Hi Everybody,I'm in a bad mood because I had a big fight with Ellie. She doesn't have to rub it in just because she is smarter than me.
So let's debunk the myth that Adobe1998 has a huge gamut advantage over sRGB. It goes along with the myth that most greens are out of gamut for both of these color spaces.I'm certain that these myths were created to advance a political agenda or just to enable some smart alec to tell you about them.
It's like profiles. An absolute profile is more accurate but a perceptual profile is more pleasing.
Yes. Adobe RGB's color space is slightly larger than sRGB's color space. However, sRGB is more pleasing to the eye. Neither one seriously omits green. Both omit some of the cyanThe differences are better shown in the Luv color space than in the XYZ color space. Here's the long and short of it as once revealed on the internet. It's been surpressed by the color police so I can no longer find the link.Phil
Stefan Lindblad said:Adobe RGB has a larger colour space than sRGB. For graphics this might not mean so much. But the more complex colour wise an image is, the better CMYK conversion when using Adobe RGB.
Exactly. If the photographs you take are for use on the web or printing on a desktop printer, then sRGB is probably the best colour space to work in. If your work is for commercial printing then the extra bit of 'space' you get in AdobeRGB gives better results when converting to CMYK or to the colour spaces used by pro photo printers who use Fuji, Canson or Hahnemuhle papers. The difference is small, but it's real, so it makes sense to make use of it.
RoyReed said:The difference is small, but it's real, so it makes sense to make use of it.
Agreed, I have two point others may have not thought of.
1. When converting spot colors to RGB for wide gamut print on ink jet devices sRGB will not reproduce some of the spot colors as they are out of gamut, also some wide gamut CMYK profiles produce dark greens and blues that are out of gamut for sRGB but are however in gamut for Adobe RGB and ECI RGB.
2. Knowing the concepts mentioned here by Stefan, Roy and I is the difference between hiring the pro and the non-pro, this allows the pro to be sought out by those who do not want the average and allow the pro to maintain the profit margins that are sorely needed today.
Here's one I like, shooting everything in color (as wide a gamut as possible) and outputting your black and white as desaturated RGB, it requires perfect gray balance on the media profile and the source file but reproduces black and white that is unobtainable wit any other process.
RoyReed said: AdobeRGB gives better results when converting to CMYK or to the colour spaces used by pro photo printers who use Fuji, Canson or Hahnemuhle papers. The difference is small, but it's real, so it makes sense to make use of it.
AdobeRGB gives better results when converting to CMYK or to the colour spaces used by pro photo printers who use Fuji, Canson or Hahnemuhle papers. The difference is small, but it's real, so it makes sense to make use of it.
I kind of have my doubts this difference is significant for most printers. It assumes a few things, it assumes that your picture has colors at the outer edge of the gamut, which is compromised of highly chromatic colors that are rarely seen in real life. Second, it assumes that your printer is capable of correctly distributing the ink at that outer boundary.
I have a real hard time believing this when we are still using things like Black Point Compensation because most printers can still not accurately display black.I doubt the ability of most printers of putting down a tiny drop of phthalocyanine or quinacridone, and do it with that much precision, that they can show the difference between small differences in gamunt. I think a lot of marketing is going on.
Can you batch process all the edits for 300 images as fast? In AfterShot Pro I have opened 300 images, did custom individual image edits for exposure, color, sharpness to mention a few to EACH IMAGE.
Simply by selecting all the images and saving them the individual image edits were applied to the saved file format as well as the selected color space and bit depth.
I'm not so sure that Photo-PAINT X7 can do that as quickly if at all.
Does your batch preset do image specific corrections?
I don't think the gamut pictures are a good representation of the difference. The difference is much smaller when you use a realistic distribution of colors. That gamut picture assumes all your colors have maximum chroma, this isn't realistic, you will have more subdued colors in your picture. The difference in gamut will be A LOT smaller than on that picture in reality, because the differences are smaller at lower gamut colors and we are far less adapted to see the difference between a subdued brown than we are at seeing the difference between a bright orange color. Another reason I'm not a fan of Adobe RGB is because the cost of a mistake is HUGE. When an Adobe RGB picture is displayed on an sRGB screen or printed, and it was not properly converted, you're going to clip much much more of the gamut than you would have gained from going from sRGB to Adobe RGB.
When you use AdobeRGB, you argue that the very small gains in your color gamut, are worth the risk of clipping a massive part of your gamut if something wrong happens during conversion from device to device, that's not a risk I am willing to take, 99% of the world and content runs on sRGB, you have to be really sure that nothing in the chain uses sRGB, from creation to distribution, and that your audience is also capable of not just having hardware that can show the gamut, but also that nothing goes wrong during conversion. That's a large bet to make, and one that will cost you dearly if get it you wrong.
