Sony Handycam HDR-UX20 Camcorder Review
The Sony HDR-UX20 ($999 MSRP) is the company’s top high definition DVD camcorder for 2008, compressing video in the AVCHD format. While the UX20 has benefited from a number of upgrades since last year, Sony has clearly split the DVD and hard disk drive (HDD) camcorders into two market segments – the point-and-shooters and the enthusiasts.
Video Performance* (9.5)*
The Sony HDR-UX20 is equipped with a single 1/5-inch ClearVID CMOS sensor. This is a major downgrade in size from last year’s HDR-UX7, which featured a 1/2.9-inch CMOS. Now, Sony didn’t just shrink it. The company goes to great pains to point out that the UX12 is equipped with the latest in Sony engineering, the Exmor sensor and Bionz processor, both of which are supposed to reduce noise. The gross pixel count on the HDR-UX12 is 2,360,000 (effective pixel count is 1,490,000). This is down from 3,200,000 gross pixels on last year’s UX7.
Our experience with the Exmor/Bionz experience up until this review was the Sony HDR-SR12. That camcorder lost size on its CMOS since last year, and its overall performance was outstanding. The case was quite different, however. Though the gross pixel count – and therefore pixel density – increased from 3,180,000 to 5,660,000, the size reduction was less severe: 1/2.9-inch to 1/3.15-inches.
Sony HDR-UX20 at 3000 lux
First, we shot a DSC Labs Chroma DuMonde color chart at an even 3000 lux. The HDR-UX20 produced a clean picture with an even palette under these conditions. We’ve seen a few camcorders this year, including the Panasonic HDC-HS9, with slightly smaller chips (3 x 1/6-inch CCDs), and even the same size chips like the JVC GZ-HD6 (3 x 1/5-inch CCDs). The Sony UX20 looks sharper and cleaner than both of them. All in all, we were pleasantly surprised with what the UX20 could produce with a "limited" chip size.
There is a clear difference in quality between the HDR-UX20 and the sibling HDR-SR12. The SR12, with its larger CMOS chip and higher pixel count, captured deeper blacks and more fine detail. The Canon HF10 continues to produce the sharpest images and best looking color performance in this particular test.
Outside of the lab, away from the ideal lighting of the lab, the results were a bit more mixed. Outside, under clouds and sun, we saw the same great color performance and auto responses to white balance and exposure that were praised in the HDR-SR12. There was little obvious artifacting. The HDR-SR12 and the Canon HF10 showed more fine detail, particularly the Canon, which remains the sharpest picture overall. However, this time out we saw some strong purple fringing in the Canon HF10's outdoor pictures. These only occurred in areas of extremely high contrast. and could be due to chromatic abberation or overspill on the CMOS chip. We also noticed the the Sonys did a better job bringing out detail in shadows when the image had mixed shadows and highlights.
Indoors, the gap between the UX20 and the Sony SR12 / Canon HF10 widened considerably. The small CMOS in the UX20 seems be a serious detriment in even moderate lighting. The difference in resolution was easy to spot, as was the loss of detail in shadows.
As you'll see in the low light section below, the UX20 failed to live up to real world requirements, at least in comparison to competing models. When the light is adequate, we were impressed with what Sony was able to accomplish with a single CMOS chip this small.
Video Resolution* (18.75)*
The resolution of the Sony HDR-UX20 was tested by shooting a DSC Labs video resolution chart at an even, bright light. The recorded footage is then played back on an HD monitor to determine line width per picture height (lw/ph). Because the camcorder records two different max bitrates – 14Mbps for DVD and 16Mbps for flash memory – we decided to test both.
Recording to the DVD, we pegged the horizontal resolution at approximately 600lw/ph and vertical resolution of 625 lw/ph.
When we shot the same thing to the internal flash memory, we found the same approximate resolution. There was no difference in resolution, at least as far as this test could show.
Low Light Performance* (1.91)*
The low light testing is done in three primary stages. First, we shoot the trusty DSC Labs Chroma DuMonde chart at an even 60 lux and 15 lux, then compare the results with similar shots from other camcorders.
