|
•Read all the test benches |
Except where noted, we used the digital (HDMI or DVI) input with sources set to 720p resolution. Contrast ratio was measured after calibration using a checkerboard pattern. We used a signal generator and DVDs to check primary color accuracy, color decoding, resolution, edge enhancement, uniformity, black-level retention, and overscan. Standard-definition video processing was evaluated with the Silicon Optix HQV Benchmark DVD. Test hardware included a Denon DVD-5910 DVD player, a Sencore VP403 signal generator, a Dish Network DVR-942 satellite HD receiver, and component-video and HDMI distribution amplifiers.
Included Test Benches:
BenQ PE7700 | Hitachi HDPJ52
InFocus Play Big IN76 | Mitsubishi HC3000U
Optoma HD72 | Panisonic AE900U
Samsung SP-H710AE | Sharp XV-Z3000
Color temperature (before/after calibration)
Low window (30 IRE): 6,776/6,409 K
High window (80 IRE): 7,254/6,470 K
Brightness (100-IRE window before/after calibration): 17/11.9
Before calibration the BenQ PE7700's grayscale varied by an average of 580 kelvin from the 6500K standard, which was fourth-best of the group of eight. Afterward it varied by just 51K, tied for second-best. (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.) Its real-world contrast ratio of 109.2:1 was also fourth-best. Primary-color accuracy was good, with green being the worst offender, and the color decoder desaturated green somewhat for both standard-def and high-def sources. When set to its optimal Real aspect ratio, the PE7700 resolved every line of a 720p multiburst patten via HDMI (but not via component-video), although the outputs from 1080i, 480p, and 480i inputs were relatively soft. Edge enhancement was negligible from all inputs; uniformity was average for a DLP, with the bottom edge of the image visibly brighter. Black levels did not fluctuate depending on program content, overscan was 0% in the Real aspect ratio mode, and focus was excellent and fringing negligible on convergence patterns. Standard-def video processing was excellent on the HQV jaggies patterns, smoothing out diagonal lines extremely well, but 2:3 pulldown detection was the worst in the group, taking nearly 1 second to engage.
See the review of the BenQ PE7700
Color temperature (before/after calibration)
Low window (30 IRE): 7,268/6,598 K
High window (80 IRE): 7,367/6,485 K
Brightness (100-IRE window before/after calibration): 8.2/11.1
Before calibration the Hitachi HDPJ52's grayscale varied by an average of 786K from the standard, which was sixth-best of the group. Afterward it varied by a fifth-best 140K. Its real-world contrast ratio of 98.94:1 was second-worst. Primary-color accuracy was fairly good, although green was second-worst in the group, and color decoding was very accurate for both SD and HD sources. The HDPJ52 could not fully resolve 720p sources via any input, while other resolutions were generally good compared to others in the group. A "screen door" pixel grid was visible from 7 feet as well as faintly from a normal seating distance of 12 feet. Edge enhancement was visible via 720p HDMI, and eliminating it resulted in too soft an image, but 720p component-video didn't exhibit any enhancement. Uniformity was the worst of the group, with subtle discoloration visible in white and gray fields at various light levels; black levels did not fluctuate depending on program content. Overscan measured 1% along the top of the screen and 0% elsewhere. Focus was problematic, with softness visible on the right side of the screen, and minor fringing was visible from 7 feet on convergence patterns. Standard-def video processing was good, although some jagged edges were evident, and 2:3 pulldown detection was nearly instantaneous.
See the review of the Hitachi HDPJ52
Color temperature (before/after calibration)
Low window (30 IRE): 6,211/6,596 K
High window (80 IRE): 6,216/6,495 K
Brightness (100-IRE window before/after calibration): 15.7/12.0
Before calibration the InFocus IN76's grayscale varied by an average of 271K from the color temperature standard — second-best in the group — and afterward it tied for second-best at 51K. Its real-world contrast ratio of 99.6:1 was sixth-best. Primary-color accuracy was below average, with the worst green in the group, and color decoding was also incorrect with green. One of the sharpest projectors in the group, the IN76 fully resolved 720p via both HDMI and component-video inputs and delivered generally superior sharpness for other resolutions and inputs. Edge enhancement was negligible on all sources except for 480i component-video. Uniformity was solid aside from a somewhat brighter area toward the bottom; black levels did not fluctuate depending on program content. Overscan measured 0%. Focus wasn't perfect, with very slight softness evident in the upper-right, but no fringing was visible. Standard-def video processing was excellent on all HQV tests.
See the review of the InFocus Play Big IN76.
