DSLRUsers.com

That feels better........well just a little.

I have for such a long time struggled with home printing despite purchasing 'decent' equipment. I curently have an Epson 3800 and exclusively use Ilford Gallerie Smooth pearl or gloss paper along with Ilford ICC profiles, and to be honest a lot of the time the results are pretty good but not always what I expect. (I generally have to increase the exposure on any image by as much as 1/2 a stop to prevent the resulting print being too dark.

I decided, after much reading that I really should purchase a monitor calibrator (have been putting it off for too long), and so I purchased a X-Rite colourmunki photo which was delivered yesterday. Now my problems will be resolved, or so I thought!

The colurmunki profiles the monitor and the printer so that the output from the printer matches what is seen on the monitor. Guess what.......it doesn't

I followed the setup procedure to the letter (or so I thought) but alas, I suppose I have done something wrong during the setup so will need to do a little more tinkering.

The usual cause for this is having the monitor too bright. Remember that monitors and printers use a completely different way of mixing colour. Monitors use an additive process ( If you mix all colours you end up with white) while printers use a subtractive process ( Mix all colours -> black or thereabouts)

If you have a Mac though there is not a lot you can do about it. Apple seems to be turning away from their traditional Graphics User power base in favour of the starry eyed. (Overly bright, only gloss screens etc etc)

That said, the Canson papers I usually use matches pretty well, but the other day I had occasion to use Epson Semi Gloss. The prints turned out too dark. Sigh.

If you use Lightroom, the new version (LR4) allows you to preset a brightness offset for printing. I can't admit to trying it myself yet though.

The most important thing is monitor profiling. I would presume that the Ilford printer profiles are fairly accurate to start with. Getting a colorimeter is not all of the battle because some of the cheaper software packages they come with do not allow you to set the brightness. That makes those ones next to useless to my mind. I presume that does not apply to the ColorMunki and so the question is what are you specifying the profile for.

I have an NEC2690 monitor that I profile for printing and an NEC2090 monitor that I profile for the web, though I usually just prepare images for print and don't worry about the sRGB conversion for the web (ie I don't soft proof for that). I also have an XRite Pulse for printer profiling and for most images I can just print off the soft proof with no problems. Mind you, that will never be true for all images and you will always have to consider whether your print really matches your vision and to consider both overall and regional adjustments.

I use NEC SpectraviewII as profiling software. The choices the ColorMunki offers may differ but I use the following settings which may be a useful starting point for you insofar as they are transferable:
NEC2690 for print: White point = D55; Gamma = 2.2; Intensity 100cd/m2; 200:1; Color gamut = Native (full),
NEC2090 for web: White point = D65; Gamma = 2.2; Intensity 140cd/m2; Contrast ratio = monitor default; Color gamut = Native (full).

Mr Darcy wrote:If you have a Mac though there is not a lot you can do about it. Apple seems to be turning away from their traditional Graphics User power base in favour of the starry eyed. (Overly bright, only gloss screens etc etc)

There isn't a lot you need to do, you can simply turn the brightness down, like with any other monitor. The ColorMunki software actually helps you set a proper max brightness for print colour matching.

Cheers
Steffen.

Steffen wrote:you can simply turn the brightness down,

True on my Macbook
My iMac varies between Nova and SuperNova

... but surely the Colormunki profiling software will not only do this but also allow you to set a specific value, irrespective of operating system (not that I know anything of Max).

The ColourMuki does adjust the screen brightness, when I profile my screens I tell the software that I have no hardware brightness control and the software does the rest. Works pretty well in matching the screen to the printer, an Epson R2400.

Thanks for the feedback......so all is not lost

So brightness of the monitor is not controlled by the calibration device , it seemed to me that when the profile was applied the brightness did reduce slightly from where I had previously had it set.

I am using LR and have just upgraded to LR4 last week. During some of my test prints last night a used the brightness control in the print module and set it to +50 which produced a more acceptable match, but this is still going to be a bit 'hit and miss'.

I also saw better results using the Ilford profiles rather than the print profile generated by the colourmunki. I thought about getting some Epson paper to see if there is any improvement, but based on Greg's comment perhaps not! I Will have another play tonight if time permits. BTW, I have a PC with a Samsung Syncmaster 223 monitor.

Geoff M wrote:So brightness of the monitor is not controlled by the calibration device , it seemed to me that when the profile was applied the brightness did reduce slightly from where I had previously had it set.

I did a tiny bit of research and you can definitely set the brightness in cd/m2. Almost certainly, I would think, you should use advanced mode so you can be sure you're getting appropriate settings, especially from your comments, brightness. This should be no more than 120cd/m2, I suggested 100cd/m2, some might prefer even lower and the most accurate level will even differ between matte and glossy papers.

You may find this video useful. It runs for an hour and it takes quite a while to get to profiling the monitor but it seems to be thorough and is specific to the ColorMunki (I'm running it and it's halfway through at the moment).

