I've been a Sony HVR-Z7U owner for a little more than a year, and, to date, it is my favorite video camera. I love its low-light sensitivity, its true progressive frame rates, its manual lens, and, especially, its professional-looking design. Unfortunately, when it was first released, it wasn't a perfect solution for my varying video needs, so I hung on to my first HD camera, the Canon XH A1, for situations where I needed a longer lens and multiple video outputs.
All of my current and previous Sony video cameras were limited to 12x optical lenses, and seeing as how I cut my teeth with wedding videos, I never used to have the need for a more powerful telephoto lens. In fact, the opposite was true, and I was constantly struggling to delicately thread and unthread a wide-angle conversion lens on the front of my old DSR-PD150s and 170s, ever fearful of the dreaded cross-thread. This all changed when I made the lateral move to the event and corporate markets and found myself at the back of large halls, convention centers, and theaters, operating my video camera at full telephoto, wishing I was either closer or had some more telephoto range. This is how my Canon XH A1, with its 20x fixed lens, earned its keep in my camera roster.
Unfortunately, Sony and Canon video cameras don't interpret color in the same way, so after multiple-camera shoots, I found myself constantly color matching in post. By Christmas 2008, I'd decided that something had to change and that I wanted a matching set of video cameras. Someone at Canon must have been listening to my past reservations with the XH A1, which I chose as an honorable mention in my Best of 2007: Editors' and Columnists' Picks and discussed in both my EventDV articles on the HVR-Z7U and HVR-S270. The new XH A1S improved on what I felt was a touchy zoom rocker on the original XH A1; it now allows for independent control of the audio input levels so that you can connect both a line- and a mic-level input at the same time. Unfortunately, the low-light sensitivity has not improved, and the LCD is smaller and has a smaller pixel count than the Sony Z5/Z7 models (2.8"/207,000 versus 3.2"/921,000).
I was reminded of this difference most recently when comparing the Sony Z7, Z5, Canon XH A1s, and Panasonic AG-HMC150 at my local camera dealer, ProVision Video. I chose these models because they all have 1/3" sensors, and I tested their performance at 1440x1080/60i for the HDV Sony and Canon models and 1920x1080/60i for the Panasonic.
The Sony models were clear winners in my comparison of LCD screens, which I tested by attempting to manually focus the individual cameras. All three manufacturers included handy focus-assist features, which I used, and I found the Canon LCD acceptable, even though it has less than a quarter of the resolution of the Sony models. The Panasonic's LCD was unacceptable and at 210,000 pixels over a 3.5", 4:3 LCD, it has the same specs as the LCD that was used on the old standard-definition AG-DVX100 and is not suited for use in a high-definition video environment.
I'm still not sure which camera I would have chosen for my second camera if I had been limited to a second Z7 or a matching XH A1S. It would have come down to a trade-off between the low-light sensitivity and convenience of the CF recorder on the Z7 and the 20x lens and the ability to have more than one simultaneous video output on the XH A1 models. Fortunately for me and my beloved Z7, Sony introduced the Z5U, basically a Z7 with a 20x fixed lens, and I sold my Canon.
The Z5 (Figure 1, below) is very much like the Z7. They share the same set of three CMOS Exmor sensors, class-leading LCD screens, and progressive frame rate options. I love the images that both cameras produce and the control I have over the audio and the video. Although the street price at the time of writing of this article for the Z7U was $4,999.95 and for the Z5U was $4,079, the "apples to apples" difference is only $36 when you factor in the $884.95 HVR-MRC1, CF card recorder that is standard on the Z7 and an option on the Z5. Although I could go on discussing all the common features these two cameras share, including what I love about both of them, I'd rather go deeper into what makes them different and what I'd like to see changed for the next model. I hope you are listening, Sony.
A video camera's controls are what camera operators use to interact with the video camera. As a camera operator, I expect controls and I/O ports to be located in intuitive and predictable locations; this is especially important when I'm switching between different cameras and/or working in the dark. I got used to the way my Z7 was laid out, and even though Sony did a good job with the placement of the majority of the controls, I question why Sony had to make so many layout changes between the Z7 and the Z5.
For starters, the tape door on the Z7 is on the back, right-hand side (RHS) and opens vertically. I love this change; the layout is different from all the other cameras I have ever used. In order to avoid getting the backs of people's heads in my shots with other cameras, I often have to hold my camera slightly higher than eye level. With a back vertical tape door, I can change the tape in a camera that is more than a foot higher than the top of my head. Unfortunately, the Z5's tapes are loaded from the top, and the door is on the left-hand side (LHS), which causes me to fumble when loading tapes while holding the camera above my eye line. With the tape mechanism moved to the LHS, the Z5's HDMI port is inconsequentially located on the RHS, and the FireWire port that was so nicely upgraded to the larger 6-pin port on the Z7 is now once again a 4-pin on the Z5.
Although there is seemingly lots of room under the I/O ports for the LANC controller jack, Sony has moved it to the RHS just above the record button, which now also doubles as the camera's power switch, while the Z7's power switch is on the LHS. The LANC jack extends the distance between the jack and the controller. But the curled LANC cable on my Manfrotto 521 Pro controller was able to bridge the extra distance, even when I switched my tripod pan handle from my usual position on the RHS to my switchfoot position on the LHS.
