AuntiPode wrote:
Interesting technique, but I'd be hard pressed to find a way to climb into deeper social doodoo than for a pakeha to dabble in faux mokos.
Found an issue with my PS setup on my proofing ... no wonder my colors have been off for a while. Try to ignore the weird red colors in the dark areas at bottom.
A Ta Moko is a Maori facial marking. To appropriate a traditional Maori moko would be considered highly offensive and disrespectful, although perhaps one can apply a Kirituhi (skin drawing), but it's a delicate line I'd suggest avoiding.
Kent, I see the leaves in your edit quite plainly on my monitor (NEC LCD2690WUXi) in Safari, FWIW. Oddly, they are less visible in Firefox on the same monitor.
Thanks Karen ... I found the problem on my end. My "proof colors" was on ... (CTRL Y), I must have tried to use a repeat command keystroke and inadvertently turned it on at some time in the past. This has been an issue for me for quite some time, but never has it been as obvious to me as it was on this one. Glad I finally figured it out.
As to the Safari vs. Firefox ... I can only guess that it is because my taking Bob's image from an sRGB space, assigning it to Prophoto RGB, re-working it and then converting it back to sRGB cause some gaps along the way that the different browser's aren't playing so nicely with.
Don't know much about Moari culture beyond a few documentaries I've seen but discussion of the their use of body language displays connected some dots with regard to how photographic poses imply aggressive / passive and masculine and feminine stereotypes via the angle of the body to the camera, angle of hips and shoulders and alignment and tilt of the head and eye line to the shoulder line.
In the 50s a portrait photographer named Joe Zeltsman realized those variables could be predictably controlled by simply placing the subject's feet and shifting weight to one foot or the other. He taught it to Zucker which is where I learned it.
FWIW - I was going for more of a Rambo cameo look with my experiment vs a bad mojo moko..
RustyBug wrote:
Found an issue with my PS setup on my proofing ... no wonder my colors have been off for a while. Try to ignore the weird red colors in the dark areas at bottom.
Hmm, I don't see them and non-calibrated monitor - FireFox. But then my color perception and sensitivity sucks too.
RustyBug wrote:
Thanks Karen ... I found the problem on my end. My "proof colors" was on ... (CTRL Y), I must have tried to use a repeat command keystroke and inadvertently turned it on at some time in the past. This has been an issue for me for quite some time, but never has it been as obvious to me as it was on this one. Glad I finally figured it out.
As to the Safari vs. Firefox ... I can only guess that it is because my taking Bob's image from an sRGB space, assigning it to Prophoto RGB, re-working it and then converting it back to sRGB cause some gaps along the way that the different browser's aren't playing so nicely with.
I'm confused Vincent Versace emphatically writes changing from sRGB to, let's say Adobe ProPhotoRGB is like adding a quart of water to a gallon bucket - you still only have a quart of water.
If that is the case, would you not only have the sRGB colors to work with?
Bob Jarman wrote:
I'm confused Vincent Versace emphatically writes changing from sRGB to, let's say Adobe ProPhotoRGB is like adding a quart of water to a gallon bucket - you still only have a quart of water.
Bob
Think cows and pasture. The bit level determines how many cows (discrete color values) there are. The color space defines the fences in the pasture space they graze in (sRGB being 20 acres and smaller than ProPhoto's 200).
In the bigger pasture the same number of cows, distributed equally, wind up with more grass between them so they will stand out more individually (easer to rope and manipulate) and the herd will look bigger perceptually than when the same number of cows are crowded into a smaller pasture.
The Catch-22 in all that is you never actually see what is actually happening in a larger editing spaces to colors outside the gamut of your monitor, you see a perceptual rendering of the larger space to fit the screen. The colors at the outer edges of the bigger gamut are remapped to the limits of the screen and all the other less saturated colors rearranged proportionately so you get an impression PERCEPTUALLY of how the bigger space is being manipulated.
