SD Q Colour

Started 3 months ago | Discussions
Macrae Regular Member • Posts: 110
SD Q Colour
6

There has been some celebrating SDQ colour recently. As I have said in the past that this has never been my experience. That said, and with a friend trying to get a handle on his new SDQ, I attempted once again to discover what I must be doing wrong. Seems thirteen years experience with Sigma and working daily with the cameras and having worked professionally for 33 years and my own studio for 25+ years just isn't enough,

So instead of beginning with my own test, I grabbed the x3f file DPR created when they "reviewed" the SDQ h. I also grabbed a few other files for comparison. It does not help that DPR cannot light their test scene target evenly. In spite of that, I worked with the x3f file in SPP6.7 and tried to match patches 19 and 24 to DPR's benchmark PhaseOne exposure, and white balanced off of patch #22, N5 as this helped CC the midtones and is the value the SPP exposure dropper is calibrated to. I also tried to balance the RED patch for saturation and density. The numbers are not exact but they are very close. I worked the Hasselblad file in the latest version of their SW, Phocus.

I found the BLUE patch from the SDQ h in the DPR exposure suspect, so I set up my own scene in the studio and shot it with my SDQ. The post production - tedious as it was- was minimal, again trying to match 19 and 24 patches and white balancing the SDQ from patch #22 and the studio standard camera on #23. The gray card right is an old Eastman Kodak and is not CC.

As I have said in the past, colour in not unlike music, it has to be in "balance" to be beautiful and harmonious. Below, you will see how the colour from the SDQ h/ SDQ sensor is out of balance in a way that cannot be reconciled. RED. RED. RED. And yet the RED patch is within spec. The wooden table in the studio scene is brown going green not red. The hair in the DPR scene is brown - not red, and the picture of the black man is over saturated in a strange way, but the MBCC is not.

Look at the files in the gallery view or do your own tests for confirmation. If you are honest however, the pictures posted here often betray these same problems especially if they contain earth tones with RED.

Sadly, I find the SDQ to be the only unusable camera I have ever owned for the reasons listed above and illustrated below. When it is time to change my equipment, I cannot imagine buying into Sigma again. When compared to Hasselblad, Pentax, PhaseOne and Fuji medium format cameras, the SDQ/h just can't compete with the subtle smooth beautiful accurate colour they produce. Would that the SDQ could produce an image like the landscape by Steve Pereira at the bottom.

Sigma sd Quattro H

Hasselblad X1D-50c

Sigma sd Quattro

Studio Standard accurate to scene

Steve Pereira Pentax 645D

Pentax 645D Sigma sd Quattro Sigma sd Quattro H
If you believe there are incorrect tags, please send us this post using our feedback form.
absquatulate Forum Pro • Posts: 11,121
Re: SD Q Colour
3

I find this quite a bizarre post to be honest, I'm not quite sure of the rationale, but here's my thoughts on it. Firstly, colour is subjective, we all see colour slightly differently, so there is no such thing as perfect colour, or even accurate colour in reality. To illustrate the point I find Fuji colour not very pleasant at all, garish greens and oranges, too much cyan in blues, it's just not pleasant to my eyes at all, and it's one of the reasons I ditched the XT-1 and left Fuji. I find Canon and Olympus colour excellent, and Panasonic used to be poor but they seem to have improved in their later cameras.

My experience of the SDQ is that it produces extremely pleasing colour, and in particular great tonal variations in colour, that bayer cameras generally aren't capable of. There are some excellent photographers here who produce really nice work and have no issues with colour, nor apparently do the viewing audience notice any judging by the comments.

I'm not disputing your background in Photography, however I do dispute your contention that the SDQ has bad colour science. Your example image does not prove anything, and isn't even a good example to my eyes. There is nothing in that image that an SDQ couldn't do, it's not even a very compelling image to be honest, it certainly doesn't make any kind of case for "better" colour, whatever that is.

The only takeaway I get from your post is that you don't like the SDQ's colour in the same way I don't like Fuji colour, that's literally it.

 absquatulate's gear list:absquatulate's gear list
Fujifilm FinePix X100 Nikon Coolpix P7800 Sony Cyber-shot DSC-R1 Panasonic Lumix DC-G9 Sigma sd Quattro +4 more
DMillier Forum Pro • Posts: 21,520
Re: SD Q Colour
5

I don't understand your claim that colour is subjective and that we all see it differently. There are umpteen colour reference charts and tools. It's trivial to test whether colour is accurate from a camera: you photograph one of these tools and then measure the deviation in different patches. There is nothing subjective about it.

