George Anchev wrote:
I am really curious to know:

How exactly do you benefit from ProfotoRGB when:

1. No camera captures colors with gammut wider than sRGB/aRGB
2. No monitor can display full ProfotoRGB
3. No printer can print the full ProfotoRGB
4. Human eye cannot perceive the full ProfotoRGB

Those are repetitive points to discuss and make no sense when you know how these things work. In our studio we pay services by a company dedicated to colour managing and they have explained us how all these things really work. They just can't recomend a working colour space that hasn't been accepted as a standard (yet).

It's not all about how the image looks, butabout what information you have at your disposal for the conversion rendering intents.

1: Most cameras capture gamuts larger than adobeRGB in some hues. Rendering from camera profile to adobeRGB can produce saturation clipping and gets rid of many colours available in CMYK.

2: It's not about the monitor. It's about getting all those levels to "exist" when you convert to a smaller gamut perceptually or in a relative colorimetric fashion.

3. True, but most printers exceed the gamut of adobeRGB in the yellows and other hues. Rather than clipping, we choose a large colour space and then render it perceptually back to our working space. AdobeRGB has been our working space all this time, but it has many flaws.

4.Not what we're talking about here. I'm focusing on rendering intents.


How do we benefit from prophoto? Converting a RAW in prophoto and then rendering it perceptually (important) back to a smaller, working colour space, like adobeRGB, has the advantage of preserving all the perceptual levels of colour. We are not working with the actual captured colours, just the perceptual levels with far less clipping, which is what matters.

Thanks

Sean Baker Photo wrote:
I don't want to ASSume, but you are working in 16bpc, right?

Yes yes yes, of course. First from raw to prophotoRGB 16bpc, then perceptual to adobeRGB 16bpc. Then work work work and then from adobeRGB to Fogra39 for the RIP to print the certified colour proof (our RIP needs fogra39 or iso coated v2 CMYK files for printing and for the spectrophotometer to certify).

The question is: would I benefit from chanigng adobeRGB to ECI-RGB since, on paper, it's so awesome? (d50, L* rather than gamma, equidistant, gamut covering most CMYK...).

Ashish Arora wrote:
I have worked for a client on this. Yes, I've been told about the same benefits. At that time, I assumed he was right.

You know how my viewpoint goes for all this: Do as the client says, even if its not better. (If it is, then fantastic, but from what I gather their printer is set to use it.)

When it comes to choice: Prints can tell not theories or anything else.

I personally have not tried the printing Adobe RGB vs. ECI, have you? If so, what were your findings? Did all the pointers actually match? Was it big change? If not, was the smaller change observed, well ask yourself would be any beneficial knowing most others wouldn't even bother to use ECI.

PS- You should be questioning printers more than anyone else around here.

This is not about the clients. Most clients we work for (ad agencies, DT, wallpaper, cosmopolitan, zara...) they all require Fogra39 certified contract proofs, and so we deliver. The thing is that, according to ECI, adobeRGB is not really the best working space for the RAW->working space->output space flow, especially for hard proofing and certifying.

FLEXmanta wrote:
This is not about the clients. Most clients we work for (ad agencies, DT, wallpaper, cosmopolitan, zara...) they all require Fogra39 certified contract proofs, and so we deliver. The thing is that, according to ECI, adobeRGB is not really the best working space for the RAW->working space->output space flow, especially for hard proofing and certifying.

That's why I mentioned about printing and observing the results for over a period of time. What you are saying, is exactly what I have heard.

But it all comes down to when you make practical observations. I never believe theories, if I see the results, I will switch.

Changes must not happen overnight and never because someone said, but because you tried it and found it better. And what you say is sure a big change from what I gather.

But then that's just me and my 2 cents.

FLEXmanta wrote:
How do we benefit from prophoto? Converting a RAW in prophoto and then rendering it perceptually (important) back to a smaller, working colour space, like adobeRGB, has the advantage of preserving all the perceptual levels of colour. We are not working with the actual captured colours, just the perceptual levels with far less clipping, which is what matters.

Thanks

Thanks to you.