So the printer gamut can't be reproduced accurately by the screen or sRBG editing space (which is based on CRT monitor gamut) but it does almost fit into AdobeRGB. AbobeRGB 1998 was created in the mid-90s when color reproduction was: transparency > drum scanner > halftone dots > Web Offset printing press. The gamut was designed to match the size and shape of SWOP CYMK.
SWOP = Standards for Web Offset Printing which defined the physical specs for ink hue, density, ink film thickness on the paper, and the color and reflectance of the paper.
I was dealing with color management starting back in the mid-70s as production manager at a web magazine printer when getting color separators and ad agencies to use SWOP standard inks and paper for press proofs was the only way to ensure the final production on the web press would match the proof. The worst offenders were ad agencies for cosmetic companies who would use non-standard reds in lipstick ads we could not match on production presses. If we tried and failed the agency, through the publisher, would demand a free reprint of the ad. It was a con game to get free advertising with the shops who did the separations and proofing on presses one generation removed from Gutenberg stacking the deck with exotic ink mixes.
We had a QA section which took the suppled film and made a SWOP standard Matchprint from it on paper matching the SWOP spec, then compared it with the supplied proofs. If they did not match we rejected the proof and sent ours to the ad agency for approval telling them that's what they could realistically expect the production press to produce and to either OK it or supply new film and proof. Bluff called they usually OK'd our proof. Eventually everyone in the supply chain saw the wisdom of proofing to SWOP, mainly because non-press methods were faster and cheaper than press proofing.
Soft proofing works the same way on a monitor. The press/printer profile is placed into the display path which allows the screen to simulate, more or less, how the file will change due to less saturated / lower contrast reproduction with ink and paper:
RGB file values > converted to CYMK printer gamut > rendered back into RGB perceptually to fit the monitor gamut per calibration profile.
Soft proofing was the Holy Grail in the 90s when ICC profile based color was introduced with grand predictions that it would replace hard copy proofs. But it never will for commercial work because monitor gamut. calibration and viewing environment affect perception of a screen image.
Instead ICC based color management is used commercially in printing to force a wider gamut ink jet to simulate with Absolute Colorimetric rendering what the smaller CYMK gamut of the production press can actually produce (as encoded in its profile used in the proofing process). That allows the ink jet printer to create fast, accurate to production hard copy proofs of 8-page press imposed layouts for review and signature approval.
In concept its the same cows and pasture game, with the pasture being defined not by the working space, monitor or wide-gamut ink jet but by the smaller gamut of the web press SWOP ink and paper.
So sRGB workflow will work fine for an all screen display workflow and even printing on photo printers like Costco, AbobeRGB still works OK if the print output is in a magazine printed offset. It's only when printing for final output on wider format 8 - 12 color ink jets that a wider working gamut like ProPhoto is needed.
But notice in the 3D wireframes that most of the colors fit inside all the gamuts. It is only the most saturated colors which are the problem requiring the perceptual reshuffling of the color perception deck via the profile guided remapping / perceptual or relative colorimetric rendering intent.
If a scene consists entirely of subtle gradients of pastel hues sRBG can usually reproduce the gradients smoother than the larger spaces, especially in 8-bit JPGs.
Again its matter of how the cows are distributed in the field, if you see wide gap between them the gradient over the herd doesn't look as smooth as when they are crowded together with no gaps between them. On a print the gaps are manifested by the banding seen in 8-bit color in flat tonal gradient like sky and reflections on cars. The banding is like if the brown cows sorted themselves into groups by color, heading for whichever brown shaded group was closest to their hide. The herd viewed as a whole from a distance would appear banded in rows of progressively browner cows.
Contrast created by banding (posterization) can also create the optical illusion of sharpness and texture: how silkscreen art and USM works to trick the brain. Shifting color spaces around can also be used to create other than normal contrast between colors and change the appearance of detail in images as Rusty is doing.