Perhaps you mean that colour preference is subjective?

There is certainly little evidence that viewers of photographs prefer accurate colour. Quite the opposite. Most people prefer some kind of romanticised colour - whether it be higher saturation, warmer hues or whatever. This preference for deviations from accurate colour is clearly individual and subjective, hence the different deviations produced by different brands.

I have no training in colour science but I learnt a bit recently in correspondence with the chief developer at Datacolor. In particular, I learned a bit how to make colour correction profiles better. It turns out the profiles are generated by a process of measurement and interpolation: a colorimeter measures a fixed number of colour patches across the spectrum and assumes a device/inkset is well behaved in between the measured patches. But not everything is - for example the particular inkset I was having trouble with turned out to have odd non-linearities and strange crossover points between channels. This made standard profiling not particularly effective.

The advice was to measure more patches: instead of 200 patches, measure 1000. Colorimeters ship with simplified colour patch charts and more complex ones. Often the simple charts are sufficient. However, the larger charts' extra data points give the profiling algorithm more to work with. Obvious really. The point being that there is nothing subjective about colour - it's simply a matching/correction process.

With respect to Foveon colour, I have never used a Quattro model. My sense from nothing more sophisticated than looking at published Q images is that the colour is improved over previous models. It seems more even, more natural, more Bayer-like.

All my Sigma cameras produce inconsistent colour: sometime it looks nice, sometimes it looks damn odd. Roland (Kalpanika) mentioned a number of times that Foveon colour can be troublesome. Because of the translation matrix from the captured data to RGB, sometimes certain colours go out of bounds and weird things happen. Foveon may or may not have greater colour discrimination but it certainly has the ability to go wrong on occasions. Sigma are forever tweaking and playing around with colour from model to model in order to arrive at an acceptable compromise. The Q models (to my eyes) look more natural but that is not to say they are necessarily accurate.

Bayer colours are subject to tweaks by each of the manufacturers and anyone can make their own custom raw profile (I've made loads - even for my Sigma cameras) so colour can be almost anything but the basic capture does seem to be more predictable and easier to work with.

-- hide signature --
absquatulate Forum Pro • Posts: 11,121
Re: SD Q Colour

DMillier wrote:

I don't understand your claim that colour is subjective and that we all see it differently.

Hopefully this will help.

https://www.livescience.com/21275-color-red-blue-scientists.html

There are umpteen colour reference charts and tools. It's trivial to test whether colour is accurate from a camera: you photograph one of these tools and then measure the deviation in different patches. There is nothing subjective about it.

Perhaps you mean that colour preference is subjective?

We all see it slightly differently and we have our own subjective tastes, the two are not mutually exclusive.

There is certainly little evidence that viewers of photographs prefer accurate colour. Quite the opposite. Most people prefer some kind of romanticised colour - whether it be higher saturation, warmer hues or whatever. This preference for deviations from accurate colour is clearly individual and subjective, hence the different deviations produced by different brands.

I have no training in colour science but I learnt a bit recently in correspondence with the chief developer at Datacolor. In particular, I learned a bit how to make colour correction profiles better. It turns out the profiles are generated by a process of measurement and interpolation: a colorimeter measures a fixed number of colour patches across the spectrum and assumes a device/inkset is well behaved in between the measured patches. But not everything is - for example the particular inkset I was having trouble with turned out to have odd non-linearities and strange crossover points between channels. This made standard profiling not particularly effective.

The advice was to measure more patches: instead of 200 patches, measure 1000. Colorimeters ship with simplified colour patch charts and more complex ones. Often the simple charts are sufficient. However, the larger charts' extra data points give the profiling algorithm more to work with. Obvious really. The point being that there is nothing subjective about colour - it's simply a matching/correction process.

With respect to Foveon colour, I have never used a Quattro model. My sense from nothing more sophisticated than looking at published Q images is that the colour is improved over previous models. It seems more even, more natural, more Bayer-like.

All my Sigma cameras produce inconsistent colour: sometime it looks nice, sometimes it looks damn odd. Roland (Kalpanika) mentioned a number of times that Foveon colour can be troublesome. Because of the translation matrix from the captured data to RGB, sometimes certain colours go out of bounds and weird things happen. Foveon may or may not have greater colour discrimination but it certainly has the ability to go wrong on occasions. Sigma are forever tweaking and playing around with colour from model to model in order to arrive at an acceptable compromise. The Q models (to my eyes) look more natural but that is not to say they are necessarily accurate.