Sony did add a useful cable holder that makes cable management easier, and it offers additional protection to the cable jacks if you accidentally pull on the cable. I've had to replace two headphone jacks over the years because I stepped on the headphone cable, so I consider the cable holder to be cheap insurance.
My biggest complaint with the Z7 (the annoying connection priority that limits the number of video connections you can use to a single output) is unchanged. The Canon XH A1 allows two simultaneous connections: a BNC connection and your choice of Component or RCA. The XH A1s seemingly mocks Sony's stingy output by allowing an astonishing four simultaneous connections: two Component, an RCA, and a BNC. In the webcast market, this will prove invaluable as it is common to use a video distribution amplifier in order to split a video camera's output to a high-bitrate encoder, a low-bitrate encoder, an IMAG feed (live in-house projection), and an output for the camera operator's portable field monitor.
Sony did make a small upgrade to its DIN connectors by allowing both the Component or RCA cables to connect to the same jack and by including left/right RCA audio connections on both the Component and RCA cables. Unfortunately, this creates confusion in my Pelican case because my Z7 and Z5 have component cables with different DIN connections. Ultimately, Sony still lacks a BNC option, multiple simultaneous outputs, and locking and/or nonproprietary video connections.
Moving over to the LHS, the ND filter wheel is now a vertical switch, and the Z5's servo-controlled, 20x G Series fixed lens no longer has the option of switching between servo and manual, and the manual/auto focus control is a switch and not controlled on the lens barrel like on the Z7.
But the real story is what is missing from the Z5. The L1/Digital extender button is gone, along with the "shot transition/focus marking" trio of buttons. The loss of four user-assignable buttons means the user has to delve into the menu more frequently, which Sony made more difficult to access than before. All the Z7 buttons, including the menu button, protrude from the camera's body ever so slightly; the Z5's are recessed, which means you have to squish your fingers into the recess in order to gain sufficient contact to push the buttons. This requires significantly more pressure, which isn't good when you are trying to keep the camera steady and is actually tough on your fingertips; they push against the straight-edge lip of the button recess. It doesn't get any better once you get into the menu. The scroll button is now smaller, smoother, and recessed, and it follows the curve of the camera, again requiring more pressure and additional turns to travel the same distance.
Things get a bit better in the audio department. The Z5 comes with the same shotgun mic as does the Z7, but it also has a built-in stereo microphone that offers increased flexibility in a variety of shooting situations. The audio controls, which are located behind a clear, plastic, bottom-hinged cover, allow me to switch between the XLR inputs and the internal microphone without accessing the menu.
The cover even has thoughtfully placed round holes that provide access to the manual audio-level dials with the cover still in place, which is nice since the cover no longer has the Z7's protruding lip-the user has to literally pry open the magnet-secured door with a fingernail.
Canon lenses are world-class. The company was founded in the 1930s to manufacture cameras, while Sony started in the 1940s to manufacture transistor radios. Canon first started developing its own lenses 70 years ago, after initially sourcing its first lenses from the predecessor of Nikon. Sony, on the other hand, relied mainly on lens manufacturers such as Carl Zeiss and Fujinon for lenses on its broadcast cameras. And when Sony introduced its first DSLR in 2005, the lenses were produced through a partnership with struggling Minolta. In 2006, Sony purchased Minolta's camera assets and, with them, decades of lens experience. The Z5 is Sony's first video camera to feature a G Series lens, the badge formerly reserved for Minolta's highest-quality lenses.
The best feature and most dramatic difference between the Z7 and the Z5 is the Z5's lens. The new lens has a 29.5mm-590mm range (35mm equivalent), which is both wider and longer than the 32mm-384mm Carl Zeiss Z7/S270 lens. The move from a 12x to a 20x lens places the Z5 in direct competition with Canon's XH A1s.
The cost associated with a longer lens is a loss in low-light sensitivity due to lens ramping, the lens's inability to hold aperture effectively while zooming. As Figure 2 (below) illustrates, both lenses have similar iris openings when fully wide (F1.6) until the 41% zoomed-in mark (F2.0), where the Canon lens takes over and has a faster performance until they both hit F3.4 at 97% of full telephoto. The Z7 is the true, low-light lens champ and maintains an F1.6 at full wide and an F2.0 at full telephoto.
Real-World Low-Light Sensitivity Comparison
Now, lens aperture is only one factor in overall low-light sensitivity, and each camera has differing levels of sensitivity based on the size, type, and quantity of sensors (CCD or CMOS); the efficiency of the analog-to-digital converter; and lots of other proprietary technology that we don't need to get into in this article. To better demonstrate real-world, low-light sensitivity, I conducted a series of simple exposure tests to demonstrate how the cameras performed at different distances. Because the four cameras I tested all have different focal lengths, I chose to test at the same absolute distance instead of by the percentage of zoom.