Color temperature (before/after calibration)
Low window (30 IRE): 7,956/6,584 K
High window (80 IRE): 8,230/6,572 K
Brightness (100-IRE window before/after calibration): 16.8/11.8
The Mitsubishi HC3000U's grayscale before calibration varied by a group-worst average of 2,630K from the color-temperature standard, and though much improved afterward it was still second-worst at 199K. Its real-world contrast ratio was third best of the group at 112.9:1. Accuracy of the Mitsubishi's primaries was better than most on red and green but the worst on blue. Resolution was below average, not hitting every line of a 720p pattern via either HDMI or component-video, and standard-def was soft via component-video. Edge enhancement was negligible on all sources. Uniformity was a bit worse than for the other DLPs, with a pronounced center-bottom hot spot; black levels did not fluctuate depending on program content. Overscan was 0% on all sides. Focus was perfect, and no fringing was visible. Standard-def video processing was excellent on all HQV tests.
See the review of the Mitsubishi HC3000U.
Color temperature (before/after calibration)
Low window (30 IRE): 7,243/6,660 K
High window (80 IRE): 7,150/6,374 K
Brightness (100-IRE window before/after calibration): 7.6/12.1
Before calibration the Optoma HD72's grayscale varied by an average of 716K from the standard, which was fifth-best of the group. Afterward it varied by a sixth-best 176K. Its real-world contrast ratio of 132:1 was best in the group. The accuracy of red, green, and blue was about average for the group, with green again the worst, and color decoding was very good. Resolution with 720p sources was perfect via both component-video and HDMI, and other resolutions were very good. Pixel structure was slightly more visible than on other DLP models but still invisible from normal distances. Edge enhancement was negligible on all inputs. Uniformity was very good, and black levels did not fluctuate depending on program content. Overscan measured 1% along the top of the screen and 0% elsewhere. Focus was very good; some fringing was visible from 7 feet but not from further away. Standard-def video processing was excellent on most HQV patterns, although some jagged edges were visible in the most-difficult sections.
See the review of the Optoma HD72.
Color temperature (before/after calibration)
Low window (30 IRE): 6,472/6,572 K
High window (80 IRE): 6,222/6,540 K
Brightness (100-IRE window before/after calibration): 10/11.6
The Panasonic PT-AE900U's grayscale varied from the standard by an average of 194K before calibration — best in the group — and was third-best afterward at 53K. Its real-world contrast ratio of 90:1 was the worst. Primary colors were relatively accurate except for green, and color decoding was very good. Resolution via 720p HDMI was perfect, while component-video was somewhat soft (as it was at other resolutions). Pixel structure was actually less visible than on two of the DLPs — excellent for an LCD. Edge enhancement was negligible with 720p sources but visible on component-video with both 480p and 480i. Uniformity was worse than for the DLPs, with some discoloration in white and bright gray fields; black levels did not fluctuate depending on program content. Overscan was 0%. Focus was very good, but fringing was visible even from normal seating distances along the right side. Standard-def video processing was excellent on all HQV tests.
See the review of the Panasonic PT-AE900U.
Color temperature (before/after calibration)
Low window (30 IRE): 7,053/6,500 K
High window (80 IRE): 6,964/6,501 K
Brightness (100-IRE window before/after calibration): 12.7/12.1
Before calibration the Samsung SP-H710AE's grayscale varied by an average of 549K, third-best of the group, but it pulled into first afterward at 45K. Its real-world contrast ratio of 125:1 was second-best. Primary-color accuracy was also best in the group — excellent for red, green, and blue, all three — and color decoding was spot-on for both SD and HD sources. Resolution via 720p DVI was perfect, but some high-frequency interference was visible via component-video; other resolutions were excellent (the DVI port could not accept 480i, however). Edge enhancement was nonexistent with all sources. Uniformity was superb, and black levels did not fluctuate depending on program content. Overscan was 0%, focus was perfect, and no fringing was visible on convergence test patterns. Standard-def video processing was excellent on all HQV tests.
See the review of the Samsung SP-H710AE.
Color temperature (before/after calibration)
Low window (30 IRE): 7,747/6,452 K
High window (80 IRE): 7,832/6,745 K
Brightness (100-IRE window before/after calibration): 12/11.6
Before calibration the Sharp XV-Z3000's grayscale varied by an average of 1,194K from the color-temperature standard, which was second-worst in the group, and afterward it fell to last place with a 203K average variation. Its contrast ratio was fifth-best at 102.5:1. As with many of the other projectors, the red and blue primaries were fine but green was highly inaccurate, and color decoding for both standard and high-def sources was also inaccurate for green. The XV-Z3000 could not deliver every line of a 720p pattern via HDMI or component-video, and except for 480p via HDMI, other resolutions also looked relatively soft. Edge enhancement was negligible for all sources except 480i and 480p component-video. Uniformity was somewhat below-average for a DLP, with a hotspot along the bottom edge of the screen and a shadow in the lower-left (perhaps unique to my review sample). Black levels did not fluctuate depending on program content. Overscan measured 0%, focus was excellent, and no fringing was visible. Standard-def video processing was excellent on all HQV tests.
See the review of the Sharp XV-Z3000.
•Read the entire review.
•Watch the video.
•Discuss this article.
Behind The Numbers
Back to Homepage
What's New on S&V