Can you please show an example where the beneficial difference is visible? I am really curious to see how converting RAW->ProfotoRGB -> working RGB produces better result than RAW -> working RGB.

FLEXmanta wrote:
in our studio, we have policies and the current one dictates to process RAWS in ProphotoRGB initially, then rendering perceptual down to AdobeRGB for working.

Note that this isn't happening if you're using the v2 ProPhotoRGB and AdobeRGB profiles. Only if both are v4 profiles will this be possible. Do you have a v4 AdobeRGB (or ECI-RGB) profile and a v4 ProPhotoRGB one? (It's a well documented limitation of matrix profiles, but it's rarely stated directly.)

FLEXmanta wrote:
3. True, but most printers exceed the gamut of adobeRGB in the yellows and other hues. Rather than clipping, we choose a large colour space and then render it perceptually back to our working space. AdobeRGB has been our working space all this time, but it has many flaws.

4.Not what we're talking about here. I'm focusing on rendering intents.

How do we benefit from prophoto? Converting a RAW in prophoto and then rendering it perceptually (important) back to a smaller, working colour space, like adobeRGB, has the advantage of preserving all the perceptual levels of colour. We are not working with the actual captured colours, just the perceptual levels with far less clipping, which is what matters.

The ColorWiki notes, based on v2 profiles states:

when converting from workingspace to working space (say, Adobe RGB to sRGB), the ONLY rendering intent available is colorimetric. Photoshop allows you to select perceptual and saturation but they make NO difference (try it sometime). Photoshop should gray out intents that don't exist in the profiles.. but that's just my opinion....

Kevin_Connery wrote:
Note that this isn't happening if you're using the v2 ProPhotoRGB and AdobeRGB profiles. Only if both are v4 profiles will this be possible. Do you have a v4 AdobeRGB (or ECI-RGB) profile and a v4 ProPhotoRGB one? (It's a well documented limitation of matrix profiles, but it's rarely stated directly.)

It is a limitation of the V2 specification or the matrix?
I see more harm than good with monitor LUT profiles...

Some interesting links...
http://www.naturescapes.net/phpBB3/view … &p=1480953
http://photo.net/digital-darkroom-forum/00YAgu
http://photo.bragit.com/Lightroom/artic … ment.shtml

Kevin_Connery wrote:
Note that this isn't happening if you're using the v2 ProPhotoRGB and AdobeRGB profiles. Only if both are v4 profiles will this be possible. Do you have a v4 AdobeRGB (or ECI-RGB) profile and a v4 ProPhotoRGB one? (It's a well documented limitation of matrix profiles, but it's rarely stated directly.)


The ColorWiki notes, based on v2 profiles states:

Seems you are the kind of guy I was hoping to have in this thread. Sorry George, I wouldn't want to hijack the thread by explaining the use of prophotoRGB with it's advantages and dangers (hint, compare a blue sky straight in adobeRGB, with one processed in prophoto and taken to adobeRGB perceptually in 16bpc). The point that it contains colors not visible to the human eye or reproducible by the monitor, although correct, makes no sense in an industry standard level. ProphotoRGB is there for a reason.

Yes, most of the ICCs we use are in their v4, including ECI.
Let's forget about the fact that we process prophoto PSDs out of the RAWs and then take it perceptual to adobeRGB for work. This was a suggestion from an expert, not a retoucher or a photographer, a guy who has a serious colour management company and is responsible for many of the main publication's colour management here in Spain including Elle, Cosmo and Yodona (known to have one of the best rotary presses in the business). The main reason is that our work comes from many different sources: large format negatives scanned with our hasselblad scanner, CGI, canon, nikon, phaseOne, Leaf... and one way to get most of the perceptuality from all these different captures methods, is to use the great prophotoRGB and then taking it to our smaller working space using perceptual. Later we can work on colour (we're retouchers, now aren't we).

I just want to know, since we softproof ISO Coated v2 and Fogra39; our studio is carefully lit at d50 as are our hardproofing lightboxes; we match fogra39 certified proofs with on screen images on monitors calibrated to 1,8 and d50, would it be right to start working on all our new projects with ECI-RGBv2 rather than adobeRGB?