Bayer colours are subject to tweaks by each of the manufacturers and anyone can make their own custom raw profile (I've made loads - even for my Sigma cameras) so colour can be almost anything but the basic capture does seem to be more predictable and easier to work with.

I disagree on this last paragraph, the SD Quattro records colour the most natural of any camera I've used, your opinion isn't really credible on this considering you're basing it on images you see on the internet. If you'd used one it would carry more weight, but I suspect even then it would be a contrarian position.

 absquatulate's gear list:absquatulate's gear list
Fujifilm FinePix X100 Nikon Coolpix P7800 Sony Cyber-shot DSC-R1 Panasonic Lumix DC-G9 Sigma sd Quattro +4 more
richard stone Veteran Member • Posts: 3,405
Re: SD Q Colour

Seriously, lots of people use the SDQ cameras and are happy with them, color-wise. Perfect? Probably not. In my view the colors in digital images can be "tweaked" so much that "accurate" color is just one possibility.

I do not entirely agree that it is ALL subjective, BUT: it is about using a camera that produces colors that you like. So use some other kind of camera, not the Q, I know of people who still believe that the SD14 was the best ever.

Although I am sorry to hear of your frustration and disappointment in regard to this issue, I like the Sigma Q cameras and colors, and my understanding from tests that have been done by various forum participants is that the Q colors are quite accurate.

-- hide signature --
 richard stone's gear list:richard stone's gear list
Sigma sd Quattro Sigma SD10 Sigma 17-50mm F2.8 EX DC OS HSM Sigma 30mm F1.4 DC HSM Art
Sjak
Sjak Senior Member • Posts: 5,147
Re: SD Q Colour
4

It's OK if you don't like the output of a camera, or are not able to get it to an image you like, or to emulate the output of another camera with SDQ-files.

As implied by yourself, with terms such as "harmonious", "beautiful",... colour is indeed a subjective matter, and it can happen we simply do not like files from some cameras.

Besides, harmony is not the only approach in music: dissonant, distortion, etc also work, and there is no reason it cannot work in photography. The question at hand is rather if it works for you; apparently not, and that's fine.

Also, you do realize you compare a 750-euro cam to models costing 5-10k or more? 

At least, you can't blame yourself for not trying to like the Quattro

 Sjak's gear list:Sjak's gear list
Panasonic Lumix DMC-LX7 Sigma DP2 Sigma DP1 Merrill Fujifilm X-T1 Pentax K100D +3 more
xpatUSA
xpatUSA Forum Pro • Posts: 16,459
Re: SD Q Colour

Macrae wrote:

There has been some celebrating SDQ colour recently. As I have said in the past that this has never been my experience. That said, and with a friend trying to get a handle on his new SDQ, I attempted once again to discover what I must be doing wrong. Seems thirteen years experience with Sigma and working daily with the cameras and having worked professionally for 33 years and my own studio for 25+ years just isn't enough,

Sigma sd Quattro H

Sigma sd Quattro

There is a lack of numerical information in this thread but there is a lot of SPV*.

To redress the balance as regards reds:

My X-rite Card data sheet, CIELAB numbers:

L* = 42.101; a* = 53.378; b* = 28.19

The posted DPR image open in RawTherapee 5.5:

L* = 41.2; a* = 55.6; b* = 37.9 ... the b* axis value is unacceptable

The OP's test shot open in RawTherapee 5.5:

L* = 39.4; a* = 43.2; b* = 27.6 ... the a* axis value is unacceptable

A sample of two Q test shots out of the many published does not inspire much confidence, statistically speaking.

-- hide signature --

* SPV = Standard Photographic Vague
Ted

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2 Sigma DP2 Merrill Panasonic Lumix DMC-LX7 Panasonic Lumix DMC-G1 Panasonic Lumix DMC-GH1 +17 more
D Cox Forum Pro • Posts: 25,216
Re: SD Q Colour
9

The faces in the DPR test setup tell you nothing at all about a camera's reproduction of skin or hair colours, as they are prints made of mixtures of yellow, magenta and cyan dyes, not real skin and hair. The spectra are different.