At 60 lux, the HDR-UX20 lost a lot of color information. When held up next to the same shot from the HDR-SR12, with its larger CMOS chip, the results make a pretty compelling case to skip the UX20 and step-up to the SR12 (or HDR-SR11). Sony seems to be making good on its claim of noise reduction with the new Exmor and Bionz technologies, but there’s only so much you can do with a 1/5-inch chip.
Sony HDR-UX20 at 60 lux (above)
Sony HDR-SR12 at 60 lux
Every camcorder we compared it to showed more ability at producing color under these conditions, though it was not always a better picture overall. The Panasonic HDC-HS9, for example, had more saturated colors but the overall image looked fuzzier. It would be a matter of preference for a shooter to decide which one they liked better. The Canon HF10 (shooting in 1080/60i mode), produced much better color and more fine detail.
Sony HDR-UX20 at 15 lux
At 15 lux, the Sony HDR-UX20’s performance is just abysmal. Awful. The camcorder had trouble focusing and there is virtually no color information. Camcorders often tank during this test, but rarely have we seen a $1000 model do so bad. The Panasonic HDC-HS9 was a little better in this test. The Sony HDR-SR12 and Canon HF10 were significantly better.
The second test looks at the camcorder’s sensitivity. We shoot the same chart with an ever-decreasing amount of light while watching a waveform monitor. We look at the lowest amount of light required for a camcorder to produce a peak of 50 IRE (a measurement of exposure). The Sony HDR-UX20 was able to produce 50 IRE at a light level of 23 lux. This was far less sensitive than the sibling HDR-SR12, which could produce the same information with only 14 lux, almost half of what the UX20 needed. It's no great surprise, given the smaller CMOS chip. This was, however, the same score as the Panasonic HDC-HS9. The Canon HF10 trumped them all, producing the same results with only 10 lux (shooting in the same 1080/60i as all these camcorders). The results for the UX20 here are disappointing.
Finally, the third test considers color accuracy, noise, and saturation at low light. We shoot the X-Rite Color Checker chart at an even 60 lux, then export frame grabs to Imatest imaging software for analysis. The Sony HDR-UX20 was able to produce a color error of 19.3. This is an abysmal score for an HD camcorder. Everything we've tested this year has performed far better. The HDR-SR12 produced an 11.4, Canon HF10 produced an 11.9, Panasonic HDC-HS9 produced an 11.3
The noise from the Sony HDR-UX20, on the other hand, was very impressive, producing only 0.78% noise at 60 lux. This is a tribute to Sony's new Bionz and Exmor technologies, though the company clearly overestimated its sensitivity in low light. It did better than the Canon HF10 and, oddly, the HDR-SR12. The Panasonic HDC-HS9 did even better with noise, producing only 0.595%. Finally, the HDR-UX20 produced the low saturation score of 45.76% in this light level.
Overall, the HDR-UX20's low light performance is very, very disappointing. With no manual control over shutter speed, the user is left with little recourse for dealing with low light environments except auto mode settings like Auto Slow Shutter and Color Slow Shutter, which can range from ineffective to drastic.
The HDR-UX20 is equipped with Super SteadyShot OIS, a stabilization system that functions by separating and stabilizing the lens element from the body of the camcorder to minimize shake. OIS is the most effective shake reductions system available on the consumer level. Electronic Image Stabilization (EIS) is less effective because it creates a digital buffer around the recorded frame, sacrificing resolution.
We tested the HDR-UX20 using our custom-built shake emulator at two speeds: Speed One and Speed Two. Speed One emulates typical stationary handheld shake while Speed Two simulates a light jog or bumpy car ride with the camcorder. The HDR-UX20 exhibited a 66.7% shake reduction at Speed One and a 60% shake reduction at Speed Two. This is a fair showing, however, the GZ-HD6 and Panasonic HDC-SD9 proved to be stronger performers.
Wide Angle* (9.6)*
We tested the HDR-UX20’s maximum wide angle using a vertical laser at both left and right angles. The HDR-UX20 was set to manual mode with the zoom pulled back to its full wide angle setting. Video was later interpreted on an external monitor in order to attain a true reading. The HDR-UX20’s maximum wide angle measurement is 48 degrees.
Get Our Newsletter
Real advice from real experts. Sign up for our newsletter
Thanks for signing up!