Reference Imaging CinePro 9x Elite & Teranex HDX Cinema MX Calibration
Light Output: CRTs are generally acknowledged to produce the best pictures, overall, of all video projection technologies. But in one respect, at least, CRT suffers: it can't produce a lot of light.
Or so conventional wisdom goes. When Chris Stephens was finishing setting up the CinePro 9x Elite and told me the light output he was getting, I couldn't believe it. But there it was, as verified by two test instruments, the Photo Research PR-650 Spectroradiometer and the Minolta LS-100 spot light meter: more than 19 foot-Lamberts! And with a great picture. Granted, this was on my relatively small, 78-inch-wide screen; the CinePro is hardly a good choice for 30-foot D-Cinema presentations. But it was the brightest CRT-projected image I'd ever seen.
Scaling back the light output to something more traditional—say, 10–12ft-L—markedly reduced the image's punch. As a big believer in video dynamic range, Stephens was reluctant to make this sacrifice. It's his contention that, once the light output drops too low, the eye no longer perceives edges in a realistic way, sharply reducing the image's 3-dimensionality. And image depth, or dimensionality, is a key strength of this projection system.
But on some material, the high light output was a little fatiguing. So during the review period I alternated between peak outputs of 15 and 19ft-L, and never felt, even at the higher setting, that I was getting anything less than the best images I had ever seen from a video projector. They were so good, in fact, that I never was tempted to try a lower light output.
Until, that is, the last two weeks of the review. The CinePro's gray-scale tracking, as measured by the Photo Research and as seen on 10-step gray-scale test patterns, was the projector's most obvious shortcoming (see below). I never found the deviations to subjectively degrade normal program material, but nevertheless, two weeks before the projector was scheduled to depart my studio, I tried reducing the contrast to a maximum of 12.5ft-L to see if I could improve the situation.
The reduction didn't make much of a change in the gray scale, but it did noticeably improve the resolution. At the higher outputs, some blooming was clearly occurring even though I had not found it subjectively annoying. And, surprisingly, the enhanced sharpness did not appear to compromise the video dynamic range. The wow factor was still very much there. Why had it looked compromised at the lower output when the projector was first installed? I can speculate only that the change had something to do with the projector settling in during the additional hours of operation.
Wondering how much wear and tear those high peak light levels put on the tubes? Judging from the image quality and measurements, very little. At the time the projector was returned, it had over 600 hours of operation (about 400 of which I had accumulated) and still produced 19.4ft-L at the settings Stephens had used. The kelvins reading at peak white (a key indicator of wear, particularly on the blue tube) was also virtually unchanged. This is a testament to the design of the CinePro, and to the quality of its tubes. By no means would I extend this argument to all CRT projectors.
Contrast: We would love to be able to measure the contrast on CRTs, but the fact is that they produce blacks so dark that it would take test equipment far more sophisticated and sensitive than ours to get a reading. Even at its remarkable peak white output level of 19.2ft-L, the CinePro 9x Elite's contrast, if we could measure it, would likely be in the multi-thousands.
Test Patterns & Gray Scale: All conventional test patterns looked superb on the CinePro, including HD resolution limits very close to the limits of the 1920x1080 format. The single exception was the 10-step gray scale. This was consistent with the measurements (shown in the accompanying diagram for the 1080p24 setup). The actual color points of the red, green, and blue tubes were very close to the standard HD points for green and blue, but red showed a slight orange shift—not very different, in fact, from the coordinates found on the uncorrected red tube of a Dwin HD-700 CRT projector.—TJN