The factory color temperature settings ranged from below 6000K in the Low setting to approximately 9000K in High. In its Mid-Low setting, the Pioneer produced a very respectable gray scale, though the x/y coordinates of the white points were off a bit more than the desired ±0.004 deviation from the D6500 standard. (They were off in a minus green direction; if you have to be off the mark, minus green is far less visible than plus green!). While the color temperature numbers after calibration don't look all that different from the before readings in the accompanying chart, they were definitely closer to an accurate D6500. They were off by more than 0.004 at only one point from 20 IRE to 95 IRE (0.005 in the y axis, at 85IRE).

After calibrating the set through an HDMI input, I found that the readings from a component input, using the same Manual color temperature calibration settings, were very close to the HDMI numbers.

As noted earlier, the PRO-940HD offers two Color Space settings. In Color Space 1 red and green were considerably off the mark, with green, in particular, badly oversaturated. Color Space 2, however, offered color points very close to the standard. Numbers aside, the visible differences between these two options were obvious on both test patterns and typical program material. Some viewers will like the punchier look of Color Space 1, but it is clearly more creative than accurate.

The Pioneer's overscan measured between 1% and 2% with 720p and 1080i, in both component and HDMI. In 480i/p, this increased to 2.5-3.5% in component, and a rather high 3.5-5% in HDMI.

Peak contrast ratios are normally measured using a full-screen, 100 IRE peak white test pattern rather than a 100 IRE window. But all plasma displays decrease their output dramatically as the amount of white in the image increases due to power supply limitations. So we normally take two peak contrast readings on plasmas, one for full screen white, the other for a white window. Its peak contrast measured 1,568:1 with a white window (64.27 foot-Lamberts peak white/0.041fL video black) and 561:1 with full screen white (22.99fL/0.041fL)—good numbers for any flat panel display, plasma or LCD. The set will also reproduce below black and above white.

The Pioneer's resolution, through the component inputs using burst patterns from the AccuPel test pattern generator, was disappointing in 480i, with the 6.75MHz maximum burst frequency missing in action. The same was true of the 13.5MHz maximum frequency at 480p. Things definitely picked up at 720p and 1080i, however, with both showing response to the maximum burst of 37.1MHz. Though the response was clearly rolling off at that frequency, the resolution lines were still clearly visible.

Using an HDMI source, the 480i response was far better than in component. It was down a bit in level at 6.75MHz, but definitely present. I obtained the same result for 13.5MHz at 480p. At both 720p and 1080i the 37.1MHz maximum burst was down in level, but visible.