Time to Get Critical

The advances in display technology over the past 10 years or so have been incredible and have opened up a whole range of new applications, especially in allowing large electronic images to be integrated into different environments. Now, however, we can stop being amazed that such displays are available, and can start being critical of their performance. Here I briefly discuss three topics: contrast, image sizing, and cost of ownership.

At the William J. Clinton Presidential Library & Museum, a series of alcoves depicts important milestones in the Clinton presidency. 40in. LCD monitors are at the center of each display, and in order to achieve good image quality these are fed from HD sources.

Poor performance can arise from poor displays, but often it arises because of a lack of understanding. I despair a little when I see reputable projector manufacturers quoting the brightness of their projectors in lumens and claiming incredible contrast ratios as if this is what the viewer will see. (For the record, the lumen is a measure of luminous flux, and on its own cannot tell you how bright a picture will be. The measure of brightness is the candela per square meter, or, if you are an unreformed cinema person, the foot-lambert).

But the viewer appreciates an image by its contrast and grayscale, and unfortunately the performance of flat-panel displays (LCD and PDP) is not particularly good. The underlying technology makes it difficult (especially with LCD) to have a bit depth greater than eight, and often the effective bit depth is less; this leads to “crushed” blacks and whites. LCD contrast performance under low ambient light conditions is also variable.

For many signage applications based on bold graphics and saturated images, this may not represent a problem in practice — but if the displays are to show a quality full-motion video image based on high production values, the results will be disappointing.

Because advances in display performance are so rapid, and because, in fact, some manufacturers have successfully addressed the problems, it is essential to evaluate candidate displays before committing to them. But this evaluation must be done using the intended source material and in the environment in which the display will be used. This shows that the realized performance of displays is highly dependent on ambient lighting conditions, and that the achievement of a good image is often in the user's hands. A display can be made to look much better not by raising the brightness, but by improving the contrast — and this means reducing the ambient light falling on the display.

When big LCD projectors first appeared the LCD panels within them were of video resolution. This meant that one of the first auditorium LCD projectors — regarded as amazing at the time — could only give its full light output on a PAL 625-line video image. If it showed NTSC you lost 16 percent of the light (conversely, NTSC displays omitted 16 percent of the image when showing PAL). Now we take it for granted that all projectors are fitted with image-resizing chips that allow displays to accept any input, and have it resized to match the native resolution of the display.

But image resizing is not a simple matter, and in practice the results can be variable. At the William J. Clinton Presidential Library & Museum in Little Rock, Ark., for example, archive video footage is shown on many different sizes of display — typically 20in., 30in., and 40in. LCD displays. The programs were well-produced by Cortina Productions on a resolution-independent production platform with final output at MPEG-2 standard definition. This looked fine on the smaller displays, but terrible on the 40in. displays, which had nothing to do with the professional production but everything to do with display resizing deficiency.

The solution was to ensure that the display was fed a signal at its native resolution. Because the Clinton Library production was resolution independent, it was a simple matter to output a high definition version of the program — and, hey presto, the display looked magnificent. This required the use of an HD player, but these are now available at very sensible prices and people looking to get the very best out of large flat panel and projected displays should really only use HD sources.

The long ticket desk at the Wales Millennium Centre in Cardiff, Wales, features a 52ft.-wide back projection display. It uses DLP projectors (Projection Design F1) rated for 24/7 operation and fitted with UHP lamps to ensure reasonable running costs. The offset optics and wide-angle lens allows the whole display to occupy a depth of only 37in.

Another critical subject is cost of ownership. Many display applications, such as lobby displays, permanent exhibits, digital signage, and control rooms, require 12/7, if not 24/7, operation. There are many products (especially projectors) that are simply not up to this work — although they may be absolutely fine for use in training rooms or occasionally used corporate facilities.

There are two issues in cost of ownership: the longevity of the display itself and the cost of consumables such as projector lamps and backlights. Budget-conscious museum directors are understandably upset when they find the LCD projectors need replacing after three years because of panel failure, and apoplectic when they face lamp bills for hundreds or thousands of dollars a month.

Thus users and systems integrators need to face up to the cost of ownership issue from the outset. Projectors designed for 24/7 use will, not surprisingly, have a higher initial cost, but will pay for themselves very quickly. It is dangerous to generalize, since progress can quickly invalidate received opinion. However, the current situation with regard to heavy-duty projection is that LCD is not ideal for this work. The jury is out on LCoS because there is simply not enough track record to form an opinion, and DLP is currently the best option.

As far as lamps are concerned, designers will often do better to base displays on smaller images requiring lamps in the 100W to 200W range, where they can use projectors based on UHP lamp technology (or its equivalent from other manufacturers) where “rated” lives of 2,000 hours to 8,000 hours are the norm. (But beware weasel words. Rated life in this field is the time taken either for a lamp to drop to half its rated output or for a given large quantity of lamps to have suffered 50 percent failure — so practical lamp life is less.)

For big projectors, it is best to go for projectors with “cinema-style” lamp houses, since these can use lower-cost conventional xenon lamps.

Robert Simpson is founding director of the Electrosonic Group. He confesses to nearly 45 years in the AV business, is a member of SMPTE, SID, and SPIE, and holds the ICIA Distinguished Achievement Award.

To comment on this article, email the Video Systems editorial staff at vsfeedback@primediabusiness.com.