Runco Reflection CL-710 DLP Projector Calibration

Calibration

I had a different sample of the Runco Reflection CL-710 from the one Joel Brinkley used in the review. JB auditioned his unit with a Stewart Studiotek 130 screen. I used a 16:9, 80-inch-wide Stewart FireHawk; all of the measurements and comments in this sidebar refer to the CL-710 as used with that screen.

As delivered, my sample's color temperature was most accurate at the Temp 1 factory setting; the Temp 2 setting ranged between 7900K (low end) and 8600K (high end), Temp 3 from 8300K (low end) to 8700K (high end). (All numbers rounded off to the nearest 100 kelvins.) As shown in the accompanying figure ("Before" curve), Temp 1 showed a slight upward drift in color temperature as the average picture level increased, although, apart from a bit too much red in flesh tones, it looked quite respectable.

Using our Photo Research PR 650 spectro-radiometer, the CL-710 calibrated beautifully. The color temperature fell very close to the optimum D6500 across the full brightness range, increasing only slightly at the brightest levels, where the eye is least sensitive to small shifts in the white balance. I measured a little red push—about 10% according to the Avia test DVD—but it wasn't visibly obvious.

Using test patterns from the Video Essentials DVD, the CL-710's horizontal resolution showed a solid 500 lines per picture height (title 17, chapter 13). With the projector's Sharpness and Filter controls set for the sharpest image short of ringing on the overscan pattern (17-8: Sharpness 0, Filter +3), the luminance sweep (17-23) showed a clear emphasis at 4.2MHz, and several more peaks in the 4.8–5.5MHz region. Reducing the Sharpness to –1 and the Filter to +1 eliminated all of these visible test-pattern artifacts.

I did find three ergonomic annoyances. First, the inputs were accessible only through a small door at the back of the projector and an opening at the bottom—a very cramped arrangement. Second, the two component-video inputs are mutually exclusive; that is, the standard component input (RCA jacks) accepts only 480i sources, and the HDTV component input (BNC connectors, RGB or component) operates only with 480p or higher-resolution material. This almost forces users with high-definition sources to employ a progressive-scan DVD player in conjunction with a high-bandwidth component switcher, such as those found in high-end A/V receivers, if they want to avoid running two sets of component leads to the projector.

And third, the setup menu containing the color-temperature adjustments (accessible only by a special code) timed out quickly; I had to re-enter it dozens of times during the calibration procedure. But I did find out, after I had done this, that the timeout can be increased in the factory menu, one of those deep, dark, secret places accessible only to high priests of the temple of Karnac and Runco service technicians. Recommendation for the next product update: delete the timeout for the onscreen setup menu and provide a way to mute it while you make measurements.

The CL-710's deinterlacing and scaling are respectable though not flawless. I rarely saw any artifacts on normal program material, but the projector was slow in establishing a solid 3:2 pull down lock, and it didn't do particularly well with the mixed-content test on the Faroudja test DVD. The Runco did well on video over film, but it had problems presenting a clear image with film over video. But even though the CL-710's performance on difficult program material—such as the waving flag and highway stripes on Video Essentials—was not up to that of Faroudja- or DVDO-class processors, it was better than many other scalers we've seen. A first-rate progressive-scan player or other outboard processor might be beneficial with the most difficult material, but the Runco's onboard processor is good enough that I'd live with it for a while before deciding whether or not I needed an outboard scaler or good progressive-scan DVD player.

Unlike some other projectors, the Runco provides full control over the Brightness and Contrast settings (though not Sharpness) when using the DVI input, which functioned very well with a 720p input from a DVI-equipped DVD player we tried: the Bravo D-1 from V, Inc. (see review elsewhere in this issue).

High-definition test patterns from a Leader LT 446 test-pattern generator showed an overscan of 5% left, 4% right, 3% top, and 4% bottom. The 720p resolution pattern looked exceptionally crisp, with a resolution of 800 lines per picture height, the maximum shown on the pattern. The 1080i pattern was also sharp, although, as you might expect because of the required downscaling to 720p, the measured resolution was no better than that from the 720p pattern.

Peak on/off contrast measured 735 (14.7 footlamberts peak white, 0.02fL black). The best peak-contrast results we've obtained with other projectors using Texas Instruments' HD2 chip have been around 2000, but the Runco's results were comparable to those of both the InFocus ScreenPlay 7200 (719) and the Yamaha DPX-1000 (1031). In addition, the Runco result may have been compromised somewhat by the need to use the 0 IRE pattern on Video Essentials for the "off" reading. With other projectors, I've used the black field created by switching to an open input for this reading, but any unused input on the Runco automatically reverts to either a blue or white screen, eliminating that option.

The Runco's average ANSI (16-square checkerboard) contrast measured 134 (144 peak). For the first time, I used a technique SGHT contributing editor Pete Putman uses for this reading: covering the entire screen with black cloth (I used a black grille material; PP uses black velvet) and taking the readings on that. This minimizes the effect of room reflections on the result.

For the primary color points, the Runco's reds were good, the greens measured a slight yellow-green shift (as JB observed), and the blue edged toward greenish-blue, which was not noticeable. None of these deviations were greater than those we've tested on other DLP projectors, and the subjective color quality, after calibration, was very good.

Comment: On dimly lit scenes, the Runco's black level and shadow detail reflected its measured contrast: satisfactory, but not as good as the best I've seen. Low-contrast scenes looked a bit grayer than on a few other DLP projectors we've tested. On the other hand, bright scenes popped nicely on my FireHawk screen, with richly textured, 3-dimensional images. Animation (Monsters, Inc.; Finding Nemo) looked great, but so did well-mastered, colorful live-action films such as the Superbit version of The Fifth Element.

The CL-710 produced exceptionally crisp images from both DVD and hi-def sources. While the projector could look great with bright, high-quality program material, those dim scenes—or less than the best transfers—required careful balancing of all the available sharpness controls (Sharpness and Filter on the Runco, plus High and Mid Sharpness on our Integra DVD player) to avoid overly edgy or subtly grainy images. If there had been more steps on the Sharpness and Filter controls, each with finer resolution, this process would have been simpler and more precise.

Finally, the Runco CL-710 was about average in its tendency to show those rainbow artifacts that come with the territory of single-chip DLP projectors. But such rainbows come and go rapidly; had I added up the time they were visible on the Runco during a two-hour movie, the total probably wouldn't have come to more than 15 seconds. Fortunately for those marketing single-chip DLP projectors, many viewers don't see these flashes of random color at all. If you don't see rainbows (aka color fringing) when you move your eyes rapidly across a dark scene with bright highlights—e.g., a quiet street at night lit only by bright streetlights—you're probably home free. Me, I'm cursed; I've seen them to some degree on every DLP projector I've evaluated, though on one or two, they were infrequent enough to be irrelevant.—Thomas J. Norton

Share | |

X
Enter your Sound & Vision username.
Enter the password that accompanies your username.
Loading
setting var node_statistics_93812