Colour profiling, soft proofing and my current digital chain

[IWC Doppel]

Hi guys,

I have an epson 3880, iMac 27" 2010 vintage, and I shoot mostly B&W, I have experimented and got to an acceptable position with three chosen papers for B&W.

I am now starting to tackle colour. My needs are printing what I see on screen, absolute accuracy is not my goal as I often adjust to give a more Willam Eggleston look, or more Fuji velvia. My 'digital chain' is Leica M9 DNG, I set the menu for adobeRGB, not sure if this affects DNG to be honest, but I do this anyway. Then into LR4.3 on my iMac, then to my epson 3880. I have an okay environment as the room is painted a dark, grey green and my one window has a thick blind, my desk is dark too. I have just bought a couple of 4700K solux spots and my screen is calibrated by a color Munki display.

My IMAC is set to photo RGB and I have used softproofing, I have a few questions after watching the LuLa videos on camera to print and Lightroom ( great vids, invaluable help )

1. Given from what I understand my iMac is somewhere between sRGB and adobe, what I see may not be what I get ( some I can't potentially see ), is this less of an issue than seeing what you wont get ?
2. I can see low but not super low IRE ( I can see 5-6, when calibrated) would I see more with something like a Eizo or top end NEC ? I say this as I will use mostly for B&W in any case.
3. With softproofing in LR4, I find using matt papers I find they look to washed out on screen, I see more contrast and punch on paper, can i use the softproofing for colour gamut only and will the colour on screen then be accurate ?
4. As at present I am only using matt papers as I have yet to bond with gloss or semigloss, are these papers showing colours below sRGB in any case and would I be better to stick to consider more carefully my digital chain and choices ? Effectively stop worrying about gamut above sRGB
5. Regarding B&W, would I find any difference when printing on matt paper shooting in adobe RGB or sRGB ?

Many thanks

[Mark D Segal]

If your display is properly calibrated and profiled and you want a predictable match between what you see on the display and what comes out of the printer, you want to examine adjust your photograph under softproofing using the same printer/paper profile in the softproof mode that you will be using for making the print. Your 3880 can see beyond sRGB, so you are sacrificing gamut by limiting your colour space to sRGB, and depending on the colours of the photograph, this would show-up in less saturated colour than your printer can give you. I would urge you to adopt AT LEAST ARGB(98) as your colour working space, if not ProPhoto RGB. Matte papers will not give you the depth of shadow detail that you can obtain using gloss/luster/baryta papers, for example Ilford Gold Fibre Silk, or Epson Exhibition Fiber, to name just a couple. There is a trade-off between the "matte feel" of the paper and the dynamic range you can get from a printed photograph by using these papers; but this is a matter of personal preference, and it is true, again subjective, that some photos look better on some papers than others.

[IWC Doppel]

Thanks, I'll stick with prophoto RGB in LR.

Any idea what gamut is achievable with matt papers ? I have printed with exhibition fibre, but prefer matt with a-b's so far. For the papers I have settled on for B&W (HFA Museum Etching, Epson cold less natural and at the cheaper end fotospeed nat text fine art) are textured and natural not bright. These do work well for me for B&W, but a little too subdued for colour. My two example shots are one of Venice with challenging blues and a lot of detail and one of a bright Orange GT3 RS with a little road tar on top on a polished surface to highlight both details of the debris and the shine and smoothness of the paint.

What I am hoping to achieve is what I am happy with on screen being replicated on paper, I know that's blindingly obvious, but I was thinking if until I have hit the print button and seen both images on print I have only ever seen the colour gamut of the display, which I am happy with (perhaps blissful ignorance)

Of course if my display is actually close to adobe98RGB then I may have a greater challenge with achieving that on certain papers.

By the way I have used the service with fotospeed/online papers for free custom colour icc's and expect to do so when I have a settled on a small (perhaps two) choices of colour matt (I expect) papers.

[Czornyj]

ad. 1) It's not an issue in case of B&W

ad. 2) It's only a problem if you can see it on a print

ad. 3) The dark surrounding can make shadow details on a screen look brighter. Try to put some white mat board behind the screen, an illuminate it with SoLux light bulbs. The monitor background shouldn't be darker than 50% gray (RGB 119,119,119 in sRGB):


ad. 4) You can get saturated colors outside of sRGB on both paper. Again - not an issue in case of B&W

ad. 5) Of course not (even on glossy papers).

