Help with color working space
- This topic has 35 replies, 12 voices, and was last updated Sep 28, 2008 at 7:18 pm by
.
-
Posts
-
I’ll let others more versed in this provide what will likely be a better answer but want to mention two things.
sRGB is from my understanding primarily for web use, I dont know that its necccessary to actually use it. I shoot in RGB and do every there. Simply because thats what the publishers want. I can’t be bothered changing the colour space for the photos I upload. Maybe ou could try using RGB and see if the same problem persist.
Photobucket.
While alot of years have gone by, I tried them as my first host and really ( read really) disliked what it did to my photos. Especially with regards to resising. Yes its free (or was) and theres alot to be said for that, but if your trying to display your photos in the best possible light they arent the host Id choose. Your problem could be there.J
I assume your monitor is calibrated, and that is not the problem.
Tagging on to what John B. said about web hosts, I have been very happy with Zenfolio.
WOW, I had no idea that a hosting site messed with photos like that. I am SURE it is Photobucket because, as I said, the exported pictures look perfect on my desktop after export (before upload to PB).
Neal.
The most popular (but certianly not the only ones) that I know of are.Pbase
Smug mug
Zenfolio.Both Pbase and Smugmug allow visitors to leave comments…( ooh, ahhh, etc). Zenfolio may add that in the future.
I started with pbase, started having problems on a recomendation i switched to Zenfolio over a year ago.
In a word I love it. Slick interface, good uploading, fantastic resising.
I upload 1024 x 750ish sized photos and can display everything from that size right down to wee tiny postage stamp images with no distortion what so ever. If I wasnt afraid of copy right infringment I could upload full 8meg files and do the same.
And about every size in between.
******Cost for a basic account is 25 bucks per year for a gig of storage.Each year you renew they up your storage limit at no extra charge (talk about Customer service).
I have 684 photos there, all roughly 1000 x 600. Of the near 2 gig I have in limit I still have 1.6 gig free.
Lots of customisable options including password protected galleries
http://jben.zenfolio.com/Until I decide to design and create my own site/domain I’ll stick with them.
Check them out, for what you get its dirt cheap.theres a free trial available. Any of us can “recomend” you. Bascially give you a code and it saves us each 5 bucks.
http://www.zenfolio.com/I would not shoot in srgb. The color space is much smaller than adobe rgb. You are losing large amounts of color data. Shoot in adobe rgb preferably in RAW or RAW + JPEG and use convert to profile in PS after you have tweaked your images to get a srgb file in the size that you want.
I would not shoot in srgb. The color space is much smaller than adobe rgb. You are losing large amounts of color data. Shoot in adobe rgb preferably in RAW or RAW + JPEG and use convert to profile in PS after you have tweaked your images to get a srgb file in the size that you want.
I also shoot in Adobe RGB – as David says, the gamut is significantly larger, so you just have more colors.
Neal, I had no idea that you were shooting in sRGB, I think it may explain an earlier issue. I agree 100% with David King; at the very least you should be shooting in RGB, for several reasons. sRGB has a smaller pallet but makes the best use of a narrower gamut. RGB has a much larger pallet and is best for broad gamut’s. Also, sRGB limits you to this smaller color pallet, even if converted to 16bit as it permanently limits the image to the use of a smaller gamut and color pallet. Very noticeable in future use of both high end inkjet printer’s, commercial photo printers (Like those at local “One’Hour Photo” intities) and commercial CMYK printer’s.
You are going to love this one. 😉 One of the gamut limitations, when comparing sRGB to RGB, is the fact that RGB goes much further into the darker shades of Green, like grass and foliage. I think this is why you had a problem post editing for the broader (color rich deeper shades of green) gamut.
I don’t know what type of package PB has on their back end but I feel pretty positive that they have some sort of software that automatically adjusts the images for better web viewing, (I don’t know for a fact though). Also, if your files are over their maximum file size, their software automatically downsizes the file and really reduces the total gamut a lot. You also have to keep in mind that we now have better monitors that will show a broader gamut than older models. The only reason to convert to sRGB 2.1 is so that it is easier for more monitors to see the best use of a narrow gamut in sRGB and broader compatibility with browser’s .
This is the most important use of shooting in sRGB and should be the only application: the photographer knows that the composition has a very narrow gamut, or purposely wants a narrow gamut. In order to get the best use of that narrow gamut, adjust to sRGB but again, it sacrifices the use of 16 bit as simply; a narrow sRGB gamut will not go into those much larger color spectrum’s.
I have attached a graphical representation that compares the two color spaces. In the left example you can see a larger color outline which is Adobe rgb gamut and the smaller gamut of srg inside. The example on the right shows Adobe rgb as a wireframe and srgb as solid color, in either case you can see how much color date you losing by using srgb.
Shoot RAW files and work on automating you workflow in PS or Lightroom to make your jpeg files for the web.
