Tuesday, January 31, 2006

A Tale of Two LCDs


A while back, I was excited that NEC and LaCie were starting to market their high-end displays based on their color accuracy, as measured against a known standard-- mainly AdobeRGB. But the question I didn’t have an answer to is how do these new, high-end LCD displays compare to the CRT monitors that have been the standard for years. While I don’t yet have a full answer, I do have some data to share with you.

One of the lessons I’ve learned in 15 years of digital imaging is that I have to buy a product and try it for myself to see how useful it is. That meant it was time to make a purchase. The display that most intrigued me was the NEC LCD1980SXi-SV. I’ve been using NEC (and Mitsubishi, who is now a part of NEC) displays for about ten years, and my experience has been that they are the best displays made for the price. This led me to favor NEC over the similar LaCie display.

Looking at the NEC website I noticed that there are two versions of the LCD1980SXi listed. The first is the standard LCD1980SXi. The specs for this model don’t make any claim as to its color accuracy, even though it is listed as a “Color Calibration Display.” The second version listed is the LCD1980SXi-SV which includes a calibration kit, and it claims it can display 69% of AdobeRGB.

My hunch was that these two displays were the exact same piece of hardware, with the only difference being the inclusion of a calibration unit. I e-mailed NEC tech support and they confirmed my suspicion. This was an important piece of data because:

A. I already own a calibration kit (the Gretag Eye-One Photo, a excellent piece of hardware), and
B. The model without the calibration kit lists for about $150 less.

Armed with this information, I went on-line and found a LCD1980SXi for $700 including shipping.

At the same time, Terrance Reimer, one of our master printmakers, purchased a ViewSonic VA912b for his own use at home (street price about $350), giving us another data point in this conversation.

There are several ways we can evaluate a display. One way is by measuring its performance mathematically with a spectrophotometer. Another way is by experience....using the display, and then forming an opinion on it based on the total experience of the tester. While measurements are important and valuable, experience is the more valuable of the two. A wine may have the proper measurements, but not taste good to the palette...and so it is with making a photograph. A lot of “perfect” measurements do not, on their own, add up to a great print, or to the creative decisions required to achieve that goal.

TESTING

The LCD1980SXi currently sits on Terrance’s desk, but with less than a week of use, we have little experience to share with you. So, in this post, we'll look at the measurements.

One of the things I was most curious to know was how the LCD1980SXi compared to the NEC Multisync FE700+ monitors we’ve been using for the last several years. Since NEC has not published the percentage of AdobeRGB that the FE700 can display, I had to find another way to measure and compare the two displays.
That task fell to a software application called ColorThink, published by CHROMiX.

ColorThink lets you compare profiles from different devices by plotting them on a color chart. For this phase of testing, I plotted four display profiles, along with AdobeRGB, as a reference point. They are:

NEC LCD1980SXi
ViewSonic VA912b LCD
NEC FE700+ CRT
Apple Powerbook 15 inch 1.5 ghz LCD

The graphs are displayed with this post. Click on it for a larger view.

In the Yxy graph, the plots for the LCD1980SXi , VA912b, and FE700+ lay practically on top of one another, which tells us they have basically the same gamuts because they can display substantially the same colors. Looking at the Lab graph, we see a similar outcome.

What does all of this mean?

For one thing, it means that in the case of the $700 LCD1980SXi , the $350 VA912b, and the $150 FE700+ (no longer in production), there is very little difference in the colors that can be displayed, despite a vast difference in price.

I wasn’t too surprised that the LCD1980SXi was virtually the same as the FE700+ as it’s much easier (cheaper) to produce a high quality CRT than it is to do the same with LCD technology.

I was surprised that the $350 ViewSonic VA912b could produce such accurate color, but despite its color accuracy, it suffers from a common “feature” of LCDs--a limited viewing angle, which in practice is far less than that the angle stated in the product specifications--at least when considering it for meeting high-end color correction needs. When viewed straight-on, it looks very good, but if you shift your head around, the color changes, and that’s a bad thing when trying to make critical color and density decisions. The LCD1980SXi has far less shift and is quite acceptable, but it can also shift when viewed at the right angle. The CRT has no color shift from any angle.

You should also note the poor performance of the LCD in my PowerBook. I know from first-hand experience that it is not as accurate as my FE700, and the measurements help confirm it. I threw it in, just for curiosity's sake.

Also note that AdobeRGB has colors that none of these displays can show. The most accurate reference for what our files really look like is still a print on a photo-quality output device like the Chromira or Epson 9800. While there are now monitors that can display 100% of AdobeRGB if you have $5,000 burning a hole in your pocket, years and years of experience, and thousands of prints, attest to the fact that you can produce the finest quality prints with a monitor that is the equivalent of the NEC FE700+. Read my previous tip for some more thoughts on the value of $5,000 displays.

Now, as I said earlier, the real test for me is experience. You’ll have to wait for that because experience takes time. Terrance has the LCD1980SXi on his desk at work and the VA912b at home, and the printers are comparing actual prints to the LCD1980SXi and FE700 to evaluate and understand it. When we have something more to share, we’ll post it.