Also, would we need to calibrate monitors for L* rather than gamma 1.8? That's the part that confuses me the most actually since i believe it would be far from the current recommended standards (d50, gamma1.8, native primaries).

I've always been uncomfortable with adobeRGB because of its d65 whitepoint and gamma2.2. I thought ECI would be perfect with its d50 and equidistant levels... but the L* not matching my monitor's gamma just confuses me a bit.

I want to make sure I'm making sense before I bring this up on our next studio meeting. Changing all our stations to this new workflow, explaining this to my colleagues and writing the policies is not a matter of minutes.

Thanks all for the help

FLEXmanta wrote:
Yes yes yes, of course. First from raw to prophotoRGB 16bpc, then perceptual to adobeRGB 16bpc. Then work work work and then from adobeRGB to Fogra39 for the RIP to print the certified colour proof (our RIP needs fogra39 or iso coated v2 CMYK files for printing and for the spectrophotometer to certify).

The question is: would I benefit from chanigng adobeRGB to ECI-RGB since, on paper, it's so awesome? (d50, L* rather than gamma, equidistant, gamut covering most CMYK...).

I did a little comparison between the two and here is what I found.

EciRGB is neutral throughout its range where as AdobeRGB isn't quite but pretty close.



The color gamut of EciRGB is fairly bigger then AdobeRGB as well.



But that doesn't quite tell the whole story. There is quite a bit of overlap between the two but there are areas where one can display colors that the other can't.

Starting at the darkest values (L*0 and up), there are no reds, greens or yellows in either but they both favor blues to an extent. As they both get brighter, the other colors begin to appear and at about L*30, they both reach their maximum saturation in the bluish region. AdobeRGB has the edge in the purplish blues while EciRGB has a slight edge in the cyans (more on this later).

NOTE: EciRGB is represented by the red outline where AdobeRGB is represented by its true colors.



As the L* value continues to increase, we see a noticible decline in the blues for both color profiles. Up to L*63, EciRGB continues to hold the edge in cyans and takes clear lead in the reds. AdobeRGB has a very slight edge in certain hues of green but nothing significant.



After which, the reds begin to decline as do the blues but the greens continue to become more saturated as the brightness continues to increase.



Beyond L*82, most colors become desaturated with the exception of the yellows. AdobeRGB can reproduce some shades of yellow that EciRGB can't (and vice versa), but yellow is one of the most difficult colors to distinguish.



By the time the brightness value increases to L*99, nealy all colors except for the yellows, are completely desaturated.



Here are some 3d graphs to help illustrate the differences between the two a little better.

NOTE: EciRGB is represented as a red wireframe where as AdobeRGB is represented with its true colors present.





Both profiles encompass the Fogra39 color gamut pretty well:

EciRGB


AdobeRGB


But AdobeRGB doesn't quite keep all the range of cyans within its boundries. At about L*31 through L*57 we see some clipping mostly in the L*52 range.





This isn't the case for EciRGB on the other hand. The Fogra39 color gamut is fully enclosed throughout all L* values.





EciRGB looks like a pretty good choice to me especially if you're adherring to the ISO 12647 standard. As far as conversion to cmyk, I would need some images to test to find which is better in which situation.

FLEXmanta wrote:
Ruben. That was amazing. Wasn't expecting such a detailed analysis.
So, ECIRGBv2 looks good.

We do work under ISO 12647-7 certifications for our color proofs and we do both Fogra39 and ISOCoatedV2 (which are actually the same).

Apart from the gamut, would there be any advantages in ECIrgbv2 given its d50 whitepoint? What about L* TRC? How do those affect converting to fogra39 perceptually (ill be using ICC v2).

THANKS!

Glad I could help. I'm sorry it took so long to post a response. I'm not as familiar with fogra as I am with gracol (american), and I knew nothing of eciRGB until your post. So I had a little bit of studying to do. But I think this stuff is interesting.

As far as the d50 whitepoint, that will better assist softproofing (which I hope you're doing), as the viewing environment is also d50 lighting. The L* tone response curve is just a different way of expressing curvature (or gamma). Keep in mind, when converting from one colorspace to another, the lab colorspace is used as an intermediary (or profile connection space aka: PCS), between the two. Because your starting tone response curve is L* and lab's tone response curve is L*, you eliminate another variable and as such, you eliminate any changes to luminance that would otherwise occur were you using a gamma of 1.8 or 2.2.

