Editors' Note: As we mentioned above, all of the hands-on observations below were made on the original KDS-60A2000 we reviewed, not on the KDS-60A2020. We have also modified the section regarding the set's ability to resolve every detail of 1080i test patterns; see the original KDS-60A2000 review if you're interested. According to Sony, the new models can resolve every detail of 1080i and 1080p sources and do not suffer from the misadjusted filter.
The Sony KDS-60A2020 delivered excellent picture quality, with deep black levels, plenty of detail in shadows, exemplary grayscale tracking, and excellent overall detail. Its color reproduction isn't as accurate as we'd like, and it has some issues with standard-def processing, but overall, it's still one of the best 1080p HDTVs we've tested.
After noting that the Custom picture preset provides the best home-theater image quality, we tweaked some settings to adjust the Sony KDS-60A2020 for critical viewing in our darkened lab. This involved reducing its peak brightness to a suitable level (about 37 footlamberts) and setting the advanced user controls as described in Features. We also set the Iris control to Minimum, electing not to go with either of the Auto settings because their increased light output had a tendency to raise black levels somewhat in scenes with both light and dark content--although some viewers may appreciate the extra punch of Auto. And while the default Warm2 color-temperature preset hewed relatively close to the standard, it still appeared a bit blue, so we used the user menu white-balance settings to achieve a remarkably accurate grayscale. Although professional calibrators may still want to go into the service menu to get the grayscale even closer, we were more than satisfied by the results we achieved with the user menu controls. Visit our page onuser menu settings for complete information.
The disc we chose to use for this evaluation was the dark, stylized, and razor-sharp The Chronicles of Riddick HD-DVD played through the Toshiba HD-A1. The Sony immediately impressed us with its depth of black; as the Necromongers' ships wheel toward the planet, for example, the blackness of space behind the star field matched the black of the letterbox bars, and both appeared positively inky next to that of the Panasonic PT-61DLX76, which we had on hand to compare. We were not able to directly compare the Sony's blacks to those of the Samsung HL-S5687W, but black-level and contrast-ratio measurements confirmed that the two were very similar in this regard. The Sony also did an excellent job of maintaining shadow detail in scenes with lots of bright material as well as mostly dark scenes, and colors remained consistent in shadows and bright light, a testament to the Sony's excellent grayscale tracking.
In terms of detail the KDS-60A2020 resolved every line of a 1080i test pattern; it was as sharp as we expect an HDTV to be. When we compared the Sony directly to the Panasonic, details looked equally sharp and lifelike, from the chainmail links in the neck of the main Necromonger to the cityscape before it blacks out, to the thin strands of curly hair on the edge of Kyra's (Alexa Davalos) profile.
The Sony KDS-60A2020 did a solid job with color, delivering neutral skin tones throughout Riddick as well as in more colorful material we watched. However, primary colors on the SXRD were inaccurate, falling outside the norm for the HD spec (see the Geek box), so they might appear a bit garish to some eyes. We turned down the color control somewhat to help compensate but were still able to maintain good saturation.
The Sony had no problem accepting 1080p/60 sources delivered by the Samsung BD-P1000 Blu-ray player, although 1080p/24 material from our Sencore VP-403 signal generator was not accepted. We also looked at resolution patterns via component-video; they were solid, although, as with any 1080p HDTV, we recommend setting your sources to 1080 output when possible. The KDS-60A2020 exhibited excellent white- and gray-field uniformity for a three-chip display, varying very little in color across the screen, although its middle was brighter in relation to the corners of the screen when compared to the single-chip Panasonic DLP. Focus was superb, and very little fringing was visible on white lines, pixel structure was invisible from further than arm's length from the screen, and the Sony suffered from less false contouring or banding than any set we've recently reviewed.
In terms of standard-def performance, the Sony KDS-60A2020 offered more than its share of surprises. A lot has been made of the fact that the step-up KDS-XBR2 line includes a different version of the company's DRC processing, and while we can't speak to whether it's better overall than the DRC used by the KDS-60A2020, we can say that the latter could use some improvements. First off, it failed to correctly detect and implement 2:3 pull-down detection according to tests from the HQV disc as well as Star Trek: Insurrection, introducing concentric lines of moiré and unnatural moving lines on the upturned boats, respectively. [08-14-06 Update: After this review posted Sony mentioned that we might experience improved performance by turning off DRC and turning on CineMotion, a setting found in the setup (not the picture) menu. In these settings the Sony still failed to pass the HQV test although moiré was reduced slightly, and on Insurrection it still exhibited sporadic moving lines via composite and S-video, while component-video was fine. Overall A2020 still delivered disappointing 2:3 performance compared to almost all late-model HDTVs.]