Geoff M wrote:I am using LR and have just upgraded to LR4 last week. During some of my test prints last night a used the brightness control in the print module and set it to +50 which produced a more acceptable match, but this is still going to be a bit 'hit and miss'.

You shouldn't be doing this. This is for people who don't have a functional colorimeter and who are willing to settle for inaccurate results, just better than no adjustment at all. Instead, you need to work out how to get a decent profile.

Geoff M wrote:I also saw better results using the Ilford profiles rather than the print profile generated by the colourmunki. I thought about getting some Epson paper to see if there is any improvement, but based on Greg's comment perhaps not! I Will have another play tonight if time permits.

Before I got the XRite Pulse, I had some profiles made for me for a couple of Epson papers by a prominent US vendor. I found that these were only subtly different from the standard Epson profiles. There was a time when supplied profiles for third-party vendors could be inaccurate, especially for non-Epson printers but this is probably less true by now. I would be very surprised if it were true for Ilford papers on an Epson 3800. So standard profiles should work OK and you should get a slight advantage from generating a profile for your own printer. I have gone from Epson semi-gloss as my main paper to Ilford Gold Fibre Silk and find my custom profile is usually pretty accurate on my 3880.

Geoff M wrote:BTW, I have a PC with a Samsung Syncmaster 223 monitor.

I don't know anything about this monitor though I notice it was released at least 5 years ago. Samsung can produce good monitors but I don't know how good their cheaper ones are if that is what this is. The best monitors are NEC and Eizo though they are not cheap. Probably the monitor is fine but there is a possibility you might benefit from a newer and better one.

Mr Darcy wrote:My iMac varies between Nova and SuperNova

Ah yes, I've heard of those (though never seen one myself). My son's iMac (21.5") can be dialled down to very dim.

Cheers
Steffen.

Thanks all for the feedback, all very useful. Tonight I have re-profiled the monitor using 'advanced' settings within the colourmunki software as opposed to the 'easy' settings. Within the 'advanced' settings you can manually set the luminance level and target white point.

For this profile tryout I set the luminance level to 100 and the target white point to D65 (default), I am using the calibration device to measure ambient light which was measured at 98.85 lux and as a result the software auto set the monitor to 80cd/m. I saved the profie and using the previous printer profile (using same paper and ink as before, so guessed that it was not necessary to repeat).

In lightroom I selected the colourmunki generated print profile, set the intent to perceptual, (as I always do) and ran off a print. Alas I have what appears to be the exact same result as previously. i.e. with the same overall, warmer and darker print to that seen on the monitor! I then ran off a print with the same settings once again except the print profile which I substituted for the Ilford profile. Once again the result is too warm and dark (in comparison to the monitor) but is a closer match than the colourmunki profile.

Question: When I completed the printer profiling, I used the ICC profile for the Ilford paper to print the test chart, should I have used this? Perhaps this is adding to the problem.

When I use Pulse to generate a printer profile, the software handles it all and no profiles are involved. Certainly I would think you shouldn't be using an Ilford profile for this, rather no profile at all. The colormunki documentation should tell you.

You say you get the same results printing from a standard Ilford profile. It would be best to get a reasonable result from that first before worrying about a custom profile. The display profile is likely to be your main problem and the less variables involved the better. I suggest you use a standard test image to eliminate more variables and for greater clarity. There are notes on how to interpret that on the page I linked to.

OK, what other variables are there?

Perceptual vs relative colorithmic won't make much difference. My rule of thumb is that relative colorithmic is better for most images; perceptual is better where you may have out of gamut colours (eg live music, flowers).
You should have draft mode printing unchecked.
You should have print adjustment unchecked.
You should have the correct profile specified.
In print settings, you should have the correct media type specified, probably semigloss or glossy depending on the paper (Ilford online documentation will tell you that).
Colour should be Colour.
Print Quality won't matter much.
[Paper Config] should be default.
Mode should be Custom/ Off (No colour adjustment) - that could well be your problem.
In the worst case, if all of those things are satisfied, there might be a problem with your monitor, that it's too far out to be calibrated.

I can't see the first page of this post????? So I am not sure which monitor you use.

As murray said, what monitor are you using Greg? I have seen funny results when I tried to calibrate old TF monitors with a calibrator.

That's a glitch we have been discussing in this post, which seems to have turned from an image post to an error discussion post. When clicking on this post in Recent Topics or New Posts you would have clicked on the 2 for Page 2 instead of the 1 or the thread name link. In either latter case the whole thread appears on a single page.

surenj wrote:what monitor are you using Greg?

Do you mean me or Geoff?
If me, an Eizo sx2762W. The brightness issue I described refers to the the iMac screen which is an All-In-One box.
I bought the Eizo in part to get around this problem. It was also an insanely good deal at last year's PMA.

If you mean Geoff:

Samsung Syncmaster 223 monitor.

