Canon Vixia HF11 Camcorder Review
The Canon Vixia HF11 is, in many ways, a minor update to an already highly regarded camcorder. The previous generation HF10 produced some of the best looking video of the year, and definitely the sharpest. The new HF11 promises to outdo even that by increasing the bitrate, making it the first consumer camcorder to record at the AVCHD spec\'s ceiling of 24Mbps. In truth, the result is a very minor increase in video quality, and most consumers will probably not see a difference. The more compelling value-add may be the increased flash memory capacity, doubling from 16GB to 32GB. Certainly not failing to live up to its pedigree, the Canon HF11 makes a compelling camcorder.
Video Performance* (10.75)*
For nearly all purposes, the Canon HF11 is identical to the Canon HF10 and the Canon HF100. It has the same 1/3.2-inch CMOS sensor with a gross pixel count of the 3,310,000 (and an effective pixel count of 2,070,000). By comparison, the Sony HDR-CX12 has approximately the same size sensor with a much higher pixel count of 5.66MP. As we saw in the results between the Canon HF10 and the Sony HDR-SR12 (which uses the same sensor as the CX12), the Sony's higher pixel density seemed to hurt its low light performance.
Canon HF11 - 3000 lux in Auto mode
First, let's look at our lab testing. We shot a DSC Labs Chroma DuMonde chart at an even 3000 lux. Under this light, Canon once again proves that it can dominate the performance category. The picture is just amazingly crisp and vivid. It's better than any AVCHD camcorder to date, and is more or less indiscernible from the tape-based Canon HV20 and Canon HV30.
Comparatively, the Sony HDR-SR12 produced a similar color performance, but the sharpness of the fine details was not as good as the Canon HF11. The SR12 just looked fuzzier. Because the Sony HDR-CX12 is using the same imaging system as the SR12, we expect it to perform the same. The Panasonic HDC-HS100 produced very bright, bold colors. Unfortunately, the resolution can't match that of either the Canon HF11 or the Sony HDR-SR12.
JVC GZ-HD40 looked very good indeed at this light, with strong even colors. However, it too lacked the resolution in the fine details. The Samsung SC-HMX20 actually topped the Canon HF10 in straight-up resolution testing, but the Canon image looked better overall, which we believe is due to a superior compression.
It's important to note that the performance from all of these camcorders was quite good. For the casual user, the Canon, Sony, Samsung, or JVC would produce extremely good results. The Panasonic has the great color, but is a little deficient in sharpness compared to the rest of the group.
Canon HF11 - 3000 lux in Cine mode
We also shot in the Cine Mode, which is an alternate gamma curve that is meant to emulate the colors produced in film. The end result is a picture that expands the range of the mid-tones, while decreasing detail in highlights and shadows. It can make your video look great, as long as your shot is properly lit.
Canon HF11 - 3000 lux in 30P mode
Canon HF11 - 3000 lux in 24P mode
The alternate frame rates, 30P and 24P, didn't look too different under this more or less perfect light. In both frame rates, there was a slight decrease in noise, but color and sharpness looked the same. As with the HF10, the HF11 video in alternate frame rates looked very good. For whatever reason, Canon is the only manufacturer that has managed to produce a smooth, decent looking progressive frame rate. The Samsung SC-HMX20 was the closest, but moving objects looked a lot more stuttery. The Panasonic high definition camcorders cannot record in 24P without also recording in xvYCC color mode, which blows out the saturation.
Out of the lab, we put the HF11 and HF10 side-by-side in some difficult shooting situations and examined the playback footage with an extremely critical eye. After much deliberation, we came to the conclusion that most people can't tell the difference between the two. You'd have to spend an insane amount of time and energy—as we did—to see a variance.
The difference is in the compression artifacting, as far as we could tell. The resolution isn't any better, but if you look very closely, you'll notice that the edges of moving objects looks cleaner. In any video or still image compression, every object in a shot creates an unpleasant 'aura' of blocky or noisy compression artifacts. What we saw as the primary difference between these two camcorders was that the HF11's aura didn't spread out as far as the HF10. That's because the HF11's faster bitrate allows the compression engine to adjust more quickly to changes.
Here are some examples above of shots that ask a lot of the compression engine. Click to open the full resolution bitmap frame grab. Look closely at the edges of the object. You should see a tiny improvement in the HF11. All shots were recorded at 1080/60i.
Overall, the Canon HF11 continues to offer some of the best video quality in all of consumer camcorders.
Video Resolution* (21.94)*
The video resolution of the Canon HF11 was tested by shooting a DSC Labs video resolution chart. The recorded footage is then played back on an HD monitor. Resolution is measured in line widths per picture height (lw/ph). Ultimately, the HF11 (shooting in 1080/60i) produced an approximate horizontal resolution of 675 lw/ph. The vertical resolution measured approximately 600 lw/ph.
Shooting in 1080/30P mode, the camcorder produced a horizontal resolution of 650 lw/ph and a vertical resolution of 650 lw/ph.
In 1080/24P, the HF11 produced a horizontal resolution of 650 lw/ph and a vertical resolution of 650 lw/ph.
These scores were identical to the Canon HF10, and very, very good against the competition. Only the Samsung SC-HMX20 produced a higher resolution.
Low Light Performance* (14.13)*
In order to make the low light performance evaluation as comprehensive as possible, we test in three stages. First, we shoot the trusty DSC Labs Chroma DuMonde color chart at an even 60 lux and 15 lux.
Canon HF11 - 60 lux in Auto mode
Under the 60 lux conditions, recording in 1080/60i, the Canon HF11 produced an image with strong colors. Strangely, the image appeared noisier than the previous generation HF10 did under the same conditions. As we'll see in a moment, these variances disappeared in 30P and 24P modes.
