The CableCARD Issue
With an antenna connected to Input A, I was able to receive all my local DTV channels with a very strong signal. Unfortunately, installing the CableCARD disables Input A (the cable-system cable goes there), and Input B cannot access the ATSC tuner, so if your cable company doesn't offer local digital channels, you lose them. In my area, Cablevision provides all local digital channels, but it doesn't include the secondary ones that NBC (4-2) and ABC (7-2) offer over the air. The NBC one is a particularly useful 24-hour local weather channel.

The CableCARD picture quality was slightly improved compared to the set-top box, but the loss of Cablevision's program guide, station ID and info, and all of the interactive on-demand channels made it a non-starter in my house. As far as I'm concerned, until there is interactivity and the actual cable system's program guide, CableCARD is useless. Losing over-the-air digital reception, as this set forces, hardly helps.

The Main Event
When a review is ready for publication, or nearly so, we send it to the manufacturer for a fact check and a manufacturer's comment if they wish to submit one. When Hitachi received the review of the 50VX915, they remarked that the sample was not typical of production and offered to send another. Since MF had moved on to other reviews, and I wanted to run the Hitachi through some of our standard test-bench measurements, I asked them to send the second set to me at our Los Angeles facilities. What follows is first a summary of MF's findings on his sample, followed by my analysis of the second one.—TJN

I have never evaluated a set as visibly askew out of the box as this one. As delivered, the Hitachi's picture was unacceptably green. After adjusting the user picture controls as best as I could, I had other issues too, but I chose to wait for ISF calibrator Kevin Miller to do his thing before making any assessment of the picture quality.

Unfortunately, even after calibration shadow scenes and dimly lit movies remained distractingly green-tinged. Your tolerance may be greater than mine. Brightly lit scenes such as live sports events and animated films had a much better balance, but even then, the green had an electric cast—a problem seen in many microdisplays.

In addition, even after sampling the Black Enhancement and Contrast Mode features (which Miller recommended against after determining their effect via measurements), the deepest blacks were mediocre at best, often with a noticeably bluish tinge to them. Dimly lit scenes were greenish-gray, drab, and lacking in detail. The overall picture appeared soft and overly smooth.

Even with the final calibrated contrast set to 90 and brightness set to 43 for HD and 53 for DVD, it doesn't take measurements to see that this set simply isn't bright enough, nor is the contrast ratio sufficient to produce a vibrant, exciting picture that pops off the screen. Compared to the bright (perhaps too bright), adrenaline rush of a picture produced by the JVC HD-52Z575 I recently reviewed, this Hitachi LCD set was strictly dullsville (though to be fair, the JVC had its own problems).

At a more-than-reasonable distance from the screen, the picture's pixel grid was still visible—or, if not visible as such, it could be seen as something getting between the viewer and the picture on the screen. This set is not alone in this fault, but it's more pronounced than on any of the DLP sets I've checked out (though, of course, this three-chipper doesn't suffer from single-chip DLP's rainbows).

Hitachi's 1080p video processing (I know, the chip is 720p, but that's what they call it) was actually quite good judging by, among other things, the latest Silicon Optix HQV Benchmark test DVD, which has some challenging visuals.

My mother-in-law is a big sports fan who comes over to watch in HD and never says anything about picture quality; when she remarked how disappointed she was by this set's picture, I knew something was wrong. Walking back into my living room where my 65-inch Hitachi CRT-based 65XWX80 RPTV resides made it all too clear that the 50VX915—and most microdisplays I've seen, actually—are losers by comparison. This LCD lost by more than most.

TJN Comments
When I received the second 50VX95, the first thing I did was to check all four color temperature settings with our Photo Research PR-650 colorimeter. None of them were particularly accurate, though Medium was the closest (see "Tests and Calibration"). I conducted my tests mainly with familiar, high-quality DVDs, including both test discs (Video Essentials and Digital Video Essentials) and good film transfers. I turned off all extraneous features. Black Enhance sounds appealing but while it appears to increase contrast, what it actually does is crush all detail out of the deep blacks. Even its lowest setting compromises the picture's accuracy. The Contrast Mode (not the same as the Contrast control) was left on Standard, and all specialized color-management and color-decoding controls were left at their factory defaults (standard and mid settings, respectively).