Zach, I don’t want you to think that I am ignoring your question as I am not, you addressed it to David and I know that he is more than capable of answering it. How-ever, if you don’t mind, I think that you are creating a lot of extra work for yourself. Why not edit the NEF’s and save them as Tiff’s first, you can then convert your sRGB’s and not have to reprocess the images again.
That is a great graft David!! Perfect example of the amount of additional data lost in sRGB and also illustrates the amount of actual color hues thrown away, and/or discarded.
Interesting point is to also interpret the gamut of your graphs. It helps illustrate a more equal gamut, in sRGB, with the blues and some of the magenta. The red’s and yellow’s illustrate a narrower gamut but workable. When we get to the greens and cyan’s, the sRGB is drastically drained and illustrates more than a 50% loss of the green gamut.
Translation: Much harder to WB landscape photography as the images will automatically go more to the warmer and/or colder side, in sRGB. With the addition of greens, in the photograph, the conversion and/or sRGB capture will sacrifice what little green gamut is available. What I am getting at there is this. Some talk about the loss of data or color shift after posting images on the web and after the conversion to sRGB; this is very true and the graft more than supports it. Going off of David’s graph: In sRGB, any photograph that has a dominant color frequency, such as red, yellow or orange, with green foliage or grass in it; post edit will send what little green gamut is available into the yellow’s or cyan’s. In simplified terms, desaturating for the red’s and blues will have a major negative impact on the greens.
The simplest way that I know how to explain, what I am trying to say, is this. If we think of post edit as a form of art and WB plus color correction as the process of painting art. Let’s say that our challenge is to color a picture that was painted with 100 crayons, the catch is this: We only get to use 60 crayons, kind of like we only get to work in sRGB with; 30 shades of blue, 25 shades of red and 5 shades on green. The shades of each color represent the total gamut data that we have available. Think of an increase in 8 bits to 16 bits as the same as just applying more pressure from the crayon onto the canvas, the colors are richer but the gamut does not change. This is just an attempt to try and simplify something that I have a hard time writing and explaining, the amount of crayons is not suggestive of how many colors are actually in these profiles.
With Raw and in post edit, we can focus on one color, either in RGB or CMYK, and we can sacrifice some of the crayons, for color correction, with out loosing quality in the image. This is due to the fact that we now 100’s of thousands in crayons to work with, in comparison.
That’s right Zach. The main thing is not to commit to JPEG or SRGB in camera. That way you have maximum color gamut and bit depth latitude with the file. JPEG really locks you down, if your exposure is a little hot you can’t recover like you can in RAW. You are going to tweak the image anyway. If you shoot RAW+JPEG when you go into Lightroom or Aperture PS etc you have the JPEG so your software dosen’t have to create previews so you can go right to work editing. With the price of media cards is coming down so storage is not such a big issue anymore.
Think about your magazine images. Make them look primo in Adobe RGB on your monitor and let the printer convert to the press or stock specific profile from a file with maximum color data. If you could get a copy of your printers profile you could soft proof it on your monitor.
I print and archive every thread like this here, I too strugle with getting the color I want. The closest I’ve seen is Cory’s work & no matter what I try I can’t come close.
I have been thinking about something that Zach wrote about how he does his processing. Every tutorial, suggestion and or advice, that I have seen or heard, about converting to sRGB before posting to the Internet is: “WRONG”. Let me explain how and it also supports that what Zach is doing, for his photograph Internet posts, is the only time I have seen it “CORRECT”.
If I got this correct: Zach is first converting to sRGB, for the images that he mostly posts on the web, and then post editing them. That makes more sense than anything that I have heard about converting to sRGB, it just took me a while to see it. Think about it this way: Every tutorial, that I have ever seen, explains that before you post to the web, convert to sRGB for better viewing. Well, if just about every browser is in sRGB then, it would not matter if we posted in RGB or sRGB. as they both are going to shift the same. What will happen is, we will see a color shift from what we post processed in RGB but unless we do a proof ICC Profile view (with the software on our personal monitors) in the sRGB mode, we won’t see it until the image is posted. The reason? In order to post edit for best viewing in sRGB, we should first shift the image to an sRGB ICC profile and then post edit. This is the only way to control the way the converted file is shown and/or viewed. It also gives more control over correcting the image for best possible use of the smaller sRGB color pallet. Still though, sRGB should “only” be used for images that are going to be posted on the internet, while saving the original JPG or RAW files, before the conversion.
In the end, there really is 2 completely different work flows and a simple final conversion to sRGB is far more reckless than converting to sRGB and then recorrecting the final image before posting. One correction in sRGB for web use and a separate RGB correction for all other use, it is the only thing that really makes the most sense. Took me a while to get it but I have never seen it put this way, makes perfect sense to me now and shows that all of those tutorials really are completely wrong about converting to sRGB last.