D Cox Forum Pro • Posts: 25,216
Re: SD Q Colour

DMillier wrote:

I don't understand your claim that colour is subjective and that we all see it differently. There are umpteen colour reference charts and tools. It's trivial to test whether colour is accurate from a camera: you photograph one of these tools and then measure the deviation in different patches. There is nothing subjective about it.

It's perhaps an exaggeration to say that we all see colours differently, but there is some variation in the cone pigments between normal trichromats, grading into the anomalous trichromats (who have three types of cone but with very little difference between the medium and longer wave detectors), and the dichromats who have only two pigments.

Also some people have more medium detection cones while others have more long wave detection cones. How much this affects colour perception isn't clear, but it makes Bayer's use of two green pixels for each red one dubious.

Perhaps you mean that colour preference is subjective?

There is certainly little evidence that viewers of photographs prefer accurate colour. Quite the opposite. Most people prefer some kind of romanticised colour - whether it be higher saturation, warmer hues or whatever. This preference for deviations from accurate colour is clearly individual and subjective, hence the different deviations produced by different brands.

I have no training in colour science but I learnt a bit recently in correspondence with the chief developer at Datacolor. In particular, I learned a bit how to make colour correction profiles better. It turns out the profiles are generated by a process of measurement and interpolation: a colorimeter measures a fixed number of colour patches across the spectrum and assumes a device/inkset is well behaved in between the measured patches. But not everything is - for example the particular inkset I was having trouble with turned out to have odd non-linearities and strange crossover points between channels. This made standard profiling not particularly effective.

The advice was to measure more patches: instead of 200 patches, measure 1000. Colorimeters ship with simplified colour patch charts and more complex ones. Often the simple charts are sufficient. However, the larger charts' extra data points give the profiling algorithm more to work with. Obvious really. The point being that there is nothing subjective about colour - it's simply a matching/correction process.

With respect to Foveon colour, I have never used a Quattro model. My sense from nothing more sophisticated than looking at published Q images is that the colour is improved over previous models. It seems more even, more natural, more Bayer-like.

More natural and more Bayer-like are not the same thing at all.

All my Sigma cameras produce inconsistent colour: sometime it looks nice, sometimes it looks damn odd. Roland (Kalpanika) mentioned a number of times that Foveon colour can be troublesome. Because of the translation matrix from the captured data to RGB, sometimes certain colours go out of bounds and weird things happen. Foveon may or may not have greater colour discrimination but it certainly has the ability to go wrong on occasions. Sigma are forever tweaking and playing around with colour from model to model in order to arrive at an acceptable compromise. The Q models (to my eyes) look more natural but that is not to say they are necessarily accurate.

Bayer colours are subject to tweaks by each of the manufacturers and anyone can make their own custom raw profile (I've made loads - even for my Sigma cameras) so colour can be almost anything but the basic capture does seem to be more predictable and easier to work with.

My experience of the Quattro is that browns and dull reds can be too saturated, so that branches of trees come out orange rather than a warm brown, and bricks are too saturated. Often this improves the picture, but you can easily adjust the HSL sliders in ACR; or use the nice colour circle in Image Analyser (look for "Color Tone Adjustment" in the menu).

chankosuto Forum Member • Posts: 57
Re: SD Q Colour

Controversial topic, what I get from it, is that color can be scientifically measured by different means to find their respective wavelengths, however human perception can differ even by sex, if I'm not mistaken I've read woman can distinguish or experience color differently. Having said that, as of now I like the a color pattern, specially in autumn.

TN Args
TN Args Veteran Member • Posts: 9,009
Re: SD Q Colour
3

Macrae wrote:

There has been some celebrating SDQ colour recently. As I have said in the past that this has never been my experience. That said, and with a friend trying to get a handle on his new SDQ, I attempted once again to discover what I must be doing wrong. Seems thirteen years experience with Sigma and working daily with the cameras and having worked professionally for 33 years and my own studio for 25+ years just isn't enough,

So instead of beginning with my own test, I grabbed the x3f file DPR created when they "reviewed" the SDQ h. I also grabbed a few other files for comparison. It does not help that DPR cannot light their test scene target evenly. In spite of that, I worked with the x3f file in SPP6.7 and tried to match patches 19 and 24 to DPR's benchmark PhaseOne exposure, and white balanced off of patch #22, N5 as this helped CC the midtones and is the value the SPP exposure dropper is calibrated to. I also tried to balance the RED patch for saturation and density. The numbers are not exact but they are very close. I worked the Hasselblad file in the latest version of their SW, Phocus.

