We've settled into our new offices at 95 Howard St now, and things seem to be running reasonably smoothly again after the initial chaos of the move. We've got a lot more room now, so we're keeping more stock and we're able to do training and so forth much more easily. We've also got a new person on board, who is sitting at our new front desk. Her name is Kylie, and we trust you'll make her welcome as she comes up to speed with the large variety of products and services we offer. She's learning fast and should bring new efficiency to the retail side of our business, as well as help us out with all the day to day office tasks.

We've just finished many, many prints for the Art Show and the APPA awards, and we wish all of you entering the shows many sales and much success with your work! We've seen some really fantastic stuff come through this year, it's quite clear that the overall level of comfort with digital techniques is rising rapidly - it's also a relief so see less formulaic work than in past years with the heavily vignetted, over processed look so popular just a year or two ago quickly falling out of favour (and not a moment too soon if you ask me - if there's one theme lately it's that work has been more cheerful than the last year or two - I've had several clients claim this is an effect of the 'KRudd' tornado!). Whether or not that change in tastes makes it through to the the actual judging remains to be seen, of course!

My Monitor Upgrade and getting down to some serious soft proofing!

As mentioned above I have recently upgraded my monitor from an Eizo S2110W to an Eizo CG241W. I was not expecting a big functional change, I really just bought the screen so that I could give proper demonstrations to client's interested in the technology. But I've been stunned at what I can now achieve, and what I am now seeing, with this amazing screen. (And all of this amazing technology is available in the new incredibly cheap CG222W as well now!).

I have spent some real time exploring hardware based soft proofing and I'm finding it to be a very effective process. Typically soft proofing is performed by using a printer profile which contains information about the black point (d-max) of a particular printing system. This is used by Photoshop to re-map the blacks from your monitors black point to the prints black point. What results is the display of a low contrast simulation on an inherently high contrast device - as monitors always have a much higher contrast ratio than prints. My CG241W monitor, for example, can display images at an 850:1 contrast ratio. The best prints, made on long tonal range papers, never exceed about 250:1 contrast ratio (and 200:1 or less is typical). So this means my monitor natively displays imagery with over 4 times the the amount of contrast than my best prints can ever express. So when I soft proof, I am asking my eye to accept a sudden drop in contrast from 850:1 to 200:1. And as anyone who has done this process knows - this is a very difficult process for the eye to adapt to. (The process of an advanced soft proof is detailed here). Various tricks are employed to reduce this difficulty for the eye, such as hiding all the photoshop palettes and setting the screen to a neutral mid tone grey, but it's never completely successful in my experience.

One of the major things an Eizo CG screen brings to the table is the ability to manipulate the contrast ratio of an LCD screen without huge unpleasant side effects, and with measured accuracy. With an Eizo CG screen, you can use Eizo's excellent and very easy to use ColorNavigator software to control your monitors black point and set a specific contrast ratio as desired - for example 200:1. And because of the fantastic high bit depth circuitry in the Eizo monitor, this adjustment is carried out without visible side effects (try this same adjustment on a typical Dell or Apple etc screen and you will immediately induce huge banding in your greyscale gradients). So at the end of the process you are left with your monitor displaying the exact contrast ratio you require - you can even input a measured paper white point and set the white point of your monitor to that of your paper so the entire display is adjusted to be as close as possible match to your final output medium! This is an amazing piece of technology that takes soft proofing and colour management to a whole new level. Indeed, if you use multiple papers, you can create multiple 'hardware based soft proofs' for your papers and dynamically flick between them at the click of a mouse using ColorNavigator. It's pretty amazing stuff and because the eye does not have to go through a huge contrast adjustment when you turn on soft proofing, it makes the whole process far, far easier. And this means more accurate prints - and that's the goal, after all.

Another thing I have noticed about my new screen is how amazingly even it is, across its entire field of display. Because the screen area is so large I regularly have more than one image up at a time, or more commonly two views on the same image (for example, I will have one view displaying the whole image and another view zoomed into the area I am retouching). The two views on the file match exactly in both density and colour terms - no matter where they are on the screen. It's uncanny. Try this same thing on an iMac, for example, and you'll quickly see that the screen varies quite visibly in both density and colour on either side of the monitor. The monitor actually came with an individual measurement sheet showing that across it's entire field there is essentially no visible difference in both colour and density...very impressive stuff as that was perhaps the one real complaint I had with my older S2110W which did have some variance across its field (still way better than most screens, though!).

