Chris T Loc: from England across the pond to New England

Stumbled across this report in my travels, so, thought it might be a great idea to pass it on to you all. I have created a list to simplify things:

The 12** Best Cameras for Color Reproduction* - Ranked (Prices are MSRP for Body Only, except where noted - some prices rounded off)
2-way tie
#1) Fuji XT-3 - 26mp APS-C MILC $1500
#1) Sony a7 III - 24mp FF MILC $1998
3-way tie
#2) Sony a7R III - 42MP FF MILC w/ 5-axis IBIS $3200
#2) Fuji X-E3 - 24mp APS-C MILC Compact $900
#2) Leica CL - 24mp APS-C MILC Compact $2800
--------
#3) Sony a9 - 24mp FF MILC w/ 5-axis IBIS $4500
2-way tie
#4) Fuji X-H1 - 24mp APS-C MILC w/ 5-axis IBIS $1899
#4) Panasonic Lumix GH5S - 10mp MFT MILC $2500
2-way tie
#5) Nikon D850 - 46mp FF DSLR - $3300
#5) Canon EOS M50 - 24mp APS-C MILC - $779
2-way tie
#6) Pentax K-1 Mk. II - 36mp FF DSLR w/ 5-axis IBIS - $1900
#6) Panasonic Lumix LX100 II - 17mp MFT Mirror-less Compact - $997 (w/ fixed 24-75)
2-way tie
#7) Canon EOS R - 20MP FF MILC - $2300
#7) Sony RX100 Mk. 6 - 20MP 1" Mirror-less Compact - $1198 (w/ fixed 24-200)
2-way tie
#8) Canon EOS 6D Mk. II - 26mp FF DSLR - $1999
#8) Nikon Z6 - 24mp FF MILC w/ 5-axis IBIS - $1996
2-way tie
#9) Panasonic Lumix G9 - 20MP MFT MILC w/ 5-axis IBIS - $1697
#9) Leica M-10-P - 24mp FF Digital Rangefinder - $7995
#10) Olympus E-M10 Mk. III - 16MP MFT MILC w/ 5-axis IBIS - $649
#11) Nikon Z-7 - 46MP FF MILC w/ 5-axis IBIS - $3396
#12) Fuji X-T100 - 24mp APS-C MILC - $599
--------------------------------------------------
** 21, actually

*according to Image Engineering
https://www.pdnonline.com/gear/cameras/the-best-cameras-for-color-reproduction-ranked/?utm_source=SilverpopMailing&utm_medium=email&utm_campaign=PDN%20Newswire%20Newsletter%20112918%20(1)&utm_content=

Note: only three DSLRs appear here - the Canon EOS 6D Mk. II, the Nikon D850, and the Pentax K-1 Mk. II ... the rest are mostly MILCs - which refers to Mirror-less Interchangeable Lens Cameras, except for the two which have fixed lenses (also Mirror-less designs.)


Legend -
MFT - Micro Four Thirds (sensor approximates 110 film)
IBIS - Internal Body Image Stabilization - 5-axis includes stabilization for up/down movement, as well as lateral
MILC - Mirror-less Interchangeable Lens Camera (as noted)
Compact - in this case, refers to a flat top - which may, or may not - reflect lack of a VF, which means - pixel-peeping on the LCD.

NMGal Loc: NE NM

The MILCs got a good showing.

Chris T Loc: from England across the pond to New England

Yes, I guess!!!!

Of course - they're all broken up into completely different categories - just six in the FF Mirror-less category - the three Sony models, the two new Nikon models, and the new Canon model. The rest are pretty evenly divided by Fuji APS-C models, Oly/Panny MFT models (quite a surprise!) and the odd compacts, and the Leica RF offering (the most expensive, by far) ... the real surprise - of course - is the vast difference in the price range - showing you can pick up a pretty good little dig cam for a few hundred bucks, and you don't have to mortgage your house to buy one!!!!!

User ID

`

If you know anything about color separation
then you know none of that matters at all.


.

Chris T Loc: from England across the pond to New England

Hey! ... I just made up the list, according to the tests done by Image Engineering (as per link quoted) ...

I'm not showing any favoritism, here ... although, they might have - missing are the EOS 1Dx Mk. II, the 5Ds / R, the 5D Mk. IV, and the D5!

Also, the only Crop-Sensor dig cams represented here, are the M50 and Fujis. There certainly does seem to be a lack of enthusiasm there - huh?

I think, User ID - the folks who put this together - were just trying to point out you can get decent color, at - virtually - any price!!!!!

User ID

I believe you're mistaken. I replied upon reading
"color reproduction" .... meaning publishing and
printing of color images. In color reproduction the
image is deconstructed in CMYK [not RGB] and is
then reconstituted as "4-color reproduction". So it
doesn't matter how "accurate" the original is, so
long as all necessary colors are represented. That
the colors may be out of balance, or muted, or in
any other way "incorrect" will be "made good" by
the color separation process.

You DID say in the title this was about magazine
and book publishing color reproduction. That IS
what I addressed in my reply ! Most of what I
shot for a living is for 4-color separation.

Digital image editing software does very much the
same thing, except it works in RGB. One ought to
avoid confusing "color reproduction" with the color
RENDERING seen in a SOOC image. IOW, cameras
*render* color, color separators *reproduce* color.

Below is a SOOC image plus an edited version. My
cameras are set to produce low saturation, muted,
low contrast SOOC results. That is how I record all
subjects. This captures as much data as possible
in the shadows and highlights, providing files most
suitable the editing process. [The "SOOC" is not a
"pure" SOOC cuz I hadda downsize it to post it.]
The process of editing for direct viewing is akin to
the color reproduction process, thus the examples.

