Sony Colors

[iiiNelson]

... and with all that said we are all free to agree to disagree. I just am losing the energy or the will to continually validate what I’ve observed for myself (and IMO that seemingly is being verified through independent testing) objectively when I was looking to move to other systems. When I was preparing to move to Fuji (XH1/GFX combo) earlier this year it specifically was due to the color science which could get me to an “acceptably finished product” faster than Sony... it wasn’t due to color accuracy.

 

[k-hawinkler]

Interesting thread. Thanks.

 

[JoelM]

Who cares about accurate colors? This isn't film. We have a choice in post production to tweak to what we prefer. I want it close to what I perceive as pleasing. I haven't really come across a camera that was crap, except some in the "early" days when it came to skin tones. Hell, we don't even all experience blue the same way because what we see is more than physical. If you're using a color checker, then I think you lost the point.

My 2 cents,

Joel

 

[pegelli]

Interesting thread, thanks for starting it :thumbup:

There was an article on Luminous Landscape a few years ago in which a shot was brought as close to fully accurate as possible by a profile that perfectly matched all colours on a Gretag-McBeth colour checker (including all the grey patches).

It looked very dull and washed out. If I remember correctly the skin tone of the person in the shot seemed to represent someone who had just thrown up

Point was that accurate colour is an interesting exercise, but for a pleasing picture a lot more processing is needed.

So I think this is in agreement what most people in this thread are saying.

 

[iiiNelson]

Who cares about accurate colors? This isn't film. We have a choice in post production to tweak to what we prefer. I want it close to what I perceive as pleasing. I haven't really come across a camera that was crap, except some in the "early" days when it came to skin tones. Hell, we don't even all experience blue the same way because what we see is more than physical. If you're using a color checker, then I think you lost the point.

My 2 cents,

Joel

Color checkers are for the camera to reproduce the conditions based on reflected light. AWB systems of any manufacturer can be “tricked” to give a different value based on a variety of factors... what color checkers provide are a constant and consistent standard “baseline” if you will.

Color reproduction accuracy is mostly important for consistency to some (assuming a person gets as much “right” in camera). I’d agree that it’s not the end all and be all for final output as many of us have stated.

 

[jerome_m]

What about the "creative style" settings? On my camera, it allows to change the colours to look different.

 

[iiiNelson]

What about the "creative style" settings? On my camera, it allows to change the colours to look different.

Yeah there’s both the creative styles and the picture profiles to switch the looks and colors. I think they were testing the standard base settings but I’m unsure. I generally stick with standard profile and white balance or S-LOG/ Cine 3 or 4.

 

[jerome_m]

Yeah there’s both the creative styles and the picture profiles to switch the looks and colors. I think they were testing the standard base settings but I’m unsure. I generally stick with standard profile and white balance or S-LOG/ Cine 3 or 4.

But then, wouldn't it be more reasonable to compare the various "creative styles" between manufacturers? For example, if one is interested in skin colour, to compare the portrait modes of Canon, Nikon, Sony?

 

[iiiNelson]

But then, wouldn't it be more reasonable to compare the various "creative styles" between manufacturers? For example, if one is interested in skin colour, to compare the portrait modes of Canon, Nikon, Sony?

Probably not. All cameras don’t have the same creative style even amongst the same brand. Creative styles generally aren’t for the manufacturers most accurate color either. I think Fuji is the only one that uses film stock/creative styles as a base look with their Provia profile but everyone else tends to just have a standard profile as base.

 

[tofa]

For fun I took some comparison pics yesterday with an X-Pro 2 and an RX1 RII. I'll add a couple to this thread, though there's nothing particularly scientific about the comparison.
There are definitely some differences, but to me they aren't that far off:

Sony


Fuji


Sony


Fuji

 

[iiiNelson]

Yeah they look somewhat similar with the Fuji having increased saturation.

 

[k-hawinkler]

Yah they look somewhat similar with the Fuji having increased saturation.

The Sony colors look better, more natural to me. :clap:

 

[iiiNelson]

The Sony colors look better, more natural to me. :clap:

Me too in this particular example. The sky, the foliage, and the dirt look more natural IMO. Looks like the red channel is beginning to bleed into the yellow channel in the dirt and looks like cyan is beginning to bleed just a bit into the blue channel within the sky on the Fuji shots. The color on either is still pleasing but the Sony seems a bit more accurate.

 

[tofa]

These were processed, the Fuji probably a bit more than the Sony.
Out of the camera the Fuji raws were, as mentioned above, more saturated, also a bit warmer with a slight reddish tint vs. the Sony. Or one could say the Sonys were slightly lass saturated or flatter, with a cooler, yellowish/greenish tint compared to the Fujis. The Sonys are probably more 'accurate'.