There's another common problem we encounter here - I must be asked about this several times a week - that of the overly bright monitor. Lots of recently released monitors can not control their brightness levels properly - with many of them being in excess of 200 candelas bright even when set to their minimum brightness. This is a ludicrous figure if paper is the thing you're trying to simulate. All this light flooding through your images will make you feel like your images are much brighter than they really are, and when you print them you will find them too dark. It's very easily solved - as long as your monitor can reach a sensible brightness figure and hold it consistently - again an area where the Eizo monitors excel. It's also just much less straining on your eyes to work in front of a monitor set to a reasonable brightness level (80 to 130 candelas are the typical values most people calibrate to).

A Quick Discussion On Shadow Detail

One thing I am often asked is why can't I easily see differentiation between RGB 15,15,15 and RGB 0,0,0 on my custom profiled printer?

Well, the first thing to realise is often that you can see this difference, but only when you print substantial areas of these tones and only when they are near each other and only when they are not surrounded by white. The human eye is particularly bad at seeing difference in dark tones when they are surrounded by white - and this is an easily testable fact. Make a print of a very dark grey tone - say RGB 30,30,30 - on a small piece of paper. Now make two small frames, one from bright white paper and one from black card. Place them immediately around your print and then swap between the two frames - you'll see your grey tone change brightness levels right in front of your eyes - it's quite hard to believe at first. There is also a famous picture which shows just how big a difference surrounding tones can have on our perception of a colour. Pretty amazing, huh? So - one thing to try is your printer test again, but cut all the white paper around the edge off and do your patches at a much bigger size. You may well see that the difference you thought wasn't there actually is!

The second reason may be simply that you're using a very cheap printing device that is really quite bad at shadow detail. A profile will go some way to helping this but profiles are good at small nudges, not huge sweeping corrections, so it may be that your printer just isn't up to the job. And if you're printing with the relative colorimetric intent (rather than perceptual), you are pretty much asking Photoshop to clip your shadow detail - try printing again with the perceptual intent and you should see in most cases much better shadow detail (more about this here).

The third reason may be your choice of working RGB space. Most photographers in Australia use AdobeRGB and with good reason as it is an excellent general purpose colour space. However, it has quite a significant compression to its shadows - more so than some other colour spaces. You can check this by creating a white to black gradient in Photoshop and then running your mouse over it with your info set to LAB mode. You will see from this that AdobeRGb is not visually linear in deep shadows - you have to go all the way up to RGB 25,25,25 to get to LAB 5 - or put simply you are numerically about 10% along the scale when in visual terms you are only at 5% of the scale. That is, there is not a lot of correlation between the AdobeRGB numbers and what we might expect - you would think that if we moved 10% up the scale in numbers, we'd move the same amount in brightness - but we don't. It's an easy trap and although your monitor profile should be sorting this out for you and placing everything in the appropriate spot visually, it tends to be an area where both cheaper calibrators and cheaper monitors are weak. So often, on your monitor, the light flowing through these parts of your image will make them look significantly brighter than these areas are going to look in print - even the most accurate print in the world. Does this mean AdobeRGB is a bad colour space? No, it just means that you need to be more careful with placing shadow detail in AdobeRGB than in, say, sRGB. And if you're testing a printer using a printer test file, you should make sure it's been properly designed for the colour space it is in - most of them are not. The PDI printer test file on our downloads page is quite good.

The final thing to look at is your lighting - printer profiles are built to simulate D50 lighting at a whopping level of brightness (2000 Lux if I recall correctly). This is *very* bright light. So when assessing your prints/profiles, make sure you are looking at them under a bright lights (a solux bulb in a couple of lamps is perfect - notes on how to set this up here). If you're under typical room lights at night, you will simply not be seeing shadow detail that is most likely easily visible under a brighter light.