Working in the optoelectronics
industry, I often get press releases from companies working on better LCD
displays on their latest advances. This is all well and good, but the display
industry suffers from rampant "specmanship
" that only performs a disservice to the industry and misleads the user.
A recent press release from Sharp (Tokyo, Japan) touts the creation of the ASV Premium LCD display panel, with a size of 37 inches (diagonal), a resolution of 1920 x 1080 pixels, and a brightness of 500 candelas per square meter (a figure also known as nits). They also tout a display contrast ratio of one million to one.
This is a bullshit spec, as are 90% of all specifications given with LCD, Plasma, and any other non-CRT display technology in existence. (The CRT guys woulds lie too if their tech weren't so mature.)
Contrast ratio, brightness, and screen-performance information are generated by using highly-tailored test patterns and performance benchmarks that have little to do with the real image delivered, but a lot to do with published specs.
For example, depending on how the core technology of the display responds, the contrast ratio test may consist of a white square, box, or dot on a black field, or a measured sequence of black-to-white screens, with the measured difference in brightness given as the contrast ratio.
The best analogy is loudspeaker specs, which unless they are linked to recognized performance specifications (like frequency response given as plus/minus decibel variance from 20 to 20,000 Hz), are completely misleading. A speaker advertised as delivering 500 Watts may only be able to handle that much power as a transient, and even then a speaker can only "deliver" the power fed into it, which means you also need a 500-W amplifier.
A very good example was at the latest Society for Information Display (www.sid.org) show. Samsung (Seoul, Korea) had both the largest LCD and the largest Plasma in existence at the show, and although the listed brightness and contrast "specs" for the Plasma was greater (they were on cards attached to the respective device), the LCD obviously had a brighter and sharper image in operation. True, the blacks were better in the Plasma, but that was the only visible distinction to the discerning viewer and only shows how little a guarantor of performance a high contrast rating is.
News of better and better displays is certainly encouraging information, and will certainly result in better-performing displays appearing on our shelves soon. But to look at any given display specification (or any other consumer electronic device's performance specs) and shout "halleluia!" is being overly generous.
The sad part is if companies started to use real video performance to generate the data, the numbers would fall so low that they would suffer against the companies that lie. Which is why everyone lies.