Author | Thread |
|
12/29/2009 07:41:03 PM · #1 |
I have been shooting RAW for less than a year now and am learning more. One issue which is still a bit confusing is that of color space. My Nikon D80 has three settings for color space- two sRGB ones (one said to be more for portraits the other supposedly for landscape whatever that means- maybe one is warmer than the other?) and of course Adobe RGB. I have been reading books by Scott Kelby and Joe McNally and both recommend Adobe RGB. I edit my photos in Photoshop CS4 (8 bits)and do not do any of my own printing. 99% or more of my photos are only viewed either on my computer or online. Older information says that some web browsers don't handle Adobe RGB very well and colors may look off but hear that this is much less of an issue today.
Does setting my camera to Adobe RGB even matter if I am using RAW? On those rare occasions where I do print is there much of a difference? After editing I keep the raw file but save the resulting image as jpeg. What do you do? Trying to optimize my settings for better results. I have definately had situations where using raw helped me get a good image from what would have been a mediocre to bad image (due to things like difficult lighting exposure- highlights and shadows) so I know the benefit of using it.
Do you leave the settings for other things like saturation and sharpening at their "factory" settings in camera as well?
Thank you for any responces.
|
|
|
12/29/2009 08:07:32 PM · #2 |
Think of it this way: RAW is the entire spectrum of colors your camera is capable of capturing. AdobeRGB would be a big box of crayons chosen from that spectrum, and sRGB would be a smaller box of crayons. Photoshop is a skilled artist that can render a pretty good image using any of these sets, but of course the more crayons it has available the better it will be able to handle subtle differences and gradations, so it's a good idea to capture and edit photos with the widest range of data available (AdobeRGB). However, most browsers (and some photo labs) assume you're working with the smallest box of crayons (sRGB) no matter what colors you actually used, so the result will be drab and lifeless unless you convert to sRGB before posting. These two images were captured in different color spaces with the same camera settings, and saved without conversion:
AdobeRGB
sRGB

Message edited by author 2009-12-29 20:08:04. |
|
|
12/29/2009 09:14:04 PM · #3 |
There is another color space called ProPhoto RGB which is even wider than AdobeRGB. As Shannon stated most monitors will not be able to display the entire gamut of AdobeRGB and certainly not ProPhoto RGB. In fact, I'm not sure the human eye can distinguish the entire gamut of either.
With that said, you should tag everything you do on the web and most print shops with sRGB.
So, you may ask, why do the two even exist? Well, to give you the flexibility of the color space without losing the information. Here is an example, I took a photo of my daughter sitting on Santa's lap (cue in the Red suit here), there is a fireplace to the left and dim lights to the right. I exposed enough fill flash. The result? A nice photo with blown red highlights on Santa's suit, blown that is, in sRGB space. In ProPhoto, the bright red is within the gamut of the space. Which means, I can map the areas INTO sRGB space at my will thus controlling how it looks.
Here is another example, going the opposite direction. Lets say I have a photo, in sRGB where every pixel is within the gamut of sRGB. Great, that means anyone looking at it on their monitor will see what I intended them to see. I have a dye-sub printer I use to print out 4x6 photos. The color space of this printer is DIFFERENT than sRGB. What does that mean? It means that certain colors or shades will not be printed as I see it on the screen, rather, it will take the closest value within the printer color space. So, I can use the soft-proof feature of photoshop to "see" what it will look like when it gets printed. Then, while proofing, I can add a levels, or curves adjustment layer to map the out-of-gamut space into the printer's space so that the photo looks more reasonable.
Hope that helps and doesn't make things more confusing.
Paul |
|
|
12/29/2009 09:42:46 PM · #4 |
Originally posted by PGerst: As Shannon stated most monitors will not be able to display the entire gamut of AdobeRGB and certainly not ProPhoto RGB. In fact, I'm not sure the human eye can distinguish the entire gamut of either. |
I don't think it's the monitors which limit the visible gamut, but the browsers ... if you can see it to edit it then obviously the monitor can display it. |
|
|
12/29/2009 10:22:48 PM · #5 |
I'm not so sure that is true. Every monitor has a limited gamut just as our eye does. The far yellows and deep reds are two in particular that I have experienced with limitations to the monitor. Even though we can "see" it on the monitor, there is still an interpolation of the digital color space values by AdobeRGB/ProPhotoRGB to what the monitor can produce.