[Mark D Segal]

What I am hoping to achieve is what I am happy with on screen being replicated on paper, I know that's blindingly obvious, but I was thinking if until I have hit the print button and seen both images on print I have only ever seen the colour gamut of the display, which I am happy with (perhaps blissful ignorance)

The main differences between your screen and paper are:
-transmitted light versus reflected light
-darker blacks and more shadow detail on the screen
-the display white point may be brighter than the paper white point, depending
-depending on the colour gamut of the display, likely somewhat smaller than that of your 3880, depending on the paper.

Of course, if your display gamut is sRGB, you will not see the colours beyond that gamut, but your printer will print them to the extent the printer/paper combination makes that possible. Does it matter much? My display, for example, is very close to ARGB(98) (NEC PA271), but my printer gamut exceeds it in certain portions of the colour space (Epson 4900 with Ilford Gold Fibre Silk). I find this much less of an issue than the four factors mentioned above.

Having regard to your basic concern of improving the ease of matching display with print, I have found that matte papers make these differences more challenging to overcome, and using a paper such as Gold Fibre Silk or its likes in other brands make these differences much easier to overcome, via softproofing of course in all cases.

[IWC Doppel]

Interesting point about back lighting.

Regarding my B&W matching as I use ABW mode softproofing is more difficult, I have ended up using the brightness and contrast adjsutent in the print module and find the final image close enough to a non softproofed 100cd/m2 calibrated display in a low lit room. My adjustments have ended up at +26 brightness and +34 contrast. Sounds Heath Robinson but it works for me in terms of tracking grey scales and test images, plus 'just' matching the test charts and not adding contrast lacked the punch when printed on matt textured paper. I do add a little shadow detail in the advanced section to bring back a little of the crushed black at low ire.

My problem is I am a little too much iterative tweaking over careful measurement

[Mark D Segal]

If you are doing a fair amount of B&W printing, and you want to be assured of more controllable screen to print matching I'd forget about the ABW driver and either just use Lightroom's B&W conversion with the standard driver, or if you want fancier, Nik Silver Efex Pro 2. I haven't found a visibly compelling reason to use ABW, although a number of users claim more neutrality - the key test is what you see looking at the prints.

[IWC Doppel]

I develop in LR into B&W and have as yet not looked to take on Nik softwares efex 2, I am reasonably competent now with LR4 and moved on from presets that emulate films (or try to) I spend 95% of my time in LR developing DNG files into B&W and at the point now where I am now considering the Nik option. I really dislike the very obvious over 'structured' look of a lot of images I have seen, they are to me very 'digital' and HDR in a bad way. But having watched a lot of videos I can see the flexibility and have a genuine interest in this product.

I just think nearly £200 is a lot for what is ultimately a plug in.

My print adjsutent are solely to match what I see on screen. I am very happy with the results using the ABW driver and some minor adjustments, I would like, split toning, but cheat a little by cooling down the tone a tad and using warmer papers. I might try a custom colour icc and compare.

[Mark D Segal]

If you are satisfied with the results you get in B&W from LR4, don't bother with Silver Efex - not worth the money if you don't think you need it. Note, it is not an HDR application, it is only a B&W conversion application.

Custom ICC profiles won't help with B&W using the ABW driver. They may help doing colour work with the standard Epson driver and an LR4 or Photoshop workflow. They may also help for doing B&W within LR, insofar as it is good to start with a properly adjusted colour image (for luminosity) before using the B&W function.

[l_d_allan]

Any idea what gamut is achievable with matt papers?

Imatest has "GamutVision" for $99 with 20 free trials. I played with it to check the "gamut volume" (GV) for the .icc's of a number papers, mostly for curiousity.
http://store.imatest.com/gamutvision.html

What I recall is that the best Canon pigment printers with OEM ink and OEM glossy paper got a GV > 800,000 when you input the Canon supplied .icc file to GamutVision. Perhaps that would be more or less close to 100% of the Adobe98 gamut? Or not?

Seems like the better glossy profiles for the Canon 9000-2 had volumes of > 700,000. This would be significantly less of the Adobe98 gamut, but I may have a flawed idea of what "gamut volume" means. I recall the .icc's for matte papers being much, much lower ... like ~ 400,000 to 500,000.

But it has been a while and I didn't get anything close to expertise with GamutVision. I didn't purchase the $99 GamutVision (not enough functionality for the amount, for me).

