• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

Color space conversion - a caution

Andrew Rodney

New member
You can guess which one is more accurate ...

I have no idea which is more accurate, I wasn’t at the scene, nor did I take spectral measurements. I can tell you, based on viewing this on a color managed web browser which I prefer visually. But I didn’t capture or render the image so it might be moot. Neither is a deal breaker. We could look at the image or the Histogram to make such decisions, but initially I’m going to look at the image.
 

Andrew Rodney

New member
Let me state that I’m playing devils advocate a bit here and agree with Doug its useful to know what’s going on.

We have an “issue” with a few RGB working spaces that are not yet V4 savvy although as Doug found, a “beta” if you will is available for sRGB. Why Adobe has not built and supported any V4 working space profiles in Photoshop I can’t answer.

We don’t live in a perfect world and we certainly don’t live in a prefect color management world! That said, we often need to convert from larger to smaller color spaces. The lack of a perceptual table for the rarer cases we deal with working spaces isn’t a huge big deal (I’ll tell you one that’s much worse and really need attention later). The rare times I’d be moving from ProPhoto to sRGB is to pop images on the web. We all know what a mess that is in terms of color management! We all know how forgiving it is. Other than that (or sending someone an email with an image), the clipping when it occurs just doesn’t seem to be a big deal. Yes, I’d like more control with V4 profiles but I suspect they are not ready for prime time.

Even dealing with output using printer profiles, we often have to clip colors. This group is savvy enough to know that they want to start with a wider gamut than sRGB. Otherwise someone could argue, just set your cameras to JPEG sRGB, you’ll never clip colors going to another RGB working space. One could also argue that the clipping happened just after you clicked the shutter and the raw data was converted to sRGB in the camera. So we really are clipping again.

If I start from raw and encode in ProPhoto, I might go off to a press (and the gamut differences there between going off to newpaper vs. an Indigo is huge). We will clip colors. I had a Epson printer with 8 colored inks that until Monday was my widest gamut output device. Then a Canon i6300 with 11 inks showed up and my gamut clipping (or compression) needs just differed a bit.

Again, why we don’t have more control converting from RGB working space to working space in Photoshop using V4 profiles is not clear.

Now onto a real problem lots of people face that doesn’t require histograms or new revisions to the ICC spec. The “my prints are too dark” issue and some kooky ideas as to why this is an issue and the fix. I provide you this blog post:

http://lightroomkillertips.com/2010/video-the-trick-to-getting-brigher-prints/#comment-14876

We have a well known imaging expert suggesting that we alter the RGB values of all our images for output to fix a problem that can be addressed with proper display calibration target values. I would welcome your thoughts on this because I think an article needs to be written that address this misconception (or the misconception that the prints are too dark, which they certainly can be for a number of reasons, some having nothing to do with the RGB values in said document).
 

Mike Shimwell

New member
...

Now onto a real problem lots of people face that doesn’t require histograms or new revisions to the ICC spec. The “my prints are too dark” issue and some kooky ideas as to why this is an issue and the fix. I provide you this blog post:

http://lightroomkillertips.com/2010/video-the-trick-to-getting-brigher-prints/#comment-14876

We have a well known imaging expert suggesting that we alter the RGB values of all our images for output to fix a problem that can be addressed with proper display calibration target values. I would welcome your thoughts on this because I think an article needs to be written that address this misconception (or the misconception that the prints are too dark, which they certainly can be for a number of reasons, some having nothing to do with the RGB values in said document).


Andrew, I followed your link and I'm flabergasted. Agree, get your calibration right first. My monitor brightness is at 8% to get about 85 cd/m^2. The default is 75%, which is way too much for me.

In terms of printers, the other issue is that of blocking up shadows for low luminance colours. There are various ways to address this - outback photo had some ideas about adjusting rgb values to 'map out' those colours. The HP I'm using allows you to reduce ink limits, which is a far better approach if you've the time and patience to test. One corollary is that different settings and profiles can be optimal for monochrome and (wide gamut) colour work as sometimes to maximise colour gamut (measured in terms of max saturation) you need more ink than is ideal for a very linear grey scale in the 000 to xxx range (where x may be bewteen 10 and 30!

Mike
 

Andrew Rodney

New member
Andrew, I followed your link and I'm flabergasted.

I plan to write an article which I'll either post on my site or perhaps Luminous Landscape which will be free to all. The article I'm envisioning would not only look into the dark print issue but go farther and have a cause and effect analysis. Something like "if you see this, go to step A, if you see that, go to step B" approach. I would encourage any and everyone here to provide peer review and suggestion before it posts.

To give you an idea just how complex this all is, and how a simple answer like "just do this" isn't effective, here's an initial taste of what I have in mind.

Your prints are too dark: Are they too dark everywhere you view them? If yes, go to step 1. If not, go to step 6.
Step 1. we need to determine if the darkness is caused by the RGB values in your document(s) or perhaps an issue with a profile, printer driver, printer maintenance issue etc. To do this, you should print out a reference image of known quality RGB values. Here are two sites where you can download test images at no cost:

http://www.pixl.dk/download/pixl_testimage2002_rgb.jpg
http://www.digitaldog.net/files/Prin...t_file.jpg.zip

Output one or both images exactly as you have output images you've been having issues with. If the image is no longer too dark, then the RGB values in your images are the root cause. Now go to step 7.

Step 6. If your prints only appear too dark when you view them compared to your display, the issue is most likely a disconnect between the print viewing conditions and the display calibration. Whatever luminance value you set for calibration (defined usually in cd/m2), lower that, recalibrate and examine the prints. Do they still appear too dark? If yes, continue to lower the display luminance. IF you cannot lower the values any farther, can you increase the luminance of the print viewing area?
----------------
Get the idea? The question and answer path should nail where the issue is happening, it doesn't assume that the RGB values are at fault because it could be an outdated print driver. It could be a clogged ink head. It could be the incorrect setting in Photoshop (or LR's) print dialogs. It could be a poor ICC profile. The idea would be to find, through process of elimination where the problem really lies! We fix that problem, not assume the problem and thus the answer is to alter the document values unless we know that's where the problem is coming from.

Suggestions welcome. Beta testers welcome.
 

Cem_Usakligil

Well-known member
...Suggestions welcome. Beta testers welcome.
Hi Andrew,

Coincidentally I have been following this thread in the LuLa forums. It is absolutely mind boggling. So yes please, I would be more than happy if I can participate and help you getting this document/product into place. Count me in as a tester/reviewer.

Cheers,
 

Asher Kelman

OPF Owner/Editor-in-Chief
Me too!

I am impressed with the opportunity to get a logical diagnostic path. It's generous of you to provide it gratis! We'd be delighted to post it as a lead article as well.

We do have a few very competent fellows here and it will be exciting to see how this progresses.

Good job!

Asher
 

Andrew Rodney

New member
Hi Andrew,
Coincidentally I have been following this thread in the LuLa forums. It is absolutely mind boggling. So yes please, I would be more than happy if I can participate and help you getting this document/product into place. Count me in as a tester/reviewer.

Cheers,


Got a very rough draft in MS Word if you two are still interested. Contact me off line.
 
Top