Color management is like a flooding river. It looks so smooth and safe on the surface but if you foolishly dive in you usually drown in confusion very quickly.
Monetarily is is like a whirlpool that will suck money out of your bank account like a black hole eats stars. You can't control what you can't measure accurately and the tools to measure color and make accurate profiles is very expensive, as is the talent needed to know how to operate them.
The dirty little secret the color management experts selling the gear and advice don't want you to know is that the brain is so adaptable to color and contrast than anal retentive attention to color management really isn't necessary. If you use a gray card to maintain neutral gray balance over the tonal scale and expose the file in the camera to record a full range of tone — which in most cases requires the use of two flashes in a key overlapping fill configuration — a photographic image will look "normal" on any device in the environment the viewer is adapted to.
Why? Because that's how the ICC profile process is engineered to work.
The best objective baseline for evaluating your monitor and printer is to go outside on a sunny day and have a person in a purple shirt hold a gray card, gray scale, white rag and small piece of foil in sunlight with shaded eye sockets on the person holding the card.
Set Custom WB off the card (away from the purple shirt) and adjust exposure so the specular highlights on the foil are clipping but the white rag isn't. Don't worry how the midtones and shadows look, that's a function of lighting contrast and camera sensor DR.
Then without screwing with anything convert from RAW to 16-bit ProPhoto.
On the monitor in the RAW editor the card should not change in color if clicked with the color correction eye dropper if camera Custom WB is functioning correctly. it should also look neutral on screen, but since your brain knows intellectually it is supposed to look that way it will adapt your perception even if the monitor is out of calibration. TRUST THE CAMERA WB (by the numbers color) not what you think you see. The neutral steps on the gray scale should all look neutral on the print. The highlights will be reproduced accurately because they were the baseline for setting exposure but the shadows and midtones will look darker than experienced by eye (your perceptual baseline for normal).
Next print it on a glossy paper your printer recognizes from the set-up menu, letting the printer manage the color. the printer should map the max 255 RGB values to the most saturated colors the CYMK and reproduce and adjust the less saturated colors like the skin so it winds up looking normal. The neutral steps on the gray scale should all look neutral on the print. The highlights will be reproduced accurately because they were the baseline for setting exposure but look duller than on screen. The shadow will look lighter than on screen and the result of the overall lower contrast range on the print the overall image will look flatter on the print than on the screen. The purple shirt will look much less saturated because its one the colors where the difference in screen and printer gamuts differs the most.
Compare print and screen side-by-side and the print will suck by comparison, but it will be the best the printer can be expected to produce. But take the print into the other room and show it to someone who wasn't there when you shot it who remembers what it looked like in person and they will likely think except for the lack of detail in the eyes and harsh dark shadows in looks pretty normal.
Do the same test in studio lighting in you have it, setting custom WB off the card, centered fill to reproduce shadows and key light overlapping to the point it clips the foil but not the white rag and the lighting ratio will EXACTLY FIT SCENE TO SENSOR, a feat not possible outdoors in direct sun which exceeds the DR of the camera....
Everything in the photo on screen and on a print will look "normal" because that's how the process has been engineered to work — when scene range matches sensor range and WB is neutral at capture. Doing those two things at capture by carrying a gray card and a pair of flashes eliminates the need for remedial fixes in PhotoChop.
lylejk wrote:
Hope you don't mind a macabre interpretation Bob. Because it is as such, just supplying the link so don't click if you don't like creepy. lol
; created an AOP file using these skulls (as a side note, this site was recently infected but they, I believe, finally fixed there site. I'm Sandboxed so pretty much immune; more then likely you all are not, but I did check the site and it appears to be fine now). One of many reasons I really like DAP.
What the hey; since I have already created this coral texture while back, why not turn your flower into a coral rendering too (since many of you all mentioned coral). lol
Vincent Versace emphatically writes changing from sRGB to, let's say Adobe ProPhotoRGB is like adding a quart of water to a gallon bucket - you still only have a quart of water. 
; created