Calibration: We began as always by setting up the Mitsubishi in a completely dark room and adjusting its picture controls accordingly. The WD-65831 can get extremely bright, but we found that turning down contrast did improve black levels and make the picture easier on the eyes in the dark. We ended up with a maximum light output of around 35 footlamberts. Prior to calibration, the Mitsubishi's grayscale came fairly close to the standard in the Low preset (see the geek box below), but was somewhat greenish. Afterward, grayscale tracking was very good, although we were disappointed that the set, unlike some others, didn't allow us to tweak the grayscale in the user menu.
We also spent a good deal of time adjusting the set's extensive color controls. Out of the box, the Mitsubishi's color decoding, measured against the standard blue reference, exhibited desaturated green and oversaturated red. The Perfect Color control allowed us to adjust the color balance and improve it tremendously. Perfect Tint, on the other hand, didn't work as well as we'd have liked. It promised the ability to adjust the colors with respect to others, making red more magenta or more blue, for example, which should fix the Mitsubishi's questionable primary and secondary color accuracy. Playing with these controls did improve colors somewhat, but after adjustment they were still pretty far from perfect. Part of the problem was that the colors actually changed slightly when we entered the menus, and that the control display obscured a good portion of the screen. But the simple lack of adequate control range was the main issue; no matter how much we shifted the tint control for blue, for example, we couldn't approach the HDTV standard on the Y axis; blue remained too greenish.
While all of those the controls are great, we'd rather the set have accurate primary colors out of the box. Compared to the Sony KDS-R60XBR2, the Mitsubishi's primary colors exhibited about the same degree of inaccuracy--albeit in different colors--while sets such as the Samsung HL-S5679W and HL-S5687W, for example, were much more accurate. When all of our adjustments were complete, the Mitsubishi's color was definitely better, but not exceptional compared to the competition. See the Tips & Tricks section above or click here for our complete user-menu settings.
Blu-ray testing: After setup, we chose to watch the excellent-looking Aeon Flux Blu-ray disc played from a PlayStation 3 via HDMI at 1080p and 1080i resolutions. We compared the Mitsubishi side by side with a variety of displays, including the directly competitive Sony KDS-R60XBR2 as well as the Panasonic TH-50PH9UK, the Panasonic TH-50PF9UK and the Pioneer Pro-FHD1, all 50-inch plasmas.
Black-level performance between the Sony and the Mitsubishi was just about equal. The WD-65831 delivered nice, deep blacks in darker scenes such as when Aeon infiltrates the theater. The shadows along the walls and curtains appeared quite clean, and I could discern lots of detail in her face as she peeks around the corner. There was no sign of visible gradations or contours in the transitions from shadow to darkness. We also appreciated that the color stayed true and neutral into the darkest areas, where on the Sony it tended to get tinged somewhat bluer. Note that the Mitsubishi actually changed black-level profiles when we switched between 1080i and 1080p sources, becoming somewhat lighter with 1080p. If you're setting up the display at home, we recommend choosing between 1080i and 1080p and sticking with that choice for each input, as opposed to switching between the two.
We've discussed color a lot already, so we'll be brief here. After the adjustments mentioned above, the Mitsibishi WD-65831's colors appeared fairly good, but there were still issues. During the gunfight at the end, for example, we noticed that the sky above the organic blimp appeared too greenish, and not as deep blue as on the Sony SXRD or the plasmas. Aeon's skin looked very good, on the other hand, without any reddish tinge, although it was slightly less saturated than on the plasmas--increasing the color control too much did turn her skin redder, although at moderate levels it was fine. The yellow in the tree leaves was a bit too mustard-colored and brownish, although the green of the grass looked much more lush and accurate than on the Sony. After our adjustments, color was a cut above many high-end HDTVs, but still not as good as the aforementioned Samsungs, for example.
Screen uniformity on the WD-65831 was a notch above many other rear-projection sets. The Sony exhibited slight discoloration on white and gray fields, namely a more purplish tinge toward the top and bottom edge, where the Mitsubishi's fields were very consistent in terms of color. Its brighter, middle hot spot was more noticeable than the Sony's, however. We also noticed more stationary screen grain on the WD-65831, which appeared as tiny spots that became visible primarily in white and other bright areas, especially during pans. The Sony had significantly less of that type of grain. Our Mitsubishi review sample exhibited excellent geometry with very little distortion or fringing along straight white lines, even along the extreme corners and edges of the screen.