I found the BLUE patch from the SDQ h in the DPR exposure suspect, so I set up my own scene in the studio and shot it with my SDQ. The post production - tedious as it was- was minimal, again trying to match 19 and 24 patches and white balancing the SDQ from patch #22 and the studio standard camera on #23. The gray card right is an old Eastman Kodak and is not CC.

As I have said in the past, colour in not unlike music, it has to be in "balance" to be beautiful and harmonious. Below, you will see how the colour from the SDQ h/ SDQ sensor is out of balance in a way that cannot be reconciled. RED. RED. RED. And yet the RED patch is within spec. The wooden table in the studio scene is brown going green not red. The hair in the DPR scene is brown - not red, and the picture of the black man is over saturated in a strange way, but the MBCC is not.

Look at the files in the gallery view or do your own tests for confirmation. If you are honest however, the pictures posted here often betray these same problems especially if they contain earth tones with RED.

Sadly, I find the SDQ to be the only unusable camera I have ever owned for the reasons listed above and illustrated below. When it is time to change my equipment, I cannot imagine buying into Sigma again. When compared to Hasselblad, Pentax, PhaseOne and Fuji medium format cameras, the SDQ/h just can't compete with the subtle smooth beautiful accurate colour they produce. Would that the SDQ could produce an image like the landscape by Steve Pereira at the bottom.

I see what you are trying to do, but these sort of manual exercises often produce vexatious results. I am not surprised by your experience using the tribulation described above.

Use your own Macbeth card, make your own, evenly lit, test files, and see how they look.

Make X3F and DNG test files. Automated profiling tools will usually work with the latter, and sometimes work with TIFF files from X3F.

Here is a colour accuracy test result for my dp0 using SPP6.3.3, with no colour adjustments at all. Link. Your sdQ sensor should be very similar.

cheers

 TN Args's gear list:TN Args's gear list
Sigma dp0 Quattro Panasonic Lumix DMC-GX7 Olympus E-M5 II Panasonic Leica Summilux DG 25mm F1.4 Olympus M.Zuiko Digital ED 9-18mm F4.0-5.6 +7 more
DMillier Forum Pro • Posts: 21,520
Re: SD Q Colour
1

But apparently it is subjective.....

-- hide signature --
DMillier Forum Pro • Posts: 21,520
Re: SD Q Colour
2

But surely we need to test all 16 million colours?

-- hide signature --
DMillier Forum Pro • Posts: 21,520
Re: SD Q Colour

No one can say for sure what another person's perception of colour is inside their mind but what we can do is test people to see whether or not they agree certain colours are the same or not. If everyone agrees a certain hue of red (say fire engine red) is the same then it doesn't really matter what's going on internally: we have common reference points. Obviously someone who has colour blindness will disagree with someone who doesn't on some colours.

The point being really, that assuming we talking about the subset of people who can all agree that the colours on two identical colour charts are the same, the we would expect them to also agree whether a colour in a photo matched the chart or not. That's your baseline for accuracy.

Personal preference for colour is something else. Two people may agree on what a particular shade is while one may consider a more saturated version preferable, artistically. That's nothing to do with correctly identifying a colour only to do with which colours you like more. Hey, there are a lot of shades of 'white' paint available!

With respect to colours under a wide range of lighting conditions, I strongly suspect that older Foveons will misfire on accurate rendition of colour more often than Bayer sensors based on both my own experience and some of the arguments I read on these hallowed pages. I have plenty of examples of quirky colour from my Foveon based cameras. Bayer sensors do seem to me to be fairly consistent even with the different tweaks that are made as crowd pleasers.

As far as the Q is concerned - yes I haven't used one so I haven't assessed the colour of my own photographs. But I saw the big print tour images and I've seen enough of forum members' work to feel that the Q colour appears to be more neutral, consistent and like Bayer colour than some of the odd renditions I've seen from my older Sigma cameras. For example, my DP1 original would often produce images with generally attractive plausible looking colours across the frame but turn red objects strongly magenta.

ps

(No need for the personal dig or supposing what I would say in order to deliver judgement on it before I have even said anything. If you want to keep adding personal digs, I'll keep pointing them out).

