SIM2 Grand Cinema C3X Three-chip DLP Projector Tests and Calibration
All the measurements and calibrations were made on a 16:9 Studiotek 130 screen (78-inches wide, 1.3-gain)
Gray Scale and Color The Before curve in the accompanying chart shows the results obtained out of the box with the color temperature set to 17—the best setting I found. The After results (at 59 hours), also at 17, include the presence of the ½f-stop neutral density filter in front of the lens to reduce light output on my smaller screen. While the two results look quite similar, shifts in the lamp's response during its break-in period made it necessary to enter the calibration service menu to obtain the results shown in the After curve. Post calibration, the x/y coordinates remained within -0.001, +0.003 of the D6500 reference point at light levels from 20 IRE to 80 IRE.
The red color point was a little oversaturated, though no more so than most digital projectors. Green was also a bit too intense, though closer to correct than we usually see. Blue was very close to the correct ATSC color coordinates.
A Software Issue?
I did run into one notable problem with the C3X. The projector allows a calibrator to fine-tune the color temperature using a code-protected service menu. You can enter different settings for each input, which I did for both component and HDMI. But when you switch inputs, the projector will often search for a few seconds for the correct color temperature setting (during which you can see the color change) before settling down. This alone was a bit annoying, but not a problem. But sometimes the color temperature the projector finally locked onto was not the correct one. This actually happened a number of times with my test pattern generator, but fortunately only once with an HDMI input from a DVD player.
Resolution (source: AccuPel HDG-3000 signal generator)
The multiburst results below are for luminance. Except as noted, the color performance was slightly less extended on most of the settings, but that is typical and in no way caused problems in the image that could be blamed on limited color bandwidth. The control settings mentioned here are those determined from the Sharpness pattern on the AccuPel generator.
At 720p, the sharpness pattern is very close to ideal. A Sharpness control setting of 6 produced the best result. The multiburst patterns were virtually perfect to the maximum 37.1MHz burst.
At 1080i, the Sharpness pattern was not quite as crisp as 720p, but typical of what I've seen from other top-rank projectors. A Sharpness setting of 5 looked best. The 18.5MHz burst was crisp though slightly uneven, but the response at 37.1MHz, while visible, was noticeably reduced in level.
At 480i the C3X worked fine with the HDMI output of a DVD player, but would not produce an image usable for measurement purposes with a 480i input from the AccuPel pattern generator.
At 480p a Sharpness setting of 3 worked best, but no setting would completely eliminate slight white enhancement lines around the pattern's black horizontal and vertical lines. The multiburst luminance pattern looked fine up to the highest (13.5MHz) frequency, though it was slightly down in level at that frequency.
At 720p, the Sharpness pattern is very close to ideal. A Sharpness control setting of 6 produced the best result. The multiburst patterns were virtually perfect to the maximum 37.1MHz burst.
At 1080i, the Sharpness pattern was not quite as crisp as with 720p but was typical of a good projector. A Sharpness setting of 3 looked best. The 18.5MHz burst was clear though slightly uneven, and the response was visible but noticeably reduced in level at 37.1MHz.
At 480i the projector produced an Out of Range warning but locked on to the image after a few seconds. There was no visible response from a 6.75MHz burst pattern (the maximum for 480i component), and close to the screen there was also visible noise in the image that was not present at other resolutions (this was essentially invisible from my normal 11' viewing distance.) A sharpness setting of 1 produced the cleanest result on the sharpness test pattern. There were some minor but visible enhancement lines surrounding the black lines in the pattern even at this minimum setting.
480p a setting of 4 worked best, but no setting would completely eliminate the same slight enhancement lines around pattern's black horizontal and vertical lines. The multiburst luminance pattern held up well to the highest (13.5MHz) burst.
The C3X's HD provided superb response at 720p and somewhat compromised resolution at 1080i. The latter is common to most of the 720p projectors we've reviewed. The lack of any response with the 6.75MHz luminance and color bursts, along with some visible noise, with 480i component was, however, unexpected in a projector at this price.
With the ½ f-stop neutral density (ND) filter in place, the projector calibrated for a color temperature of D6500, the lamp set to its minimum 200W setting, the brightness and contrast controls set to the points providing the deepest blacks (as determined by a PLUGE pattern) and whitest whites while (still allowing some headroom above 100 IRE), and the Gamma set to 2.4, I measured a peak contrast ratio of 1618:1 (14.56 foot-Lamberts peak white/0.009 video black.) Without the ND filter, the white and black levels increased (19.35fL and 0.012fL respectively), but the peak contrast ratio remained nearly the same, at 1613:1.
With the ND filter removed and the Gamma control increased to a User setting of 2.8 (see "Gamma," below), the other conditions as above, the peak contrast measured 1538:1 (18.45 peak white, 0.012 video black.) Increasing the lamp setting to 250W increased both the peak white and minimum black levels (to 22.09fL and 0.015fL, respectively), for a peak contrast ratio of 1473:1.
The C3X displayed both below black and above (100 IRE) white with a component input, and above white on HDMI. It would not, however, display below black via HDMI.
When I measured the on-screen gamma at a Gamma control setting of User 2.4, it actually measured between 1.8 and 1.9 at all levels except 20 IRE, where it measured 2.2. What this means is that the image was lighter than it should be based on a desired gamma of 2.2-2.4 (the CRT gamma at which most program material is created, and which therefore will produce the most accurate results on playback.) This clearly contributed to the projector's brighter appearance when compared side-by-side with the Yamaha DPX-1300, which in its gamma "B" setting ranged from 2.2-2.5. Increasing the setting on the Gamma control on the C3X to User 2.8 increased the measured gamma to an average of about 2.2 (ranging from 2.15 to 2.4.) This setting, combined with the punch the projector's peak light output provided on my modestly-sized screen contributed hugely to my generally very positive final impression of the SIM2's image—an enthusiasm tempered only by its rather disappointing deep blacks and peak contrast.
With a 480i component input, the projector sailed through all the tests on my Faroudja test DVD. It also performed reasonably well on the Benchmark HQV test DVD. It sailed through that disc's jaggies tests, though it did exhibit a few wavy lines on the cadence tests and failed the difficult racetrack bleacher test. I never encountered visible deinterlacing or scaling artifacts on regular program material, with either HD or DVDs, in the over 100 hours I put on the projector.
The 480i component overscan was 2% on the left and right and under 1% on the top and bottom. With 480p it was 2.5% left and right and 0% on the top and bottom. In 720p and 1080i, component, and all HDMI resolutions, it was 0.5% or less on all sides. All readings were taken with the Overscan control set to "0."