[Mark D Segal]

By comparison, in ColorThink Pro's Profile Inspector the gamut volume of Adobe RGB98 is 1,207,520. The Gamut Volume of the profile for an Epson 4900/Ilford Gold Fibre Silk is 966,999. I generated this profile with the discontinued but otherwise excellent XRite Pulse Elite kit. The gamut volume for the same printer/paper combination I ordered from a top-of-the-line external service provider using the best of the current XRite offerings has a gamut volume of 925,006. The difference between 966K and 925K would hardly be noticeable in most photographs. More important perhaps is the shape of the gamut volume, and the smoothness of tonal gradations the profiles provide. While the gamut volume of the Epson 4900/Ilford GFS combination is overall smaller than the ARGB98 working space as given in these calculations from ColorThink Pro, the fact is that because of differing gamut shape, there are colours outside the gamut of ARGB98 that the Epson 4900 can print. Using ProPhoto RGB in a print workflow avoids such issues. 

[Iliah]

Imatest has "GamutVision" for $99 with 20 free trials.

You can view gamut and volume statistics using free http://www.argyllcms.com/doc/iccgamut.html and http://www.argyllcms.com/doc/viewgam.html

[IWC Doppel]

Thanks, interesting information.

On a practical note, I assume if my LR is set to pro photo RGB and my camera output adobeRGB and I simply use the LR printing workflow I will get the maximum gamut available ? Ie I am not clipping or compressing along the way 

[Mark D Segal]

Set your camera to produce raw files (to be imported into Lightroom). In the Print dialog you have a choice whether to let the printer manage colour, or to use an ICC Profile. Choose the latter option and select the correct profile for the printer and paper you are using.

[l_d_allan]

On a practical note, I assume if my LR is set to pro photo RGB and my camera output adobeRGB and I simply use the LR printing workflow I will get the maximum gamut available ? Ie I am not clipping or compressing along the way 

Correct, but my understanding is that if you shoot RAW, then the camera setting for AdobeRGB or sRGB doesn't matter. That camera setting only matters if you shoot jpeg. Just to check ... are you shooting RAW?

To illustrate: my Canon 5dm2 has 14-bits per RGB channel. The RawDigger utility shows that there are about 14,000 discrete values per channel. You could think of those values as the camera's "gamut". ACR and LR leave those alone. As you post-process RAW with Exposure, Clarity, Spot-Healing, etc, those don't change the bit, but rather are a series of .xmp instructions.

If I stay in RAW within ACR or LR, the image file is untagged. For eventual output to print or web, at some point I'll export/save as .jpg, .psd, or .tif from ACR or LR with a selected Color-Space. In ACR, the options are sRGB, Adobe98, or ProPhoto from ACR.   "Edit + Assign Profile" in Photoshop is another way to associate a color-space with an image file.

This LR semi-newbie has the impression that LR assumes a very wide gamut, almost-ProPhoto work-space in order to accomplish printing.

So in a round-about way, I agree that the OP is retaining maximum gamut with a LR printing workflow ... assuming they are using RAW.

[Iliah]

if you shoot RAW, then the camera setting for AdobeRGB or sRGB doesn't matter.

There is a small detail here, your histogram and overexposure warning depend on colour space settings. So it is better to have camera set to AdobeRGB.

[Jim Kasson]

This LR semi-newbie has the impression that LR assumes a very wide gamut, almost-ProPhoto work-space in order to accomplish printing.

A good assumption. For computations, LR uses a color space with the ProPhoto RGB primaries and white point, but a gamma of one. The LR color space question is a bit more complicated than that; here's a good explanation.

Jim

[digitaldog]

3. With softproofing in LR4, I find using matt papers I find they look to washed out on screen, I see more contrast and punch on paper, can i use the softproofing for colour gamut only and will the colour on screen then be accurate ?

You need to tweak the display calibration (and/or print viewing conditions) to produce a visual match. IOW, in soft proof mode, that appearance should match the print as close as the technology allows.

http://www.luminous-landscape.com/tutorials/why_are_my_prints_too_dark.shtml

[Mark D Segal]

A good assumption. For computations, LR uses a color space with the ProPhoto RGB primaries and white point, but a gamma of one. The LR color space question is a bit more complicated than that; here's a good explanation.

Jim

The reference material you provide is indeed authoritative, but you will note the header to the paper saying it is for readers of the book (Martin Evening's Lightroom 4 book) and not intended for distribution. I would highly recommend readers making or planning to make extensive use of Lightroom to purchase a copy of Martin's book. Of all the technical manuals I have on hand, this is the one I make the most use of and so far there hasn't been a Lightroom question on my mind that isn't answered there.

[Rhossydd]

you will note the header to the paper saying it is for readers of the book (Martin Evening's Lightroom 4 book) and not intended for distribution.

That's the publisher's own web site and openly available.
As read; the wording of the header only specifies that it shouldn't be resold or distributed.

There's a long precedent of publishers making a chapter or two of books available to allow prospective purchasers a taste of the book to encourage sales. It's happened with most of Martin's previous books and very good they are too.