Second, it didn't do a very good job of removing jagged edges from moving video, such as the stripes in a waving American flag. The results of these tests were the same, regardless of which DRC preset was used or how we adjusted the custom "reality vs. clarity" DRC matrix--don't ask.
On the other hand, the Sony's detail with standard-def sources was excellent, and its ability to reduce video noise was among the best we've tested. We looked at standard-def resolution patterns, and the Sony aced them in DRC Off mode, while High Density looked a little softer and Progressive (the other two DRC modes) looked unstable. Turning from test patterns to real video, the shot of the bridge from HQV was commendably sharp in all three modes via component-video, although via S-Video it softened up considerably in Off. The shots of noisy sky from HQV, which wreak havoc on many HDTVs, looked much smoother in the KDS-60A2020's highest noise-reduction setting with no apparent loss of detail, and even the Auto mode--available in composite and S-Video only--did a good job of engaging the correct level of noise reduction.
Xbox 360 performance: We also tried playing a couple of video games on the Sony with our Xbox 360-- and --and the picture looked excellent. We didn't notice any delay from controller to screen whether or not the set's Game mode was engaged. With composite-video, turning Game mode on did have the effect of softening the picture considerably, obscuring details in the ridges of one of the guns in Prey, for example, but that's to be expected, since engaging Game Mode skips video processing. With component-video 1080i, we noticed no difference in picture quality regardless of Game mode setting.
We also tried connecting the 360 via the VGA adapter, and the console did not fill the Sony's screen, regardless of which adjustments we made on the console of the TV itself. Component-video looked much sharper and also filled the screen. (Thanks to David Rudden for 360 testing).
PC performance: We hooked the Sony up to a PC via VGA, and as expected, the set maxed out at 1,366x768 resolution. The desktop did not nearly fill the screen, and furthermore, according to DisplayMate, the Sony couldn't resolve every line of resolution; text indeed looked pretty soft, even after we engaged the set's automatic calibration for PC.
When driven by an HDMI input connected to a DVI adapter, however, the Sony's PC picture quality improved significantly. First off, the set had no trouble accepting a full 1080p source, and it did resolve every line, although the image was overscanned significantly; the edges of the screen, including the entire Windows toolbar for example, were invisible. We couldn't adjust it properly in the Sony's menu system, but our GeForce 7900GTX video card's HDTV Overscan Compensation came in really handy. When we selected the Underscan option, it fit the screen perfectly; displaying at 1,824x1,016 pixels. Of course that's not full 1,920x1,080, and we'd prefer not to have to make these kinds of tweaks--a "dot by dot" aspect ratio would be a big help--but it's safe to say that computer display via HDMI really outclassed the Sony's VGA input.
KDS-60A2020 picture-quality comparisons
Vs. existing KDS-R60XBR1: We were not able to test the XBR1 and the A2020 side-by-side, but we really do not expect the latter's ability to accept 1080p signals to provide a major picture quality improvement. Most other aspects of performance will be identical, including black levels and color reproduction, although white- and gray-field uniformity seem to be improved on the A2020. Overall we recommend the XBR1 over the A2020, as long as it's available, because of its better standard-def performance.
Vs. Samsung HL-S5687W: Again, we did not have this Samsung DLP on hand to directly compare, but in previous testing, it delivered similar black levels, better primary color accuracy, and superior standard-def performance. On the other hand, its characteristic DLP rainbows may be visible to some viewers, and its tendency to accentuate reds is something the A2020 doesn't share. The Sony scored higher in performance, but the Samsung's 61-inch equivalent is probably a better value to viewers who aren't bothered by rainbows.
|Before color temp (20/80)||6,821/6,776K||Good|
|After color temp||6,529/6,547K||Good|
|Before grayscale variation||+/- 303K||Good|
|After grayscale variation||+/- 59K||Good|
|Color of red (x/y)||0.680/0.320||Poor|
|Color of green||0.287/0.698||Poor|
|Color of blue||0.144/0.046||Average|
|Black-level retention||All patterns stable||Good|
|2:3 pull-down, 24fps||No||Poor|
|Defeatable edge enhancement||Yes||Good|