Point is that I would NOT need to submit the edited
image to the color separator. The SOOC has all of it
within itself, and the separator will render the image
according to editorial expectations. If I'd submit the
edited image, I'd limit their options.

I'm not convinced that you knew what the survey or
report was actually reporting about, colorwise.

Anywho, don't forget to click "download" to properly
view and compare the pair of example images.

(Download)


(Download)

Chris T Loc: from England across the pond to New England

THIS - is the headline, to which I responded, UserID ... and I made up the list according to the ranks given by the writer -

The Best Cameras for Color Reproduction, Ranked

June 25, 2018

By Greg Scoblete

Of course, when I started reading the text, I'd no idea what to expect ... I thought it would compare all the highest rated ones, but, it did not!

User ID

`

I read the linked PDF. Cannot figger out WHAT they
are viewing to pass judgement ? Prints ? Screens ?
JPEGS ? Can't be viewing RAW. Seems very heavy
on science and consistency and very lightweight on
relevance or usefulness :-( We could shoot video
of "a day at the lab" and peddle it under the title of
"Geeks Gone Wild".

Don't FW updates sometimes alter the JPEG engine
in cameras ? None of this makes any sense at all.


`

Chris T Loc: from England across the pond to New England

Then I suggest you contact the author - Greg Scoblete, the lab who actually made the comparisons: Image Engineering, or the Newsletter -
PDN Photo District News ...

All I did was make up the list, plus - added a short description of each device, and then added the going price, looked up in the B&H Catalog.

After spending a couple of hours, doing that, User ID, I thought it might be worth all my effort, if I shared the result ...

I'd no idea it would create all this kerfuffle ... but, now that it has - it needs to be pointed out my limited responsibility in the matter ...

I must admit - it's interesting to see one of the newest cameras on the block - the Fuji XT-3 (a 26MP APS-C device) ... made #1 ...

Also, although I did NOT mention it earlier, I read somewhere that publishers require photos taken with 25+MP sensor cameras.

That means, the XT-3 is in-like-Flint, as is the slightly older Canon EOS 6D Mk. II (also 26MP, but FF) and, of course - anything with more res.

BebuLamar

Chris T Loc: from England across the pond to New England

This is at the beginning of the report - details the chart and color analyzer used -

"How Image Engineering Tests for Color

Image Engineering sets the camera color space to sRGB using the factory default for a color profile.

To measure color reproduction, they use color patches within the TE42 test target. The colors in these patches have been selected with reference to the well-known X-Rite ColorChecker SG color target. Each of the 96 color patches is individually measured using a calibrated spectrophotometer. Reference data is provided to the analysis software along with the image being tested. With the knowledge of the RGB color space (the commonly used sRGB), they convert the RGB data to the CIE-LAB color space. Based on the LAB color space, they calculate color error and differences in brightness, saturation and color tone.

Each camera test produces a chart which shows color deviations (Delta E). Red cells indicate that the camera produced a strong deviation from the reference color, light green represents a “noticeable” deviation and dark green represents a moderate. For our list, we tallied the total number of noticeable and strong deviations for each camera. Strong deviations were weighted double and the list was generated accordingly. The actual color data, plus the deviations, is included in each entry."

User ID

`

You just posted exerpts of the doc that I read.

In what way is that answering the question I had
raised about the testing ? I will REPEAT the same
question:

What is it that they're viewing ? Print ? Screen ?

Plz don't cut-n-paste. If you found any answer, I'd
like to know what they looked at for a comparison
to the original subject/color-chart. I just can't see
how a print would suffice, too many variables. And
I can't see how a screen image tell them suitability
of any camera's output for publishing 4-color repro
on paper.

There's no hidden agenda, brand-wise, on my part.
I use 5 brands, 3 formats, various models. IOW no
skin in this game.


`

Chris T Loc: from England across the pond to New England

Since I had absolutely nothing to do with the testing, whatsoever, I fail to see why you're insisting I should know the answer to your question.

I also use multiple brands - all four who make both FF and APS-C models - Nikon, Canon, Sony and Pentax, plus a Pentax MF Film ... plus FF 35mm film cameras, made by both Nikon and Canon ... plus Bridges made by all of the aforementioned, plus Fuji, Panasonic and Olympus ... so, I have every major brand covered, too (except for Leica) ...

Plus, I've been in this "game" just short of 50 years ... and am equally baffled by the supposed science behind this comparison ...

In fact, it occurred to me - a bunch of college sophomores* decided to pool together all the dig cams they owned, to see how they checked out!!!

There is a note further down - something to the effect they will add more cameras, as they go along ... that tell you anything, User ID?

*Perhaps, that'd explain the omission of cameras costing upwards of $4G (except for the Leica M10-P, but they probably got that from a parent!)

Largobob

I've been a 'science guy' all my professional life....through training and experience. In order to satisfy my scientific curiosity, I generally prefer seeing a description of the methodology used (how the tests were preformed, including the equipment used) and the corresponding raw data. (Were there sufficient replicates collected/used to eliminate random error?). Been at this research game long enough to know it is possible to insert bias into any result, based on the chosen experimental design.

I seldom rely on the summary or conclusions drawn from any unknown group or individual...until I can see the data myself. I have little/no experience with most of the cameras on the list...so I make no judgements here. I do wonder about the conspicuous absence of proven industry leaders from the list. Coincidence? Bias? Something else?

jerryc41 Loc: Catskill Mts of NY

That's an interesting rating/ranking. Post processing can do so much. I can't imagine anyone buying a certain camera on the basis of this listing alone. I guess it's just one more piece of information to help make a decision.

If I was going to buy a mirrorless, I would get the Fuji XT-3, so I'm glad to see it at the top of the list.