One of the chief uniformity complaints we've had about DLP televisions is the rainbow effect. Maybe it's a result of the company's six-segment color wheel, but we barely saw any rainbows whatsoever on the WD-65831. They cropped up only during very high-contrast scenes and only when we moved our eyes across the picture--something that happens rarely if you sit an acceptable distance from the screen. Of course, Sony and JVC LCoS displays don't engender rainbows at all, but this is the first DLP we've reviewed (aside from the LED-powered Samsung HL-S5679W) where rainbows weren't an issue.
As a 1080p display, the Mitsubishi WD-65831 should be able to deliver every detail of 1080 material. It did, according to the pattern on our Sencore VP403 signal generator, with both 1080i and 1080p sources. The set accepted 1080p at both 60 and 24 frames per second. Although we've heard (from HD Guru Gary Merson) that the Mitsubishi can properly detect and implement 2:3 pull-down detection with 1080i material, we can't test that capability ourselves since we don't yet have access to the proper test material. We did notice that it exhibited moire and moving lines on the stairs at the beginning of Chapter 8 on the Mission: Impossible Blu-ray discs (artifacts that disappeared when we switched to 1080p output), but we saw the same effects on all of the TVs in our test facility. If you're keeping track, only the JVC HD-56FN97 has passed this test.
In the highly detailed sections of Aeon Flux, such as Chapter 9 when she confronts the Keeper, the high resolution of the WD-65831 came out beautifully. All of the fine vertical lines were perfectly visible, and when Aeon ascends the stairs, they remain perfectly clear and artifact-free. The Mitsubishi didn't exhibit the same level of razor-sharpness that we saw from the Pioneer Pro-FHD1, but it was very sharp--about on the level of the Panasonic TH-50PF9UK. It was also a hair sharper than the Sony next to it, delivering very slightly sharper details in wrinkles on Aeon's lips, for example, or the stubble on Trevor Goodchild's face.
Standard-def testing: Tearing ourselves from high-def discs, we checked out the WD-65831's standard-def performance using the HQV test DVD. The Mitsubishi turned in a more consistent performance than the Sony, for sure, acing the 2:3 pull-down test, although it took a long half-second to remove the moire from the bleachers. The biggest issue was some instability we saw on the resolution pattern, although the WD-65831 did resolve every line of the 480i source. It did a mediocre job of smoothing jagged diagonal lines, and there was some softness to details such as the stone bridge and the grass in the stationary city shot. On the other hand, the WD-65831 exhibited the most effective noise reduction we've seen from any HDTV we've tested recently. The three levels of noise reduction, Low, Medium, and High, offer progressively better removal of video noise with trade-offs in detail, which is par for the noise-reduction course. High mode especially virtually eliminated noise on the disc's difficult, snowy-looking cloud and sky shots.
PC testing: We tested the Mitsubishi's PC capability by connecting its DVI input to a home theater PC, and we appreciated the set's ability to handle a 1,920x1,080 source. The WD-65831 also resolved every line of the source according to DisplayMate, and text looked perfectly legible. As we've seen on other rear-projection sets connected to PCs, the WD-65831 overscanned the image significantly, so, for example, the taskbar along the bottom of the Windows desktop along with the far-left column of shortcuts were invisible. You may be able to correct this issue using your video card's drivers, at the expense of scaling the image. We appreciated the option to adjust vertical and horizontal position.
We also tried connecting the Samsung BD-P1000 Blu-ray player to the DVI input using an HDMI-to-DVI adapter and didn't have any problems playing a disc at 1080p. Other resolutions were problematic, however, including 1080i, which appeared soft, didn't fill the screen, and caused an error message. Connecting the Toshiba HD-A1's HDMI output to the Mitsubishi's DVI input at 1080i resolution didn't work at all; all we saw was a red screen. In other words, the DVI input seems to work fine as a third HDMI input (as opposed to a PC input) as long as the source is 1080p. Be aware, however, that it offers fewer picture controls than the two HDMI inputs, so we recommend using them first.
|Before color temp (20/80)||6,300/6,277K||Good|
|After color temp||6,523/6,427K||Good|
|Before grayscale variation||+/- 219K||Good|
|After grayscale variation||+/- 131K||Average|
|Color of red (x/y)||0.663/0.327||Poor|
|Color of green||0.289/0.631||Average|
|Color of blue||0.143/0.087||Poor|
|Black-level retention||All patterns stable||Good|
|2:3 pull-down, 24fps||Y||Good|
|Defeatable edge enhancement||Y||Good|