Ruben Vasquez wrote:

Glad I could help. I'm sorry it took so long to post a response. I'm not as familiar with fogra as I am with gracol (american), and I knew nothing of eciRGB until your post. So I had a little bit of studying to do. But I think this stuff is interesting.

As far as the d50 whitepoint, that will better assist softproofing (which I hope you're doing), as the viewing environment is also d50 lighting. The L* tone response curve is just a different way of expressing curvature (or gamma). Keep in mind, when converting from one colorspace to another, the lab colorspace is used as an intermediary (or profile connection space aka: PCS), between the two. Because your starting tone response curve is L* and lab's tone response curve is L*, you eliminate another variable and as such, you eliminate any changes to luminance that would otherwise occur were you using a gamma of 1.8 or 2.2.

That confirms what I thought.

Today I started my first job with ECI-RGBv2. It's for Cosmo, so i hope it all goes well. So far, it has.

I do soft-proof Fogra39, but I also hard-proof with certified UGRA-Fogra prints in a D50 JUST viewing booth.

For the curious, here's my station. That contract proof there is an old one with the old color wedge. We now have the 3rd version of that wedge with one more row of patches.
http://www.facebook.com/photo.php?fbid= … e=1&ref=nf

Also, do you think with a monitor like mine, setting its TRC to L* would help the color engine better instead of setting the monitor to gamma1,8 and making ACE do the job?

George Anchev wrote:
I am really curious to know:

How exactly do you benefit from ProfotoRGB when:

1. No camera captures colors with gammut wider than sRGB/aRGB

My camera does have photoRGB. Which is what i happen to shoot .

Sergei Rodionov wrote:

My camera does have photoRGB. Which is what i happen to shoot .

You mean prophoto? Doubt it. And if you are shooting RAW, the file has no profile assigned.

...but going back to the original topic.

Today I did the first photo of a 7 image editorial. It happens to have some of the trickiest colors to take to a standard offset CMYK like fogra39. It's a fluorescent color block editorial for Cosmo, shot Terry Richardson's style: camera flash (actually my references are Tony Kelly's). I can tell eciRGBv2 is already helping a lot with those super bright yellows and retaining perceptual volumes in the super saturated colors.

FLEXmanta wrote:

You mean prophoto? Doubt it. And if you are shooting RAW, the file has no profile assigned.

...but going back to the original topic.

Today I did the first photo of a 7 image editorial. It happens to have some of the trickiest colors to take to a standard offset CMYK like fogra39. It's a fluorescent color block editorial for Cosmo, shot Terry Richardson's style: camera flash (actually my references are Tony Kelly's). I can tell eciRGBv2 is already helping a lot with those super bright yellows and retaining perceptual volumes in the super saturated colors.

To actually make a perceptual difference under controlled lighting , I'd be surprised if there where that much difference. As said above differentiating between yellows is much more difficult  than dark primaries. As far as saturated yellows go I cannot understand how there would be a lot of difference between Adobe RGB and ECI-RGB whereas both are well contained in Fogra 39.
From the above Colorthink graphs, as expected the Cyans are clipped, but ever so slightly with Fogra 39 ( tolerances for Fogra 27 were different) .

Distribution though may be better with a power curve rather than a straight gamma function.

As I said in the first post, ECI was derived to better fit an offset process, your validating that so it has been a good thread to follow. Albeit for MM a bit techie!

FLEXmanta wrote:
You mean prophoto? Doubt it. And if you are shooting RAW, the file has no profile assigned.

If you know anyone who shoot with Leaf - ask them to show these options to you If not - give me a shout if you happen to be around Dallas or in bits of Europe that i am passing (northern mostly, unless i go to Venice next Feb)  - i will show you

However it is a wee bit offtopic indeed.

FLEXmanta wrote:
That confirms what I thought.

Today I started my first job with ECI-RGBv2. It's for Cosmo, so i hope it all goes well. So far, it has.

