Test Bench for Web JVC HD-70FH96 70-inch LCoS projection HDTV by David Katzmaier

Unless indicated otherwise, all tests were conducted with 1080i-format signals via the TV's HDMI input. Color temperature (Low color temperature, Theater Pro mode before/after calibration) Low window (20-IRE): 6,678/6,791 K High window (80-IRE): 6,227/6,529 K Brightness (100-IRE window before/after calibration): 95.6/24.8 ftL

The JVC HD-70HF96's out-of-the-box color temperature in Theater Pro mode came reasonably close to the standard of 6,500 K, although it was a bit less accurate than other 1080p TVs I've tested recently, with the darkest areas the closest and the majority of the grayscale measuring too warm. Grayscale tracking was relatively linear, trending toward red as the image grew brighter. Calibration brought the entire scale much closer to the standard, varying by an average of only 90 K, which is excellent performance. (Calibration needs to be performed by a qualified technician, so discuss it with your dealer before purchase, or contact the Imaging Science Foundation at imagingscience.com or 561-997-9073.)

I was surprised to discover that the Theater Pro preset limited peak brightness to just 26 footlamberts (ftL), perfect for a completely darkened home theater (the 24.8-ftL peak brightness number above was measured with the Dynamic preset). As a result, the contrast ratio I measured after calibration wasn't as high as on typical DLP 1080p sets, though the JVC's average black level, measured on the checkerboard pattern from the Sencore VP403, was the best I've seen yet at 0.101 ftL. The set clipped detail in brighter areas at a contrast setting higher than 0 but was fine at -1 and lower.

The Smart Picture setting dynamically adjusted contrast according to picture content, so I left it off after setup. The Dynamic Gamma setting actually made blacks brighter in dim scenes, so I left it off as well. The HD-70FH96 failed to hold a constant level of black in all picture situations but did maintain it well in most cases. With standard-def sources, the VNR noise-reduction settings worked well to cut down on mosquito noise in lower-quality broadcasts, but I couldn't detect any effect when I engaged the MPEG noise-reduction feature. Deinterlacing showed evidence of 2:3 pulldown but took longer than normal to detect film-based material.

The JVC was among the best TVs I've seen in its reproduction of multiburst resolution patterns. It fully resolved 1080i sources through its HDMI input and scaled lesser resolutions extremely well. Even more impressively, component-video sources looked almost as good - it was very difficult to tell the difference. Like most other 1080p HDTVs I've reviewed in the past year, the JVC couldn't accept 1080p sources via any input.

Artificial edge enhancement was nonexistent with sharpness reduced to all the way to -30. The ramp pattern was very smooth from brightest to darkest. Color decoding was accurate for both HDTV and standard-def, although there was a slight overaccentuation of red with HD. There was a significant amount of red and green fringing along fine lines both horizontally and vertically, which worsened toward the edges of the screen. This wouldn't be visible from normal seating distances with regular program material and will vary between individual televisions. Focus also became noticeably fuzzier toward the screen edges. Geometry was very good on my review sample, with a minor trapezoid effect that made the bottom of the image a few pixels wider than the top. Overscan was minimal, averaging about 2% on all sides, but the entire image was shifted about 2% to the right and therefore off-center. Uniformity was much improved over previous LCoS HDTVs, although I did notice a faint yellowish discoloration along the right edge of the screen and even fainter darker red areas toward the corners at all light levels. These would be invisible with most content.

Back to Homepage What's New on S&V