Hitachi HDPJ 52 LCD Projector Measurements

Tests and Calibration

506Hitachi.1.jpg

All the results shown below (unless noted otherwise) were taken with the iris set to 3/Auto 1, the High Contrast gamma setting selected, and the Whisper mode on Normal. The sample tested was different than the one Steven Stone used for his viewing tests.

The Before curve in the accompanying chart shows the best out-of-box result (using the factory 6500K setting). It drifts further from 6500K (not to mention D6500) as the light level of the source grows dimmer. After calibration, however, the results were very close to the D6500 standard. The actual coordinates of the white points were within +/- 0.002 from 30 IRE to 95 IRE—a superb result. The Hitachi offers color temperature adjustments not only at the top and bottom of the brightness range, but in the middle as well—a welcome first in my experience.

The red, green, and blue color points, as in most projectors we've tested, weren't precisely on the money. All the colors were a little oversaturated, with green deviating the most from both the NTSC and ATSC standards. But the results were nevertheless closer to the correct points than we usually measure.

The Hitachi clipped below black signals (sourced from the Digital Video Essentials DVD on my Pioneer DV-79AVi DVD player), but did show below black in component. With the Overscan control set to the lowest level (oddly, a setting of 10 out of 10), overscan measured zero in both HDMI and component at all tested resolutions (480i/p. 720p. 1080i)—another first for our tests.

HDMI resolution was measured using my AccuPel HDG-3000 test generator. In all resolutions (480i/p. 720p. 1080i) the Hitachi responded out to the bandwidth limit of each format. At the maximum 37.1MHz burst in 720p, the response was a little uneven, but still visible in both luminance (black and white) and color (with the color response rolled off more than the luminance—as is typical). At 1080i, the 37.1MHz response was rolled off noticeably, but the resolution lines were still visible.

In component, the resolution was fine at 480i, noisy but otherwise good at the 13.5MHz maximum at 480p, visible at 37.1MHz at 720p but showing heavy noise and strobing at that frequency, and present and accounted for but down in level at 37.1MHz and 1080i.

At an iris setting of 3/Auto-1 I measured a peak contrast ratio of 1359:1 (10.87 foot-Lamberts peak white/0.008fL video black, measured in component). The contrast ratio degraded gradually as the iris was opened further. A wide open iris (setting of 10, in Auto-1) resulted in a peak contrast of 864:1, but also provided a peak white level of 26.78 fL—useful for watching sports with a little room lighting if the screen isn't overly large. (All of the readings here were taken on my 78-inch wide, Stewart Studiotek 130 projection screen, gain 1.3).

Finally, while Steven did much of his viewing using outboard scalers, I don't believe most users will need one with the Hitachi. Most projectors today have at least respectable video processing; with the processing chips available today there really is no excuse for anything less. And the Hitachi delivered; its deinterlacing and scaling are excellent. They sailed through my battery of de-interlacing and scaling tests on both the Faroudja and HQV Benchmark test DVDs, including the waving flag, jaggies tests, and unflagged 3/2 pulldown test. The projector also grabbed a passing grade on the difficult Benchmark test that features a pan across empty bleachers. It scored only fair on the Faroudja bad edit test, and showed a few ripples in the 2:2 and 3:2 cadence tests, but in general performed about as well as any built-in de-interlacer/scaler I've tested with the possible exception of the Silicon Optix device featured in the Yamaha DPS-1300 DLP projector (over 3x the price of the Hitachi).

X