Help - Color Management Confusion!

Question

I'm hoping someone can give me some clear guidance as to how to configure my software to integrate color management so that what I see on my monitor is comparable to what I print or post on the web.  I've read all I can find on-line and looked through a number of books and magazines, but as my reading increases, so does my confusion - too many menus, options, links and connections.  I'll describe what I have in terms of hardware and software in as much detail as possible.  My hope is that someone can help me figure out how to coordinate color management across all platforms and outputs.
I run an HP desktop with an ASUS IPD monitor.  I have ColorMunki and have reguarly conducted their diagnostic, which gives me an updated monitor profile every two weeks - when a new profile is created, it becomes my default under Windows 7 Control Panel Color Management.  I print with an Epson Stylus Photo R1900 with Epson paper and ink.  I use the printer and paper profiles supplied by Epson for the R1900 and specific papers.
On the other end, I have Nikon photo equipment.  I set the color space on my cameras to ProPhoto and I shoot exclusively in Nikon raw format - .NEF.  I shoot a D90 and a D7000 with 12M and 16M files respectively.  I download directly from the SD card using Windows Explorer and then import into Lightroom 3.
I have Lightroom 3 and Photoshop CS5 Extended.  I use Lightroom to import raw files into the catalog and to do basic editing.  From there, I publish to Smugmug as .TIFF files - directly from Lightroom.  I also export files from Lightroom to Photoshop CS5 Extended and save them as .PSD and .JPEG files.  The files that I export to Smugmug look pretty good - similar to what I see on my screen.  My real conundrum is with the Epson printer and Lightroom/Photoshop.  What I see on my monitor is very different from what prints.  From both platforms.  As a rule, the shots are too dark and the colors are often very muted.  This is especially true with images of people and skin tones.  They often come out looking pale gray or blue.
Right now, my solution is to tweak and print until I get something close enough to print, while the image on my monitor looks horrible.  Needless to say, this is not a viable solution for the long term - it is too expensive and time consuming.  I've tried to read and then adjust adjust both programs, but I'm confused as to what to set where.  I've taken to explorting Lightroom files into Photoshop to use the soft-proof capability.  it works better but still not great and not consistent.  When I do the Photoshop soft-proof, I see signficant portions of most images as "out of gamut>"  I also notice that these areas are often the ones with odd colors.
My problem, (well, one of them) is the multitude of menus that have color management inputs - how do I make sure they are all working together and reinforcing each other instead of working at cross purposes?  In Lightroom, I have Page Setup and Print Preferences, as well as Color Management options in the Print Module right hand panel.  In Photoshop, I have Print and Page setups, as well as Color Management menus.  There are also the Proof menus to consider.
In Photoshop, softproofing, I think I ned to select teh printer profile so that Photoshop is showing me how the printer will interpret the color space.  I also think I need to be consistent in how I define the workspace - from Nikon to Lightroom to Photoshop to Epson.  If I use ProPhoto with my camera do I need to use that in every circumstance?  Should I switch to Adobe RGB?  I have printer setting color space turned off in both LR and PS.
Can someone walk me through any or all of this?  Right now, all of these moving parts is making me dizzy.  It's said that a little knowledge is a dangerous thing.  This proves the point.  Three months ago, I had color management problems but had no idea what such a thing was.  Now, I know a lot more about it, but my results are no better - worse in some respects.  Anyone who can help will earn major Karma points and hae my undying gratitude for several months.  Sorry - that's all the compensation I can offer!
Thanks in advance for any help.

Answer 1

Lundberg02 wrote:
is that the hand of God in the picture?
That was a nearly mature tornado forming in Nebraska, and it dissipated at the last moment because at that time the sun went down and the temperature dropped rapidly.  I was glad, because I was camping in a travel trailer at the time, and everyone knows trailers attract tornados. 
Lundberg02 wrote:
Please explain why and under what circumstances anyone would want to use a device independent profile as a monitor profile, which should be a device dependent profile.
Here's one example, to answer your specific question:
You have a monitor that provides sRGB performance (e.g., via a specific sRGB setting).  You judge (and/or ensure via your own calibration) that it provides satisfactory sRGB color accuracy for your needs, so you set the monitor profile to sRGB, which is accurate in this case. 
What does this do for you?
Images displayed in Internet Explorer, which assumes your monitor is sRGB regardless of your profile, are now properly color-managed because you have made IE's assumption valid.  By the way, Microsoft does not appear to be going to change this behavior any time soon.
Untagged images, considering a majority of untagged images assume sRGB encoding, are displayed properly by apps that just pass them through to the display.  Some browsers do this.
sRGB images (the majority) are displayed properly by non-color-managed apps.  If you choose to set your preferred working space to sRGB, then your own processed images will display properly using your non-color-managed apps.  Such apps include various viewers, the thumbnails in Windows Explorer, etc.
The sRGB IEC61966-2.1 profile is a clean, well-formed color profile and works well in virtually every color-managed application. It's the Windows default, so you can imagine most apps are very well tested with it.
Screen grabs are already in the sRGB color space, so if your working space is sRGB then you gain simplicity and don't have gotchas when mixing screen grabs back into your workflow.
Since the color-gamut is not wide, the distance between adjacent colors in a 24 bit color environment is smaller - the display of gradients looks smoother, and you might not crave 30 bit color quite so much.
There are other subtle advantages as well, simply because so much of software development throughout history has assumed the representation of color images on computers is sRGB. It's akin to "going with the flow".
In short, setting up a soup-to-nuts sRGB system means more images match more often across more applications with a system set up this way.  If I'm not mistaken, that's about what the original poster is asking about, which is why RikRamsay's response is not unreasonable.
And there are, of course, some specific disadvantages to doing this. 
For one thing, one does not have the direct ability to work with a wider gamut of colors.  There are those who wish to work in wider gamut color spaces, have wide gamut monitors, and wide gamut printers.  Brighter, better managed colors may well help such a person set his/her work off from the crowd.  This is starting to become more and more important in this day and age of better and better wide gamut hardware.
It's not straightforward to set up a system so that its sRGB response is accurate, though it is doable.  Can you trust the factory sRGB calibration of a monitor that advertises sRGB response characteristics?
-Noel