. Using the "highlights" slider to bring down  sky can often yield bad artefacts on thing like treetops. That depends on tone mapping used in Lightroom. The tone mapping doesn't work for small areas surrounded by dark detail.

Using DNG compatible raw converters it is very easy to make a decent camera specific calibration using Adobe DNG Profile Editor with a ColorChecker card. The colour checker Passport software is a similar product. A third solution is QPCard which uses a different set of colours. All these software solutions are free, but you need to buy a ColorChecker or QPCard.

Most raw converters I have tested can use DNG Colour Profiles even on non DNG raw files. Raw converters I have tested with DNG Profiles are Lightroom, ACR, Raw Therapee, AccuRaw and Iridient Raw Developer. Capture One is a notable exception.

This article covers some of that ground: http://echophoto.dnsalias.net/ekr/index.php/photoarticles/79-p45-colour-rendition

Sandy McGuffog, the author of AccuRaw and some other nifty softwares, pointed to this interesting article in his blog:
https://www.hansvaneijsden.com/colorchecker-perfect-skin-colors/ worth reading.

Personally, I have made Adobe DNG Profiles for all my cameras. On the Sony SLT 99 I mostly use Adobe Standard profile as it works fine for me. With the P45+ I use my own DNG Profiles.

As a side note, a ColorChecker card is not good enough for a full calibration, for that more stuff is needed (*).  

The reason Adobe DNG Profile Editor works is that they use ColorChecker to tweak a hpefully well designed base profile.

Best regards
Erik

(*) Something like this may give a proper starting point:
http://image-engineering-shop.de/shop/article_IE-camSPECS%2520express/camSPECS-express.html?sessid=jD6bM1hTL2TlDlB1AhLiLIit6AJVf3f7stolSdsx0jh1jstKDPOweX6p6nyy8OcJ&shop_param=cid%3D565%26aid%3DIE-camSPECS%2520express%26

I personally own a D800, and also several medium format film cameras as I don't have the money to buy any digital back (until recently, because now the p45 seems to get cheaper every minute).

So my opinion regarding medium format digital backs is completely based on the material I have gathered on the internet.

My personal opinion, and I believe Synn has pointed out something similar on his leaf puchase thread, is that the d800 (for example) has this sort of color bleeding (I've made this name up to describe it somehow). I see the effects of this very clearly in some situations.
For example,the dynamic range seems huge, but there is no way of using it without Photoshopping/masking many layers, and even then it depends a lot on your skills. If you try to raise the shadows or lower the highlights you will immediately get this Tonemapping sort of look (bad HDR look).
Try to compare this to one of Eriks samples,http://echophoto.dnsalias.net/ekr/Articles/Shoots/BernardSamples/Distagon_50/ When you raise or lower the shadows it seems to only affect the shadows, if you do the same on the D800 you will get halos around every detail, and the color quality goes down the drain dramaticly. There seems to be some sort of inertia spanning over larger areas (dozens of pixels) which gives this sort of tonemapping look. I'm sorry I don't have any technical term for this, but I think you all know what I'm talking about. Synn's samples of nightscapes are another example where color seems to bleed, or just being ugly. Of course if you go all crazy with this with the samples provided by erik you will get that problem too, but there is a big difference to the d800.

Another scenario is in skin colors. I guess you are familiar with the way c41 films render skins (portra 160 for example). I'm talking about this slightly over exposed skin look. Well, try to do the same on a d800, there will be no detail, the skin seems like one big blub that only get's sort of details when it goes into the left side of the histogram. I'm not talking about recovering highlights (which is also pretty terrible, but that's where film wins over both digital backs and cmos), but about this strange behaviour where the subtilities of lighter skin can't be represented by the D800, but both C41 film and some samples I've seen of digital backs have no problem to represent it.

I would love to know if there is some way of fixing this because I can't think of a reason to explain this "inertia" (I do have very good lenses, it's not an aberration or contrast problem)

[eronald]

Erik,

 If it's relevant, I'm aware of Dietmar's device, and obviously I'm aware of the Passport as I was involved in its design as a portable target
 It's late and I'm not thinking straight, but I believe that once you have the sensor's spectral sensitivity curve (eg. Dietmar's gadget), you still need to measure the illuminant in situ with a spectro. Even if you were to scan your scene with a teleradiometer or an XX channel camera,  to turn the spectral measurements into *color* you still need information about the illuminant

 This whole discussion reminds me why I've given up on color. I'm taking up video at the moment, and just noticed that what comes out of the cameras is close to what Michael calls "jpeg", with colors completely off if you haven't set wb. And I'm trying to figure out an acceptable workaround in the context of a Lumix GH4.

Edmund

Hi,

I am using LR as raw converter, even for the P45+. Now LR5 has some issues on P45+, in my view mostly in the demosaicing department: http://www.luminous-landscape.com/forum/index.php?topic=94812.0

I have only Sonys to compare with, but what I see is that I can pull incredible shadow detail out of the Sony images using Lightroom's shadows control, without obvious artefacts. Using the "highlights" slider to bring down  sky can often yield bad artefacts on thing like treetops. That depends on tone mapping used in Lightroom. The tone mapping doesn't work for small areas surrounded by dark detail.

Using DNG compatible raw converters it is very easy to make a decent camera specific calibration using Adobe DNG Profile Editor with a ColorChecker card. The colour checker Passport software is a similar product. A third solution is QPCard which uses a different set of colours. All these software solutions are free, but you need to buy a ColorChecker or QPCard.

Most raw converters I have tested can use DNG Colour Profiles even on non DNG raw files. Raw converters I have tested with DNG Profiles are Lightroom, ACR, Raw Therapee, AccuRaw and Iridient Raw Developer. Capture One is a notable exception.

This article covers some of that ground: http://echophoto.dnsalias.net/ekr/index.php/photoarticles/79-p45-colour-rendition

Sandy McGuffog, the author of AccuRaw and some other nifty softwares, pointed to this interesting article in his blog:
https://www.hansvaneijsden.com/colorchecker-perfect-skin-colors/ worth reading.

Personally, I have made Adobe DNG Profiles for all my cameras. On the Sony SLT 99 I mostly use Adobe Standard profile as it works fine for me. With the P45+ I use my own DNG Profiles.

As a side note, a ColorChecker card is not good enough for a full calibration, for that more stuff is needed (*).  

The reason Adobe DNG Profile Editor works is that they use ColorChecker to tweak a hpefully well designed base profile.

Best regards
Erik

(*) Something like this may give a proper starting point:
http://image-engineering-shop.de/shop/article_IE-camSPECS%2520express/camSPECS-express.html?sessid=jD6bM1hTL2TlDlB1AhLiLIit6AJVf3f7stolSdsx0jh1jstKDPOweX6p6nyy8OcJ&shop_param=cid%3D565%26aid%3DIE-camSPECS%2520express%26