JVC Everio GZ-HD40 Camcorder Review
Video Performance* (10.0)*
The GZ-HD40 contains two firsts for JVC. It's the company's first CMOS chip in a consumer camcorder, and its first to use AVCHD compression. With that in mind, it was very interesting to test the GZ-HD40 against previous JVC Everio HD models, and AVCHD camcorder from other manufacturers.
The GZ-HD40 features a single 1/3-inch CMOS chip, a big step away from JVC's normal outfit of three CCDs for its top-tier camcorders. It's a development we're quite pleased to see, frankly. Sony and Canon have been using CMOS chips about this size in the last few generations of their HD camcorders. They've performing extremely well, usually better than the JVCs with their three small CCDs.
JVC GZ-HD40 3000 lux (MPEG-2 TS)
We tested the camcorder in both MPEG-2 TS and AVCHD in our controlled lab environments and out the lab to see how it performed. Let's start with the MPEG-2 TS in the FHD quality setting, which is purported to offer the highest performance. At 3000 lux under more or less ideal conditions, the GZ-HD40 performed quite well. It's funny that even though the sensor was completely overhauled, we still could have identified this test chart as the product of a JVC camcorder in a double-blind test. Much of the color processing must have remained the same, because it has the habit of really saturating the greens and reds.
Overall, this looks to be an improvement over the GZ-HD6, the previous-generation HD Everio. The color performance is largely the same, except for a clear and inexplicable difference in the yellows, which are now less saturated and more natural looking. The GZ-HD40 looks sharper than the GZ-HD6 - cleaner lines and better edges. In extreme zooms, we could see that the HD40 is capturing more fine detail, but the increase is incremental. What you can see quite clearly is that the GZHD40 does not produce the same blue discoloration along areas of high contrast that we saw with the HD6.
|JVC GZ-HD40||JVC GZ-HD6|
|Canon HF10||Sony HDR-SR12|
Above, 100% crops from the test charts
So what about our top performers this year, the Canon HF10 / Canon HF100 and Sony HDR-SR12 (the finding of which also apply to the Sony HDR-SR11)? At a quick glance, the Canon HF10/HF100 appears to have a more natural looking color pallet. This is what we've always liked about Canons - great color. Canon also has the upper hand on high contrast detail (like type), although the performance gap is much smaller than it was with the JVC GZ-HD6.
Comparatively, the Sony HDR-SR11 / HDR-SR12's colors are much less saturated, which will be to the liking of some, but not others. This is more of a personal preference. The Sony's colors tend to test very accurately, but most people prefer more saturation. Sharpness to the eye was almost identical with the GZ-HD40. We were interested to note, however, that Sony managed to produce a more even color within each color chip, while the GZ-HD40 showed more compression artifacting - minor blotchiness where color tone should be flat.
Out of the lab, the JVC GZ-HD40 (shooting in MPEG-2 TS) looked very good. On a sunny day, you're going to love the camcorder's propensity for saturation and warming the colors. The blue in a blue sky, the green in the grass, the red in the brick - they all pop with the kind of intensity that you want.
The auto white balance adjustments do not come very fast. This can be a a little frustrating when you point into a shady area and all the color is suddenly sucked into a bluish haze. Give it a few seconds and the colors will naturalize again, but the GZ-HD40 does seem slower than others.
JVC GZ-HD40 3000 lux (AVCHD)
Now let's move on to the AVCHD footage from the JVC GZ-HD40. When looking at the color chart, there is virtually no difference in color performance between AVCHD and MPEG-2 TS footage, which is understandable. The camcorder is still using the same processor and same sensor. The difference is in what the processor is choosing to discard. Most of that comes out in the fine detail and color gradation. Perhaps our shots outside of the lab illustrate the point more effectively.
|*JVC GZ-HD40 MPEG-2 TS in FHD quality (click for full res)*|| |
|*JVC GZ-HD40 |
AVCHD in EP quality
Shooting outside, we could clearly identify the difference in quality between MPEG-2 and AVCHD. The heavier compression really starts to show in color gradations. Take a look at these three shots of a motorcycle. Note how bad the compression artifacting gets as we move from MPEG-2 TS (FHD quality), then AVCHD (XP quality), then AVCHD (EP quality). While JVC is not entirely wrong to tout the camcorder's ability to record full 1920 x 1080 even in the lowest AVCHD quality, it clearly has a debilitating effect in the creating smooth tones.
|*Note the increasing color compression in the tonal gradations as the quality is lowered*|
|MPEG-2 TS in FHD quality|
|AVCHD in XP quality|
|AVCHD in EP quality|
Overall, the video quality of the JVC GZ-HD40 is excellent when shooting in adequate light (low light performance is discussed below). The colors are more saturated than its competition, the Canon HF10 / Canon HF100 and Sony HDR-SR11 / Sony HDR-SR12, but most consumers will appreciate the boost. The ability to upconvert video to 1080p for playback is a small but useful feature to squeeze a little extra juice out of the camcorder, though it doesn't actually add any resolution to the image.
Video Resolution* (20.25)*
The video resolution of the JVC GZ-HD40 was tested by shooting an DSC Labs video resolution chart at an even, bright light. The recorded footage was then played back on an HD monitor. As we stated in the JVC GZ-D6 review, switching between 1080i and 1080p output had no impact on this type of resolution test. In real world shooting, you will see a small increase in how diagonal lines are displayed if you switch the camcorder to 1080p output, of course providing that you have a TV that supports 1080p.
Ultimately, we found that the JVC GZ-HD40 produced an approximate horizontal resolution of 675 line widths per picture height (lw/ph) and a vertical resolution of 600 lw/ph. These scores compared very well with the competition. This was better than the previous generation JVC GZ-HD6, as well as the Sony HDR-SR12. The scores are approximately identical to the Canon HF10, making the GZ-HD40 one of the highest resolution camcorders we've seen so far.