-- hide signature --
Johan Borg Senior Member • Posts: 2,930
Re: SD Q Colour
1

Instead of trying to make every camera look identical, I prefer to let each play to its strengths. It's not just the Sigma that struggles with the adjustments above, the blue playing cards went all indigo in the Hasselblad file...

absquatulate Forum Pro • Posts: 11,121
Re: SD Q Colour
1

DMillier wrote:

No one can say for sure what another person's perception of colour is inside their mind but what we can do is test people to see whether or not they agree certain colours are the same or not. If everyone agrees a certain hue of red (say fire engine red) is the same then it doesn't really matter what's going on internally: we have common reference points. Obviously someone who has colour blindness will disagree with someone who doesn't on some colours.

The point being really, that assuming we talking about the subset of people who can all agree that the colours on two identical colour charts are the same, the we would expect them to also agree whether a colour in a photo matched the chart or not. That's your baseline for accuracy.

Personal preference for colour is something else. Two people may agree on what a particular shade is while one may consider a more saturated version preferable, artistically. That's nothing to do with correctly identifying a colour only to do with which colours you like more. Hey, there are a lot of shades of 'white' paint available!

With respect to colours under a wide range of lighting conditions, I strongly suspect that older Foveons will misfire on accurate rendition of colour more often than Bayer sensors based on both my own experience and some of the arguments I read on these hallowed pages. I have plenty of examples of quirky colour from my Foveon based cameras. Bayer sensors do seem to me to be fairly consistent even with the different tweaks that are made as crowd pleasers.

As far as the Q is concerned - yes I haven't used one so I haven't assessed the colour of my own photographs. But I saw the big print tour images and I've seen enough of forum members' work to feel that the Q colour appears to be more neutral, consistent and like Bayer colour than some of the odd renditions I've seen from my older Sigma cameras. For example, my DP1 original would often produce images with generally attractive plausible looking colours across the frame but turn red objects strongly magenta.

I gave you the opinion from scientists who research colour perception, it's quite clear they don't believe we all see colour the exact same way. So yes, we can mostly agree on whether something is blue or red, but there is a further subtlety to that distinction that can't be easily measured. It's the different shades of colours where opinion can start to differ, and I don't believe that's solely down to subjective taste. The distinction between something like orange and red can often lead to conflicting opinions on which is which for instance.

On a side note, I find that the SD Quattro sensor is very good at displaying subtle tonalities in dark and bright areas where the bayer sensor is not so accurate. The Merrill sensor is not so sensitive and often fails in shadow areas. In the blue hour the SD Quattro sensor is particularly good at rendering the light accurately, much more so than bayer sensors to my eyes, which tend to black rather than blue. The Quattro is able to record the subtle pastel tones in this kind of light really well.

ps

(No need for the personal dig or supposing what I would say in order to deliver judgement on it before I have even said anything. If you want to keep adding personal digs, I'll keep pointing them out).

Sure, just as I'll point out where you are consistently negative about something you allegedly like, a regular contradiction you adhere to, so I guess we both know where we stand.

 absquatulate's gear list:absquatulate's gear list
Fujifilm FinePix X100 Nikon Coolpix P7800 Sony Cyber-shot DSC-R1 Panasonic Lumix DC-G9 Sigma sd Quattro +4 more
DMillier Forum Pro • Posts: 21,520
Re: SD Q Colour
1

Hey, many say I am negative about everything.

I seem to recollect reading years ago that one of the differences between Foveon and Bayer is that Foveon preserves hue in highlights better than Bayer which tends to go white early. No idea whether I made that up. Even if my recollection about the claim is correct, I'm not sure whether it is even true.

The Merrill goes colourless in shadow regions but that appears to be a deliberate decision by Sigma in an attempt to hide the horrible big patchy splodges that lurk there.

-- hide signature --
xpatUSA
xpatUSA Forum Pro • Posts: 16,459
Re: SD Q Colour
1

absquatulate wrote:

DMillier wrote:

No one can say for sure what another person's perception of colour is inside their mind but what we can do is test people to see whether or not they agree certain colours are the same or not. If everyone agrees a certain hue of red (say fire engine red) is the same then it doesn't really matter what's going on internally: we have common reference points. Obviously someone who has colour blindness will disagree with someone who doesn't on some colours.

The point being really, that assuming we talking about the subset of people who can all agree that the colours on two identical colour charts are the same, the we would expect them to also agree whether a colour in a photo matched the chart or not. That's your baseline for accuracy.