What program are you printing out of? apparently in cs6 and possibly cs5 (not sure as I don't print my images at home off my mac) you go into print options and you can select your paper type etc but unless you switch off the Adobe color auto changes (have to turn it off otherwise its on all the time) the printer gets confused and even though you've selected the paper type etc the Adobe program modifies the image to what it thinks is best to print. But if you were to turn it off it prints as you have it on your computer/mac. As I said i can't say for sure, but it worked for a friend who was using an epson printer also who had similar issues and no longer has them after he made these changes. Hopefully someone on here might understand what i am trying to say and explain it in the correct way with the correct terms, but it might be worth a look, a simple fix would be a relief no doubt!

Which I think is what I said in terms of the 3800/3880 driver as "Mode should be Custom/ Off (No colour adjustment)". And yes, that could well be the main problem.

Murray Foote wrote:OK, what other variables are there?

Perceptual vs relative colorithmic won't make much difference. My rule of thumb is that relative colorithmic is better for most images; perceptual is better where you may have out of gamut colours (eg live music, flowers).
You should have draft mode printing unchecked.
You should have print adjustment unchecked.
You should have the correct profile specified.
In print settings, you should have the correct media type specified, probably semigloss or glossy depending on the paper (Ilford online documentation will tell you that).
Colour should be Colour.
Print Quality won't matter much.
[Paper Config] should be default.
Mode should be Custom/ Off (No colour adjustment) - that could well be your problem.
In the worst case, if all of those things are satisfied, there might be a problem with your monitor, that it's too far out to be calibrated.

Murray - thankyou for your input and advice. All the varialbles noted above are confirmed and those are the options which I am selecting with the exception of the print intent (perceptual/relative), as previously stated I always have printed with the percuptual option checked (read somwhere that this is the best option) but I have run off another test print selecting relative and I think we are onto a winner as I hnow have the closest match yet. There is a difference to that seen on the monitor but it is negligible.

The difference between perceptual and relative colorithmic rendering is that perceptual compresses the colour gamut of the image to accommodate out-of-gamut colours whereas relative colorithmic clips the gamut of the image to preserve colours that are already inside that gamut.

This means that if you are photographing a band with bright lights (preferably not LEDs which have their own problems), perceptual will better preserve those bright colours. However if those colours are outside the gamut of your monitor, then obviously you will not be able to see an accurate preview of them and in the worst case may have to resort to extensive hard proofing (test prints). Also, perceptual can cause colour shifts because in-gamut colours get moved to accomodate out-of-gamut colours. Relative colorithmic works best with most images because most images are within the gamut of a good monitor and your printer.

For most images you should be able to see the difference between rendering intents by soft-proofing in Lightroom 4 (or Photoshop). My guess is that if you are having this problem with an image that does not obviously have colours that are likely to be out of gamut then your monitor may not be good enough. Either its gamut may be too small or its colour rendition inaccurate or it has deteriorated over time. If so, then the most "economical" high quality replacement monitor at the same size is probably the NEC P221W which displays 96% of Adobe RGB and costs about $600 from discount online stores. This way, you should get a pretty accurate image on screen and have confidence in the monitor. You can get still better monitors from NEC and EIZO but they will cost more. Cheaper monitors are likely to range from poor to not as good and it can be difficult to know how good they really are for photographic purposes.

Looks like a replacement monitor will be on the shopping list next! There is always something to spend your money on in this hobby of ours

They had a deal on Eizo's at The Digital Show in Melbourne on the weekend where every purchase came with a free Wacom Intuos5.

Wink wrote:They had a deal on Eizo's at The Digital Show in Melbourne on the weekend where every purchase came with a free Wacom Intuos5.

If you can get in on this deal, go for it. My Eizo is brilliant. The only trap is that they don't have a VGA port.
Mine has mini Display Port, Display Port & DVI - D.

Any Eizo is going to be a really good monitor but I think an NEC is a better proposition. I have seen it suggested that the relatively new NEC PA range is as good as the Eizo Coloredge range (therefore better than the Eizo Flexscan range) and they are much cheaper, cheaper also than the Flexscan range. Several significant web people use NEC monitors including colour management guru Andrew Rodney.

My earlier suggestion for a monitor assumed you wanted to economise. For the best practical monitor I would suggest the NEC PA271 which is a 27" monitor. If you get a larger monitor you won't be coming back and saying "I wish I got a smaller monitor". And you can also use the Samsung as a secondary monitor provided your configuration supports that. It won't particularly need to be colour managed; you can use it for tools and Outlook and ancilliary programs.

The NEC monitors (including the P221W) also permit direct hardware calibration using Spectraview software, as for the Eizo Coloredge range (with different software) but not the Eizo Flexscan range. There is a lot of information about NEC and Eizo on the Image Science site. They currently have specials on bundling their monitors with Spectraview software which are probably worth considering but expire at the end of May. Spectraview II will still work with the ColorMunki spectrophotometer.