Low Light Performance* (5.1)*
The low light performance of the JVC GZ-HD40 was tested in three stages (and sometimes more). First, we shot a DSC Labs Chroma DuMonde chart at an even 60 lux and 15 lux, then compared then with similar test charts from our ridiculously enormous bank of camcorder test results. (Note: if you were ever interested in tracking the qualitative progress of the video industry for the past decade, our offices would be a good first stop).
JVC GZ-HD40 60 lux - AGC on
At 60 lux, shooting in MPEG-2 TS, the JVC GZ-HD40 retained a decent amount of light. The increase in noise was noticeable, but not overwhelming. It should be stated that these shots were taken with the auto gain set to AGC, which allows the auto gain to be on. There are two other settings: OFF and AUTO. The OFF setting turns the auto gain off completely, which renders the picture too dark. Just do yourself a favor and never turn it off.
JVC GZ-HD40 60 lux - AGC Auto (slow shutter mode)
The second setting, AUTO, allows the shutter speed to drop lower than it normally does, probably to 1/30th instead of 1/60th. It wouldn't hurt to leave it in this setting if you think your shots are coming out too dark. We do not run our test with this setting on, but you can see the results above. In short, we didn't see a difference at 60 lux. While shooting outside of the lab, in all the dark corners of our offices, we concluded that it seems to have an impact only in very low lighting.
How did the competition do under the same circumstances? To the eye, it was also a big improvement over the previous generation JVC GZ-HD6. The colors, in particular, looked more natural. Overall, the GZ-HD40 appeared brighter than the Canon HF10 and Sony HDR-SR12. This is a very good looking performance. The only complaint is that the noise (or perhaps it's compression artifacting) appears as a discoloration - a sort of yellowish or bluish fuzz. On the Sony and Canon, the noise tended to be black or grey, which can be a little less distracting. However, we're pleased that the GZ-HD40 is able to keep noise to a minimum. It's a huge improvement over how the noise was rendered in the older GZ-HD6.
If you choose to record in AVCHD rather than MPEG-2 TS, the results won't look too different. There was some slight color compression artifacting, but overall you could safely shoot in either mode.
JVC GZ-HD40 15 lux - AGC on
At 15 lux, the JVC GZ-HD40 kicks up a lot more noise, which is expected. There's also a big loss in fine detail retention, due primarily to the noise. However, the camcorder still does an admirable job with color retention. Granted, the color performance is a far cry from what it can do in adequate light, but testing showed that the GZ-HD40 can pull color from virtual darkness. It was much better than the Sony HDR-SR12, and a touch better than the Canon HF10.
The second stage of testing measures color accuracy, noise, and saturation in low light. We shoot an X-Rite Color Checker chart at an even 60 lux, then export frame grabs to Imatest imaging software. According to Imatest, the JVC GZ-HD40 produced a color error of 9.1. statistically identical to the previous generation JVC GZ-HD6. My, isn't that interesting? It seems that even with the switch from CCD to CMOS, and even with the big increase in surface area (three 1/6-inch CCDs to a 1/3-inch CMOS), the color accuracy is exactly the same.
The noise measured approximately 0.915%, a decrease since the GZ-HD6. This is a most welcome sign of improvements. The GZ-HD40 scored better than the Sony HDR-SR12 and Canon HF10. Finally, the saturation measured 96.1%.
The third test measures sensitivity in low light. We shoot the DSC Labs Chroma DuMonde color chart under a steadily decreasing light while the camcorder is connected to a waveform monitor. The point at which the camcorder can produce a maximum exposure of 50 IRE gives us the score for this section of the test. We found that the JVC GZ-HD40 required 14 lux of light in order to produce 50 IRE. This is a huge improvement over the older GZ-HD6, which needed 23 lux of light to perform the same output. The Sony HDR-SR12 produced the same score. The Canon HF10 was still better, needing only 10 lux.
Overall, the newer, bigger CMOS sensor on the GZ-HD40 is a major step forward for JVC's low light performance. The images stay brighter and cleaner in lower light than they ever did on previous HD Everio models.
JVC opted to regress to the same digital image stabilization (DIS) system found on the GZ-HD3, rather than the superior optical image stabilization system (OIS) from the previous generation. Unfortunately, the results are nearly identical to the older DIS system. DIS operates by sacrificing pixels to compensate for shake. OIS functions by isolating the lens from the body of the camcorder. The GZ-HD6 flaunted a OIS system that yielded the highest Stabilization score and best performance to date. Why JVC opted to downshift systems is baffling after the success of the GZ-HD6's OIS.
We tested the effectiveness of the GZ-HD40's DIS using our custom-built shake emulator at two speeds: Speed One, and Speed Two. Speed One is akin to typical handheld camcorder shake while Speed Two deals with more of a rocky terrain, like a light jog or jittery car ride. The GZ-HD40's DIS exhibited a 50% shake reduction at Speed One and a 66.67% shake reduction at Speed Two. This poor performance is almost a carbon copy of the GZ-HD3's stabilization performance. OIS is the way to go, and JVC needs bucket of cold water dumped on its head to snap out of this preposterous DIS funk.
Wide Angle* (8.8)
*We tested the GZ-HD40's maximum wide angle capability using a vertical laser. Test video was recorded with the Zoom pulled back to its widest setting and DIS disabled. The video was later interpreted on an external monitor in order to obtain an accurate wide angle reading. The GZ-HD40's maximum wide angle measurement was 44 degrees, which is a 6 degree reduction from the GZ-HD3 and GZ-HD6.
Get Our Newsletter
Real advice from real experts. Sign up for our newsletter
Thanks for signing up!