The Hitachi's pre-calibration picture looked respectable on normal program material, at least on bright scenes. Even before calibration, I measured no excess of green in the set's color temperature from the darkest grays to peak white. Despite this, however, flesh tones did show the subtlest trace of green. But it was not nearly as obvious or bothersome as it apparently was on MF's sample (which I did not see).

My post-calibration observations were essentially the same. The deepest blacks had a hint of blue, as MF also observed. My viewing notes for Charlotte Gray, which I rank among the dozen or so best-looking DVDs I've ever seen, state that the color is "okay." If I looked very carefully, at some lighting and angles, I could still spot the merest hint of green on flesh tones, but not enough to disturb me. I know that some viewers are more sensitive than I am to any green shift, however small, but I found the flesh tones and overall subjective color respectable, though still not very accurate on measurement (see "Tests and Calibration").

The images were nicely detailed—I was not bothered by softness in my sample that MF observed in his—and only rarely did I notice any video noise or artifacts. I also liked the overall brightness level. Some, like MF, will find the set too dim. Many competing models are two or three times as bright, but to me, they look unrealistic. At just over 30 foot-lamberts, the Hitachi will more closely resemble the (lower) brightness of film projection than the standard torch settings used by much of its competition.

But the Hitachi's most evident shortcomings were its shadow detail and black level. The grayed-out look on dark material so common with many LCD displays was clearly visible. Dark, low-contrast scenes simply look grungy on most such sets, including this one. It takes you out of a movie because you know instinctively, from years of viewing, that this is not the way filmed material should look.

The culprits are contrast ratio and the absolute black level a set can produce. Calibration will do nothing to improve this. Bright scenes looked good on the Hitachi, but as the scenes got dimmer, the image lost snap and dimensionality faster than most DLPs and plasmas, much less a CRT. It's not at all surprising that MF preferred his well-calibrated, 65-inch Hitachi rear-projection CRT; an LCD will seldom measure up in such a face-off, and it will come off best only (perhaps) in sharpness and brightness. (New developments, such as dynamic irises, may improve this in sets we'll see later this year or next.)

Conclusion
Michael Fremer: Hitachi is a company accustomed to having complete control over all aspects of the technology used in its televisions—not to mention in the bullet trains it builds in Japan. In the case of this TV, the company outsourced the LCD imaging chips. I don't now whether these chips, or Hitachi's implementation of them, is responsible for the mediocre picture I saw. Based on the other Hitachi products I've reviewed, and knowing the competency and talents of the people involved as I do, I'd bet it wasn't Hitachi's fault. The folks at Epson may have another perspective, however.

Either way, I can't work up enough enthusiasm for the 50VX915 to recommend it. Perhaps with the next generation of Epson LCD chips, future versions of this elegant-looking design will yet produce a great picture. In fact, according to Hitachi, the next generation of this display addresses all of the problems I have cited here, so stay tuned.

Thomas J. Norton: I had a little better luck with the second sample than MF had with his. Though its color was far from accurate, it was watchable and not obviously tinged with green. It was definitely dimmer than many competing sets, but I unlike MF, I saw this as a plus, resulting in a more film-like brightness than the excessive light output from many digital displays.

I suspect that many owners of this set (assuming their samples are more like mine than MF's) will watch it contentedly, but it really isn't a set for the hard to please videophile. Its blacks and shadow detail are poor, and while that description can be applied to many competing LCD displays, it isn't really a good excuse when there are DLPs, plasmas, and, yes, CRTs that do significantly better. The Hitachi also proved devilishly tricky to calibrate, and the results may not be worth the effort and expense.

Like all manufacturers, however, Hitachi is working to improve their sets. For a sneak peak at the new Hitachi line, scheduled for release late this summer, see the related news story.

Highs and Lows

Highs
Attractive cosmetics
Plasma-like form factor
Excellent remote

Lows
Compromised contrast ratio and color
Highly reflective screen
No ATSC over-the-air reception with CableCARD in place