The color gamut isn't really the problem either I feel, it's high dynamic range, it's the difference between really bright and really dim lit areas, the difference of the massive amount of light an outside scene can give you and the tiny amount of light indoor lighting gives you, the fact your eyes need to adapt when you go from inside to outside, that massive difference is what is missing to create more realism, not the fact one gamut has a 1% higher color gamut at the outer edge of the max chroma of a color.
It's not about hues, we can show hues just fine, it's about the difference in value, the massive amount of contrast high dynamic range can give you in real life, and the pitiful contrast a printed page or screen can give you and the fact so few devices can accurately capture that range.
You have some points, in reality a scene may only have a few colors and a small area that expand the dynamic range of colors in the scene. However it is that area when reproduced in a high quality conversion that make the difference between work and great work.
It is unfortunate today that we lose sight of how to utilize all the available technology to get improved results and to get the client to be willing to pay the extra.
The advantage of using larger gamuts for print including printing press is in the conversion to either the media profile or to the press CMYK, specifically using perceptual rendering. Black Point Compensation is viable but non-ICC compliant. With a process such as this two colors that are very close is special relationship, one just in gamut and one just out of gamut will convert to CMYK with a perceived difference maintaining a more natural look to the end product. Mind you I did not say accurate, I said perceived difference, always remember there is no way to fit a size 12 foot into a size 8 shoe so this is always a judgment call.
Unfortunately today with profit margins and the state if incoming files as they are many output providers in general are just doing the quick fix. Well, they're doing what they can afford!
Phil1923 said:5. The end result is that I will get an sRGB print which looks like the image on the monitor.
No you won't. Depending on the printer you'll get a CMYK print with the printer software itself having done the conversion from sRGB. (Or if your printer has more than four colours you'll get something like CMY lightBlack darkBlack.) A print cannot be any form of RGB. The conversion to some sort of subtractive colour space has to happen somewhere. In your model you are choosing to delegate the settings to the printer.
Ok Phil here you go I'm posting a screen capture of some images of my favorite subject, (for the web I had to set the three properly color managed documents side by side and capture all to sRGB) on the left is the RAW converted 16 bit prophoto tif, next to the right is on top a properly converted sRGB, below an assigned sRGB, then on the far right, on top a properly converted Adobe RGB and below as assigned Adobe RGB.
Pay attention to Morgans dress, the red and yellow toy in the background and the red end of the box of animal crackers. Most colors in this capture remain very similar, however this is why we use Adobe, my Epson Stylus Pro will print the differences. Not sure if it will show up on the web or on poor displays.
Here's what's interesting, this image displays great from the web on the display in which it was created, it's ok on a decent LED graphics display but crappy on my tablet and all laptops. Anyone who wants the original let me know
The bottom rows were assigned the profile middle sRGB and right Adobe RGB. An improper setup. The interesting issue that popped for me is that how poor modern monitors are, my High End ViewSonic LED shows some differences between the three top images, my tired and true CRT really shows that it has a much wider gamut as when I view the capture or the original CDR files there are significant differences between the three top images, the left Prophoto being the best with significant changes in the Adobe and then even more with the sRGB. I'm going to look into the original CDR files on my ViewSonic and see what I see.
This really demonstrates the near impossibility od doing real color correction or color management tutorials on the web.
BTW I will do a chart for you showing how RAW works as you like sharts.
Hi Everybody, I'm not too concerned with the answer to the minor question. Based on what I have read, I'm fairly certain -- but not positive -- as to what happens when an Adobe RGB photo gets printed as an RGB image. It's all in Fraser's diagram and the conversion profiles shown in the diagram. (These profiles are conversions between abstract spaces. These days, profiles converting voltages to color spaces are within the peripheral devices.)
Following Fraser's diagram, an Adobe RGB photo gets converted to the central color space (usually Lab) and then gets converted from Lab to sRGB. At one time in the distant past, the profile conversions were done by look-up tables plus the addition of magic potions. Today, they seem to be done with "simple" matrix algebra.
So let's follow the process using matrix algebra. The conversion is via a "simple" 3 x 3 matrix. Here's the matrix without numerical values because I don't know what values are used. The left column is the sRGB values for a pixel. The right column is the Adobe RGB values.Next consider a case where a pixel is only green in Adobe RGB. The matrix is then as follows.Let's multiply this out to get the RGB values in sRGB. It's:Well look at that. The pure green has become desaturated by the addition of red and green.Whether or not it gets darker depends on the values of the coefficients. I don't know what is being used.
I have seen the conversion described most frequently as that the Adobe image looks muddy in SRGB. That fits the math.
CONCLUSIONS:1. Use Adobe RGB if you can print in Adobe RGB or if you are recording for posterity.2. Use sRGB everywhere else.
So I withdraw the minor question.
Now, can anyone add me and PhotoPaint to Fraser's diagram. One picture or diagram is worth a million words. For me it's worth fifty million words.
Phil