As we stated in the HF10 review, the larger CMOS chip on the Canon HR10, Canon HV20, and Canon HV30 produced a brighter image with better color quality. Comparatively, in 60i the Sony HDR-SR12 was also brighter, though the Sony does not offer alternate frame rates. The JVC GZ-HD40 held up very well under the same test, retaining a lot of color. We did, however, notice that the noise produced by the JVC was of a yellowish tint that was more distracting than the greyish noise from the Canon and Sony.
The Panasonic HDC-HS100 appeared as brightly exposed but with very dull colors under this 60 lux light. It looked as if the auto gain has been pushed way up, and the noised masked or supressed. The Samsung SC-HMX20 was the big surprise when we tested it, producing an amazing color and light retention that rivaled any of the camcorders in this group.
Canon HF11 - 60 lux in 30P mode
However, Canon keeps an ace up its sleeve with the ability to record in both 30P and 24P. Each of these settings slow down the refresh rate of the sensor, letting in more light. At 30P, the HF11's brightness picked up a little, and so did the saturation.
Canon HF11 - 60 lux in 24P mode
When the HF11 was put into 24P mode, the low light performance was (naturally) the best of all. The noise went down considerably from the 60i shooting, and the saturation picked up quite strongly. In this mode, under these conditions, you would have no problem capturing all the light and fine detail you needed. The only potential issue is the inevitable trailing that occurs with 24P shooting. Of all the camcorders on the market recording in some kind of 24P mode—Canon, Panasonic, and Samsung—the Canon camcorders are the only ones that seem to pull it off without heavy trailing or overly-stuttering motion.
Canon HF11 - 15 lux in Auto mode
When we lowered the lights to 15 lux, the Canon HF11 produced a ton of noise. Yes, the color performance was good, but the overall image was hard to look at. Fortunately, the camcorder manages to produce a high enough resolution that no matter what the light, it can salvage a lot of fine detail.
Canon HF11 - 15 lux in 30P mode
At 30P in 15 lux, the Canon HF11 reduced a lot of that noise.
Canon HF11 - 15 lux in 24P mode
At 24P in 15 lux, the HF11 started to look salvageable. Sadly, the camcorder had a little trouble white balancing under these conditions, and appeared somewhat reddish.
The second stage of the testing examines color accuracy, noise, and saturation. We shoot an X-Rite Color Checker chart at an even 60 lux, then export the stills to Imatest imaging software. According to Imatest, the Canon HF11 (in 1080/60i mode) produced an approximate color error of 12.2, which is not a great score. This is worse than the JVC GZ-HD40, Sony HDR-SR12, and Samsung SC-HMX20. It's statistically identical to the Canon HF10. The noise measured 1.9225%, which confirms what our eyes saw in a separate test under the same conditions—the Canon HF11 is noisier than the HF10, at least at 60i. Finally, the saturation measured 1.9225.
In 1080/30P, Canon HF11 produced a color error of 10.2, according to Imatest. The noise dropped to 1.3175%, and the saturation increased to 89.97%.
In 1080/24P, the color error measured 9.7%. The noise dropped further, to 1.225%, and the saturation increased to 94.74%.
The third and final test examines sensitivity. We lower the light in a slow and steady manner while the camcorder is connected to a waveform monitor. We see how much light is required for the camcorder to produce 50 IRE (a measurement of exposure). In 1080/60, the Canon HF11 required 10 lux.
In 1080/30P, the sensitivity increased and the HF11 needed only 5 lux of light to produce the same exposure.
In 1080/24P, sensitivity increased further to 4 lux.
Overall, the low light performance on the Canon HF11 is still very, very good. But in order to take full advantage, you'll want to use the alternate frame rates: 30P and 24P. If you only plan on shooting in 60i, there are camcorders that produce a brighter image and better color (but not the sharpness, unfortunately). Check out the Sony HDR-SR12 and Samsung SC-HMX20.
*Disclaimer – We received some unexpected test results during our evaluation of the Canon HF11, particularly a higher noise percentage in low light. We contacted Canon, and they agreed with our assessment that the results were counter to what should have been expected. In order to deliver to you, the reader, the most up-to-the-minute information, we have published our original results here. Canon plans to send us a second review model so that we can have the opportunity to retest for the sake of accuracy. The review will be edited to reflect any new test results, with a note explaining what was changed and why.
Like the HF10, the HF11 is equipped with Optical Image Stabilization (OIS), the leading form of shake reduction in the camcorder circuit. OIS operates by separating the lens element from the body of the camcorder, so as not to sacrifice pixels around the border of the frame. Electronic Image Stabilization (EIS) is less effective, operating by creating a digital buffer of pixels around the sensor, sacrificing resolution.
We tested the efficiency of the HF11’s OIS using our custom-built shake emulator at two speeds—Speed One and Speed Two. Speed One emulates typical stationary handheld shake, while Speed Two’s shake pattern is more intense, producing a frequency that is more akin to a light jog with the camcorder in hand. At Speed One, the HF11 displayed a 75-percent shake reduction, while Speed Two showed an unimpressive 50-percent reduction. This is an identical performance compared to the HF10. JVC's GZ-HD6 still holds the title for having the best OIS so far this year.
Wide Angle* (9.6)*
We tested the HF11’s maximum wide angle capability using a vertical laser. During testing, OIS was disengaged and the zoom was pulled back to the full wide setting. Video was then later interpreted on an external monitor to obtain an accurate reading. The HF11 displayed a maximum wide angle measurement of 48 degrees, which is identical to the HF10.