Originally posted by GeneralE:
I don't think it's the monitors which limit the visible gamut, but the browsers ... if you can see it to edit it then obviously the monitor can display it. |
|
|
|
12/29/2009 11:11:35 PM · #6 |
Originally posted by GeneralE: I don't think it's the monitors which limit the visible gamut, but the browsers ... if you can see it to edit it then obviously the monitor can display it. |
Every device has a limited color gamut of some sort, and monitors are no different. Some newer, wide-gamut models are touting 97% of AdobeRGB as a feature, so presumably a "standard" display covers significantly less than that. |
|
|
12/31/2009 12:48:43 PM · #7 |
But doesn't RAW just capture everything the sensor can capture, and only on conversion does colorspace matter? I was wondering this last night when somebody on a blog said to shoot raw and assign AdobeRGB, which didn't make sense to me. Here's an article (2nd section) that implies it matters at conversion and not capture. |
|
|
12/31/2009 12:56:39 PM · #8 |
Originally posted by Ken: But doesn't RAW just capture everything the sensor can capture, and only on conversion does colorspace matter? I was wondering this last night when somebody on a blog said to shoot raw and assign AdobeRGB, which didn't make sense to me. Here's an article (2nd section) that implies it matters at conversion and not capture. |
You're correct, it does not matter at capture; the RAW data will be the same. It only matters once you convert. Now, your converter might or might not understand what the camera was set to and could default to the camera-set space on conversion. In that case, for the most seamless process, set the camera to your desired output space.
Most 3rd-party converters don't care (or even know) what the color space setting was in-camera. They are configurable to whatever you want on export. Brand-specific converters (think Nikon Capture NX) do know what the camera setting was. They still can override it on output at your discretion. |
|
|
12/31/2009 01:04:49 PM · #9 |
Originally posted by kirbic: Originally posted by Ken: But doesn't RAW just capture everything the sensor can capture, and only on conversion does colorspace matter? I was wondering this last night when somebody on a blog said to shoot raw and assign AdobeRGB, which didn't make sense to me. Here's an article (2nd section) that implies it matters at conversion and not capture. |
You're correct, it does not matter at capture; the RAW data will be the same. It only matters once you convert. Now, your converter might or might not understand what the camera was set to and could default to the camera-set space on conversion. In that case, for the most seamless process, set the camera to your desired output space.
Most 3rd-party converters don't care (or even know) what the color space setting was in-camera. They are configurable to whatever you want on export. Brand-specific converters (think Nikon Capture NX) do know what the camera setting was. They still can override it on output at your discretion. |
And when I first installed CS3 & CS4 the hard way that it defaults to AdobeRGB when I wanted sRGB. |
|
|
12/31/2009 01:52:14 PM · #10 |
Originally posted by scalvert: Originally posted by GeneralE: I don't think it's the monitors which limit the visible gamut, but the browsers ... if you can see it to edit it then obviously the monitor can display it. |
Every device has a limited color gamut of some sort, and monitors are no different. Some newer, wide-gamut models are touting 97% of AdobeRGB as a feature, so presumably a "standard" display covers significantly less than that. |
If the monitor can't display the entire gamut, how are people editing or seeing differences between the color spaces? Color monitors are (theoretically) capable of displaying 16.7 million colors -- how many are there (theoretically) in AdobeRGB? |
|
|
12/31/2009 02:36:34 PM · #11 |
You can't, which is one of the reasons there is always a slight discrepancy between what the print gives and what the monitor shows unless the print gamut is contained within the monitor gamut. All the color rendering software (i.e. windows monitor profiles) can do is give the best estimate of what the color is within the gamut.
Its also not about the number of colors that a monitor can display, since this is set by the bit depth of the color model (Compare today's colors to the old CGA colors)! Rather, it is about where in the color space those 16.7 million colors are displayed.