Thanks Zach, I never really thought about the fact that sRGB should never be the last action 🙂
If you have a image that is color correct to your liking in any color space then correct conversion to another color space especially a small color space like SRGB should be a given, its the whole reason for the embedding of profiles. I think you need to shoot and correct in the largest color space possible and then use convert to profile to convert to SRGB. The Apple or Adobe color engine will automatically convert to SRGB using a Look Up Table and reproduce what the file looked like on your monitor.
There is no reason why you couldn’t correct in SRGB first if you wanted to and you just used the files for the web. If I were going to spend the time working on a select file I would stay in Adobe RGB optimize and save a PSD and then save a SRGB version. If I did all my work in Aperture I would export a JPEG version and could re-edit the image as I wanted in the future without color or bit depth loss. You could probably do the same thing in Lightroom.
My main point is don’t shoot JPEG or SRGB it limits your options from the moment you make the exposure. If you shoot and save your RAW files you have a lot more control.
Boy, I am just all over the place on this one, seems that I couldn’t even agree with myself [ch9786]. I have to recant my last post, as you are correct David, about complete post processing in RGB, at the very least. I decided to talk with a person whom I consider to be one of the word’s leading experts on color correction. For those of you who may not recognize the name, Andrew Rodney, he is a very well known lecturer, consultant, as well as author, for many of the country’s largest organizations. You may recognize his book “Color Management for Photographers”, he also has a lot more great links and advice on his web site
http://www.digitaldog.net/Anyway, I was talking with Andrew about why converting to sRGB and why make it the final step? The conversion went much deeper than I thought it would but I now understand far more than I thought was involved with web posting. Just a few browser’s, like Safari and FireFox, are ICC aware, meaning they will recognize the ICC profiles of both sRGB and RGB. Most others are non ICC aware, they simply send the RGB numbers to the viewers monitor display with out knowing anything about your color space or ICC profiles. As he explained it to me, most currently owned monitors are roughly based on sRGB, “So the idea is, convert to a color space that suites the lowest common denominator.”
However, this is sure to change as Andrew also told me (He gave me permission to quote what he wrote out for me) “Now what’s interesting is that more and more affordable displays are NOT producing an sRGB behavior but instead wider gamut (closer to but not exactly Adobe RGB (1998)). What’s going to happen now is that when a user with a non ICC aware browser views images using such a display, they will not appear very well. So the idea is, as we move away from sRGB like displays, the idea of posting sRGB on the web will fall apart.”
Converting to sRGB for current optimal viewing is correct for large audiences and when there is doubt on the viewers monitor, it should also be the last step. The reason for this is, just like what we were saying earlier and what David King just explained, larger pallet and gamut. I was talking with Andrew about the much narrower gamut of green and he brought this back to my attention: “If you view say ProPhoto RGB, then a document you converted to sRGB, you’re not going to see much if any difference. Both are based on an emissive display so no difference in dynamic range. And if you’re working with an sRGB display (most are), then you can’t see anything outside that space. So the ProPhoto document might have colors that are not visible on this sRGB output device. IOW, they look pretty much the same. The greens may be much wider in terms of gamut in the ProPhoto image, but you can’t see it! This is one reason we’re seeing display manufacturers making wider gamut displays (if you converted from Adobe RGB to sRGB on such a wide gamut display, the differences would be more apparent).” This is the reason for processing in the largest ICC profile as there is color and gamut that we may not be able to see on our personal monitors but others, with wider gamut monitors can see and pro-commercial printers can print.
I found this interesting David and thought that you might too. I always thought that I could just do a soft proof in sRGB and be able to distinguish between the 2. I am sure that you have much wider gamut monitors in your studio but I found this very interesting. As Andrew was explaining to me, what difference would I expect to find when soft proofing sRGB on a monitor that is not much wider than sRGB? As he put it: “Worst of all, 95% of the people looking at that image on a web page are not using an ICC aware browser, so what they see and what you see is going to be different (there is the soft proof setup called Macintosh RGB and Windows RGB, that shows you what this image will look like on YOUR machine in a non ICC aware browser). It’s only somewhat useful because again, while using this soft proof produces a match in the browser to this soft proof, it’s only correct on your machine. Its not what the image will look like on anyone else’s machine.”
So, back to the way that I always have been doing it, (RAW-TIFF/PSD-RGB and then sometimes to sRGB). Sorry for the little side trip but I needed to wrap my hands around this sRGB thing better.
Completely agree about shooting in RAW David, way to much available information to not take full advantage of it. The extra data can also be thought of as a fail-safe feature as, the extra data can really help save a photograph as well…
Well Andrew Rodney aka the digitaldog pretty much knows it all when it comes to profiling and color management. He has a good site at digitaldog.net if any you are interested.
Going forward we will probably see more intense efforts to get really accurate color on the web. It makes for a pretty good “Green Story” as they say in the carpet business. If you can save resources energy paper etc thats a Green Story. It makes sense from a enviromental and business standpoint.
- You must be logged in to reply to this topic.