I hate you. The last job I got hired for was to manually focus stack images of micro fossils together. Oh well. At least she was a paying client.

FLEXmanta wrote:
I do soft-proof Fogra39, but I also hard-proof with certified UGRA-Fogra prints in a D50 JUST viewing booth.

For the curious, here's my station. That contract proof there is an old one with the old color wedge. We now have the 3rd version of that wedge with one more row of patches.
http://www.facebook.com/photo.php?fbid= … e=1&ref=nf

Nice setup. That would be my ideal but I can't justify the cost right now. Maybe one of these days...

FLEXmanta wrote:
Also, do you think with a monitor like mine, setting its TRC to L* would help the color engine better instead of setting the monitor to gamma1,8 and making ACE do the job?

Absolutely yes! Again, that will help with softproofing. What software/hardware are you using to calibrate your screen? Neither the proprietary software that came with my Eizo or the spyder3 software offer L* for TRC.

FLEXmanta wrote:
Today I did the first photo of a 7 image editorial. It happens to have some of the trickiest colors to take to a standard offset CMYK like fogra39. It's a fluorescent color block editorial for Cosmo, shot Terry Richardson's style: camera flash (actually my references are Tony Kelly's). I can tell eciRGBv2 is already helping a lot with those super bright yellows and retaining perceptual volumes in the super saturated colors.

Will these images appear in the next cosmo? I'd be really curious to see them.

If you'd like, I can map an image to any of the color spaces. You can learn a lot seeing exactly where the image falls in the various color spaces and seeing just how the colors shift from one space to another. PM if you're interested.

Sergei Rodionov wrote:
If you know anyone who shoot with Leaf - ask them to show these options to you If not - give me a shout if you happen to be around Dallas or in bits of Europe that i am passing (northern mostly, unless i go to Venice next Feb)  - i will show you

However it is a wee bit offtopic indeed.

We have a couple leaf backs... I'll ask one of the digital techs to show me. Although, i really still think the RAW file comes, per definition, untagged (As it really isn't an image). I'm assuming the setting is there for when the camera is set to shoot TIFFs os as a preset that will show when you process in Leaf Capture. We only use Leaf Capture when shooting tethered with the Leafs.

Ruben Vasquez wrote:

FLEXmanta wrote:
That confirms what I thought.

Today I started my first job with ECI-RGBv2. It's for Cosmo, so i hope it all goes well. So far, it has.

I hate you. The last job I got hired for was to manually focus stack images of micro fossils together. Oh well. At least she was a paying client.

FLEXmanta wrote:
I do soft-proof Fogra39, but I also hard-proof with certified UGRA-Fogra prints in a D50 JUST viewing booth.

For the curious, here's my station. That contract proof there is an old one with the old color wedge. We now have the 3rd version of that wedge with one more row of patches.
http://www.facebook.com/photo.php?fbid= … e=1&ref=nf

Nice setup. That would be my ideal but I can't justify the cost right now. Maybe one of these days...


Absolutely yes! Again, that will help with softproofing. What software/hardware are you using to calibrate your screen? Neither the proprietary software that came with my Eizo or the spyder3 software offer L* for TRC.

I use NEC spectraviewII. It does have L* setting for TRC. My second monitor doesn't.

FLEXmanta wrote:
We do work under ISO 12647-7 certifications for our color proofs and we do both Fogra39 and ISOCoatedV2 (which are actually the same).

Hey, I just wanted to point out, that those profiles are not the same. They are based on the same characterisation data but have different seperation characteristics for example.

Here's a colorful chart for your enjoyment.



Julian

Ruben Vasquez wrote:
Absolutely yes! Again, that will help with softproofing. What software/hardware are you using to calibrate your screen? Neither the proprietary software that came with my Eizo or the spyder3 software offer L* for TRC.

You can try ColorEyes Display Pro http://www.integrated-color.com/cedpro/ … splay.html
or basICColor display 4 http://www.basiccolor.de/english/index_E.htm
they both have support for Eizo and also have L*

IMPORTANT, when installing basICColor it will ask if you are in the USA, say NO
or you are not going to have L*.