Here is an example from my monitor, a NEC MultiSync FE992 (CRT) display as it is compared against 3 profiles, sRGB (Left), Adobe RGB (Middle), and Wide Gamut (Right). The colored regions display the monitor profile, the shaded regions show the comparison. Notice that my monitor comes closest to the sRGB space. Today's LCD monitor profiles are generally larger than my older CRT.
 |
|
|
12/31/2009 02:51:40 PM · #12 |
Originally posted by GeneralE: If the monitor can't display the entire gamut, how are people editing or seeing differences between the color spaces? |
They see the part of the gamut that the monitor can display, which is still larger than sRGB on most modern monitors.
Originally posted by GeneralE: Color monitors are (theoretically) capable of displaying 16.7 million colors -- how many are there (theoretically) in AdobeRGB? |
Apples and oranges. Color gamut is more than just number of colors— it's the range of those colors. Switching your display from 16 bit to 24 bit gives you more colors for smoother transitions, but it doesn't extend the range to make green more saturated. Using the crayon analogy, a box of 64 gives you more colors, but also provides an extended gamut (like metallic gold or fluorescents) that you couldn't possibly mix with the other crayons. The difference between sRGB and AdobeRGB is somewhat like mixing colors with muted red, blue and green crayons vs. really intense, pure hues.
The entire range of colors a human eye can perceive is defined as the NTSC color gamut (unrelated to the TV standard). sRGB and average CRT monitors cover about 72% of that gamut, while AdobeRGB reaches 92%. Most common LCD monitors cover 75% (usually extending the range of greens), but those with wide-gamut CCFL backlights cover 85-100% of NTSC and the best LED-backlit monitors can reach 114% NTSC to show colors the human eye can't even see. |
|
|
02/01/2010 02:18:41 AM · #13 |
Color space can be a confusing issue. Hopefully this will simplify it for you:
You only have a certain number of bits to work with (for colors), usually 8 bits or 16 bit per channel. Let's say your photo has a red that (in 8 bit) is 243,23,46. Those are just numbers and don't mean anything in real life until you assign a color space.
People new to this immediately want to use the widest possible color space they can use, but that is not usually the best thing for them. Why? Because wider color spaces map those 8 or 16 bits onto a wider space. This makes for wasted space for the majority of photos. They won't have reds that are out of gamut (clipped, etc..), but it also makes for less subtle gradations.
For example, if your RAW photo would fully fit in the sRGB gamut and you convert it to AdobeRGB instead, your resulting file would actually have fewer unique shades/colors. Because so much of the 8 bit or 16 bit range would be unused in your file. It should be noted here that this is a good reason for using 16 bit files because it usually isn't an issue in that case, but it for sure comes into play if you only have 256 individual shades per channel like you do in 8 bit.
Don't get thrown off by comparing the gamut of various monitors to the gamut of various color spaces. The whole purpose of color spaces and our color management engines is so that our color is always mapped to our various devices. Whatever the reddest red is in our working space will be mapped to the reddest red that our monitor can display (ideally) so that nothing is clipped and we can still see all the gradations in our file.
If all of your color profiles for your devices are fairly accurate, this color mapping and whatever conversions need to be done (when printing etc..) are done automatically and with good results, but being aware of what is going on will be good to know if you come across some extreme situations. Example: You're working on a photo of some really vibrant oranges and you know that it will be sent to a printing press that will load up a special neon orange ink just for this job. In that case you will want to be much more careful about the gamut of your color spaces, and which conversions you use.
This brings up a good point, it is better to think about the color gamut of whatever your end output will be than the actual real life color of your subject in the photo, or which colors your monitor can display.
In pretty sure that sRGB is a good fit if you'll be printing to photo labs (taking your pix to the Fuji machines at Target) but might not be if you'll be printing to a new high gamut epson inkjet with 8+ colors of ink, etc..
-mh |
|
|
02/01/2010 08:56:17 AM · #14 |
|
Home -
Challenges -
Community -
League -
Photos -
Cameras -
Lenses -
Learn -
Help -
Terms of Use -
Privacy -
Top ^
DPChallenge, and website content and design, Copyright © 2001-2025 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 06/26/2025 08:42:19 AM EDT.