Personal preference for colour is something else. Two people may agree on what a particular shade is while one may consider a more saturated version preferable, artistically. That's nothing to do with correctly identifying a colour only to do with which colours you like more. Hey, there are a lot of shades of 'white' paint available!

With respect to colours under a wide range of lighting conditions, I strongly suspect that older Foveons will misfire on accurate rendition of colour more often than Bayer sensors based on both my own experience and some of the arguments I read on these hallowed pages. I have plenty of examples of quirky colour from my Foveon based cameras. Bayer sensors do seem to me to be fairly consistent even with the different tweaks that are made as crowd pleasers.

As far as the Q is concerned - yes I haven't used one so I haven't assessed the colour of my own photographs. But I saw the big print tour images and I've seen enough of forum members' work to feel that the Q colour appears to be more neutral, consistent and like Bayer colour than some of the odd renditions I've seen from my older Sigma cameras. For example, my DP1 original would often produce images with generally attractive plausible looking colours across the frame but turn red objects strongly magenta.

I gave you the opinion from scientists who research colour perception, it's quite clear they don't believe we all see colour the exact same way. So yes, we can mostly agree on whether something is blue or red, but there is a further subtlety to that distinction that can't be easily measured. It's the different shades of colours where opinion can start to differ, and I don't believe that's solely down to subjective taste. The distinction between something like orange and red can often lead to conflicting opinions on which is which for instance.

On a side note, I find that the SD Quattro sensor is very good at displaying subtle tonalities in dark and bright areas where the bayer sensor is not so accurate. The Merrill sensor is not so sensitive and often fails in shadow areas. In the blue hour the SD Quattro sensor is particularly good at rendering the light accurately, much more so than bayer sensors to my eyes, which tend to black rather than blue. The Quattro is able to record the subtle pastel tones in this kind of light really well.

ps

(No need for the personal dig or supposing what I would say in order to deliver judgement on it before I have even said anything. If you want to keep adding personal digs, I'll keep pointing them out).

Sure, just as I'll point out where you are consistently negative about something you allegedly like, a regular contradiction you adhere to, so I guess we both know where we stand.

LOL gents

(takes a ringside seat, with a 64-oz Coke and a tub of popcorn ...)

... ... ...

-- hide signature --

Too many variables, not enough grey cells ...
Ted

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2 Sigma DP2 Merrill Panasonic Lumix DMC-LX7 Panasonic Lumix DMC-G1 Panasonic Lumix DMC-GH1 +17 more
D Cox Forum Pro • Posts: 25,216
Re: SD Q Colour

absquatulate wrote:

I gave you the opinion from scientists who research colour perception, it's quite clear they don't believe we all see colour the exact same way. So yes, we can mostly agree on whether something is blue or red, but there is a further subtlety to that distinction that can't be easily measured. It's the different shades of colours where opinion can start to differ, and I don't believe that's solely down to subjective taste. The distinction between something like orange and red can often lead to conflicting opinions on which is which for instance.

Often this is a disagreement over the names of colours rather than a difference of perception. One person's beige can be another person's khaki.

The XKCD survey is I think very interesting.

https://blog.xkcd.com/2010/05/03/color-survey-results/

On a side note, I find that the SD Quattro sensor is very good at displaying subtle tonalities in dark and bright areas where the bayer sensor is not so accurate. The Merrill sensor is not so sensitive and often fails in shadow areas. In the blue hour the SD Quattro sensor is particularly good at rendering the light accurately, much more so than bayer sensors to my eyes, which tend to black rather than blue. The Quattro is able to record the subtle pastel tones in this kind of light really well.

xpatUSA
xpatUSA Forum Pro • Posts: 16,459
Re: SD Q Colour

D Cox wrote:

absquatulate wrote:

I gave you the opinion from scientists who research colour perception

Often this is a disagreement over the names of colours rather than a difference of perception. One person's beige can be another person's khaki.

The XKCD survey is I think very interesting.

https://blog.xkcd.com/2010/05/03/color-survey-results/

Thanks! An absolutely excellent reference.

-- hide signature --

Too many variables, not enough grey cells ...
Ted

 xpatUSA's gear list:xpatUSA's gear list
Sigma DP2 Sigma DP2 Merrill Panasonic Lumix DMC-LX7 Panasonic Lumix DMC-G1 Panasonic Lumix DMC-GH1 +17 more
Keyboard shortcuts:
FForum MMy threads