 

[k-hawinkler]

These were processed, the Fuji probably a bit more than the Sony.
Out of the camera the Fuji raws were, as mentioned above, more saturated, also a bit warmer with a slight reddish tint vs. the Sony. Or one could say the Sonys were slightly lass saturated or flatter, with a cooler, yellowish/greenish tint compared to the Fujis. The Sonys are probably more 'accurate'.

:thumbs: :grin:

 

[Knorp]

The Sony colors look better, more natural to me. :clap:

+1 the Sony colours (as shown here) are absolutely better and IMO more accurate.

 

[tofa]

Okay, last comparisons, I promise. I did these because I'm getting ready for a big sell-off - Fuji bodies and lenses, Sigma, Nikon, Pentax, accessories, etc., maybe Sony . . . Wanted to see the actual difference in IQ between RX2 and XP2 with a decent lens (23/2 in this case, to match the RX2 as closely as possible; don't have the 23 1.4). Trying to decide what to sell, what (if anything) to keep. The more I sell, the more $ I'll have to upgrade . . . One reason for the tests is that the XP2 is such a joy to shoot - I'd forgotten about that in the last year or two I've been with the RX2.

On these two images I tried to match the colors more closely. What I'm really interested in is the quality of the image: the roundness, detail, three-dimensionality (not as in-focus/out of focus, but in the look itself: how do the twigs look, the leaves, the near objects vs the far, etc.). Obviously the RX2 has more resolution, but I make more books and calendars than 30x20 prints, so how is the look itself? Is the XP2 too much of a sketch camera compared to the RX2 (or to an A7rIII if if I go that route)?

Anyway, one thing I'm noticing in these, especially the first one, is the DoF. The conventional wisdom is that APSC gives about a step more DoF, so I shot the Fuji at 5.6 and the Sony at 8 - figured that would give about the same DoF. Check the trees and power tower in the 1st pic - this is clearly not the case. The Fuji has much narrower DoF then the Sony. Another truism bites the dust for me.

Sony


Fuji


Sony


Fuji

 

[PeterA]

Funny thread tbh.

 

[seb]

tofa, to your DOF-issue:
Is the Fuji really 23mm and the sony 35? Where was the focus point?
And it looks like a good mixture of field curve, resolution, sharpness (to the corners) on top.

Back to the colours. Just to bring it in again: If you use RAW files a colour comparison like this or the others within this thread are only about the profile. It has nothing to do with the capability of the camera to measure light nor is it a choice of Sony how the colours look like.

There may be two exceptions:

• The photosites on the sensor can't measure a certain spectrum. Sensors are not perfect to measure visible light (in numbers it's the sensitivity metamerism index). The SMI does not tell you where the sensitivity is not perfect but how good it is in general. Within profiling, it's possible to check the "where" while you look at how good the sensor measure certain colours in comparison to others. I think we would know from somewhere if there is a visible issue with one or another camera.
• Adobe measure colours in OOC-JPGs to create their default profiles for each camera. This is a possible solution to get the brand-look into their default LR-profiles. But as we can create and use own profiles this exception can be negated too.

Therefore IF you sell your gear because of the colours it's simply the wrong solution to solve your problem.

For a good friend, I made recently a summary of colours and dependencies around the processing of RAW. It's very basic and it's rather a summary to get deeper into certain things within a discussion. But maybe one or another is interested in it too.

 

[iiiNelson]

Okay, last comparisons, I promise. I did these because I'm getting ready for a big sell-off - Fuji bodies and lenses, Sigma, Nikon, Pentax, accessories, etc., maybe Sony . . . Wanted to see the actual difference in IQ between RX2 and XP2 with a decent lens (23/2 in this case, to match the RX2 as closely as possible; don't have the 23 1.4). Trying to decide what to sell, what (if anything) to keep. The more I sell, the more $ I'll have to upgrade . . . One reason for the tests is that the XP2 is such a joy to shoot - I'd forgotten about that in the last year or two I've been with the RX2.

On these two images I tried to match the colors more closely. What I'm really interested in is the quality of the image: the roundness, detail, three-dimensionality (not as in-focus/out of focus, but in the look itself: how do the twigs look, the leaves, the near objects vs the far, etc.). Obviously the RX2 has more resolution, but I make more books and calendars than 30x20 prints, so how is the look itself? Is the XP2 too much of a sketch camera compared to the RX2 (or to an A7rIII if if I go that route)?

Anyway, one thing I'm noticing in these, especially the first one, is the DoF. The conventional wisdom is that APSC gives about a step more DoF, so I shot the Fuji at 5.6 and the Sony at 8 - figured that would give about the same DoF. Check the trees and power tower in the 1st pic - this is clearly not the case. The Fuji has much narrower DoF then the Sony. Another truism bites the dust for me.

Sony


Fuji


Sony


Fuji

FOV looks different on both pictures but is more obvious on the top set where the Fuji looks wider. I think there’s a case to be made for either. You’ll have more detail with the Sony and more flexibility with the Fuji. If it were me I’d go for flexibility. The Fuji colors have a good reputation for a reason and you could make either camera work for you.