Saturday, January 28, 2006

New Data Quantifies the Accuracy of Color Displays

This is the first of a series of posts we’ll be making about monitors for imaging. It was previously posted on our Photoshop Tip page, and I am re-posting it here, since it is the jumping-off point for my next post.

In digital imaging, your display is a critical tool because it is the only way you can view your image without making a print. This is even more true if you are using a digital camera. With film, you have a physical original you can view on a light table. With a digital file, you only have numbers, which must be converted to light to be displayed. If that conversion is not done accurately, you have no way of knowing what you actually captured. Was the dress red or orange? Without an accurate display, you won't know for sure.

Several display manufacturers are making it easier to know just how accurately a display shows color. This helps set our expectations, so we don't expect an imperfect display to be perfect. It also helps us make a more informed decision when purchasing a display.

Specifically, NEC and LaCie are starting to list the percentage of different RGB colorspaces their high-end displays are capable of reproducing. After 15 years of spending sometimes outrageous sums on “color accurate” displays, it’s nice to finally have some meaningful data on how accurately a given display can reproduce color before I make a purchase. It allows us to easily compare displays, and make a more educated decision.

Here is a quick rundown on some current offerings and their color ratings:

19 inch

NEC MultiSync LCD1980SXi-BK-SVII $999
70% of NTSC
69% of AdobeRGB

LaCie 319 LCD $1249
72% NTSC


21 inch

NEC MultiSync LCD2180UX-BK-SVII $1749
72% of NTSC
69% of AdobeRGB

NEC MultiSync LCD2180WGLEDBKSV $6999
100% SMPTE-C

LaCie 321 LCD $1849
72% NTSC

NEC Diamondtron UWG RDF225WG CRT $4999
97.6% of the AdobeRGB
93.3% of NTSC




But what do all of these numbers mean?

The displays listed above show the degree of accuracy for three separate color spaces: AdobeRGB, NTSC, and SMPTE-C. As you might guess from the numbers, these color spaces are fairly similar. They are not the same, but they are close, and they give us a good point of reference for how one display compares to another. I’m guessing here, but I don’t think there would be much difference between a display that shows 70% of AdobeRGB versus one that shows 72%. I would hope (and expect) that there would be a difference between a $1749 display that can show 72% of AdobeRGB and a $6999 display that can show 100% of Adobe RGB.

As a side note, if in early 2006 it costs $999 to achieve 69% of AdobeRGB, how much of AdobeRGB do you think your less expensive or older LCD display can reproduce?

Now, before you e-mail me, let me answer your next question....is it worth it to spend $6999 on a display? For most photographers, probably not. In a prepress environment where a press run costs from a few thousand dollars to tens of thousands of dollars and you are doing many of these runs day in, day out, it is of huge importance to be able to proof on-screen as accurately as possible. It can save huge amounts of time and money because it is very expensive to proof on a press. However, for the average photographer engaged in noncommercial, personal expression, who has the time to make and evaluate proofs, and can print them at a very low cost, it's not as critical. While you want a really good display, the money to get "the best" would be better spent on time off to make more photographs.

And while it’s great to know how much of AdobeRGB a display can reproduce, even 100% of AdobeRGB is smaller than what a Chromira with Fuji Crystal Archive or an Epson 9800 can print. In the end, we still need to make hard proofs to really judge the color. Hard proofs and experience have and will continue to let us bridge the deficiencies of displays.
This brings us to the truth that even very accurate displays are merely guides--the ultimate definition of what a photograph should look like must come from the print itself. Since displays transmit light, and prints reflect it, even the best displays can't perfectly show what your final print will look like.

It's very rare that a photographer has an unlimited budget, so most of us have to be very wise about where and how we spend our money since there are limitless things to spend money on when it comes to photography. We have to make sure that we are actully getting some real value for our money. My experience using CRT displays over the last 12 years tells me that even a good display can be used to make the finest quality prints. This is because perfect equipment doesn’t give you vision or experience.

Ansel Adams was once said, "There is nothing worse than a sharp image of a fuzzy concept." This is as true today as when Ansel said it. Vision and experience can't be bought, and displays are only tools that help the vision that we've developed. They don't help us take better pictures...they give us a more accurate view of what we've already captured.

The next step is for WCi to buy one of these displays and see how good they really are. We’ll keep you posted.

p.s. Did you notice that the CRT that can show 97.6% of Adobe RGB is $2000 less than the very best LCD display? It’s still cheaper to make a very accurate CRT than it is to make a very accurate LCD. That’s changing slowly, but it just goes to show how expensive it is to make a really accurate LCD display.

Welcome!

Welcome to the West Coast Imaging blog! We’re stepping into the blogosphere because there is so much we want to talk about that doesn’t fit into a neatly organized box on our website. For instance, I’m about to post about some testing we’re doing with LCDs as compared to CRTs. In the past, I might have been inclined to call it a tip, but it’s not really. It’s valuable information, but it’s not an end-all-be-all answer. It’s information that is part of the journey of making expressive prints, with technology that is ever-changing...and sometimes even improving! We’re all on this journey together, so pass the road map, put on some good music, and let’s hit the road!