My first test was conducted with four cameras lined in a row at a distance of 9', and my subjects were Teemu Selanne and Nikolai Khabibulin hockey figurines with a Lego crocodile and a lion cub. In order to set my exposure, I set the Zebras on both cameras to 100 to tell me when the highlights on Selanne's helmet were starting to bloom.
Getting identical exposure between cameras was critical for comparison purposes, so I verified my results using the waveform monitor in Adobe Premiere Pro CS3 and disqualified any results in which the exposure was more than 5 IRE different between cameras. (Note: Because the HMC150 records in AVCHD and does not have a tape option, I did not record or review its footage using a waveform monitor, and my results are provided solely for discussion purposes for the first two tests.)
The 20x Z5 and XH A1s were zoomed in to 89% and 87%, and the Z5 had an iris setting of F6.2, while the Canon was slightly behind with an iris setting of F4.8 (Figure 3, below). The 12x Z7 and the 13x HMC150 were zoomed at 97% and 99% and had F-stop settings of F6.8 and F4.0.
In this test, a higher F-stop indicates a camera that has better low-light sensitivity, so both Sony models had clear leads over the Canon and Panasonic models, as shown in Table 1 (below).
I conducted a second test with the same camera placements but had the cameras at their widest zoom setting; the results were predictably the same, but I did notice that the Panasonic LCD had too low a resolution to allow me to get an accurate indication of when Selanne's helmet crossed the 100 IRE threshold.
My third test was another fixed-distance test, this time at 6', but I was testing both how open I could get the iris when I zoomed-in and how bright the resulting image would be. The XH A1s made up some ground, due in part to a faster lens when zoomed in more than 40%, and its iris was F2.8 at 81% zoomed, while the Z5 was F3.1 at 83%. Ultimately, what matters in this test is not as much the smallest F-stop number but the overall exposure, and the Sony Z5 bested the XH A1s with an image that was brighter (Figure 4, below).
Back in the editing suite, I compared waveform indicators and focused on a chunk of highlights and midtones. The highlight I compared was 90 IRE on the Z5 and 85 IRE on the XH A1, and the midtone was 40 and 35, indicating an exposure advantage of 5 IRE for the Z5 at this specific distance and focal length. Predictably, the Z7 bested both the Z5 and XH A1s by a considerable margin, due largely to a faster lens that was F2.0 at 91% zoomed. Its midtone sample was 55 IRE, and the highlights were off the chart-but likely with a 20 IRE advantage over the XH A1s at this test, as shown in Table 2 (below).
I also want to note that in all the tests of the XH A1s, and from my extensive use of my old XH A1, I noticed that the colors appeared somewhat dull. You can see for yourself in the side-by-side comparisons. For testing purposes, I didn't use any picture profiles and I white-balanced each camera to a Vortex Media white-balance reference card to ensure consistent results from these cameras in their out-of-the-box color settings. I tested this observation with Premiere Pro's Vectorscope, which agreed that the XH A1s' chrominance trace was less saturated in the green, yellow, blue, and red hues, which correspond with the crocodile, the lion cub, and the Winnipeg Jets hockey figures.
Of the four cameras tested, only the Z7U allows you to change lenses. Replacing the stock 12x zoom lens with a lens that is similar to the focal range of the 4.1-82mm Z5 G series lens will cost $2,995. The Fujinon Th17x5BRM broadcast lens has a 5-85mm focal length, which provides a bit of extra room on the telephoto end at the expense of the wide angle. The benefit of this broadcast lens is that it improves on the stock 12x Zeiss lens that starts at F1.6 and ramps down to F2.0 at full telephoto. The Fujinon 17x lens has a maximum relative aperture of F1.4, which it maintains through its full telephoto range. If price is no object and you want a wider lens, you will want to look at the $6,599.95 Fujinon 13x3.5BRM (3.5mm-46mm) or the $8,995.95 Fujinon HTs18x4.2BRM (4.2mm-76mm), or-for even more telephoto range-the $8,819.95 Canon KT20x5B KRS (5mm-100mm).
Replacing the stock Zeiss lens with a 1/3" bayonet-mount lens means you may lose some of the camera's automatic controls, optical image stabilizer, and the functionality of the LANC controller. Each lens has its own replacement zoom controller, so for the 13x and 17x Fujinon lenses, budget an additional $899.95 for the SRD-92B zoom rate demand unit and more for the 18x model.
As you can see, some of the options will end up costing more than the camera itself, so this leaves Z7U users who are looking for more zoom with two options. The first option is to use a digital extender. The Z7U allows you to supplement the optical zoom with digital zoom at the cost of resolution. Not the perfect solution, but it doesn't cost a dime.
The second option is to use a telephoto conversion lens. I've been using a Century Precision Optics 0HD-16TC-Z7U 1.6x telephoto converter lens (Figure 5, below) and have been very pleased with the results. I tested its light loss in my first test; to achieve the same framing, it reduced my zoom from 97% to 87%, but it increased my F stop from F6.8 to F6.2, which brings it in line with the light sensitivity of the Z5U. Unlike my old conversion lenses, the Century lens is not a threaded model, but it uses a bayonet mount for a quick and secure connection. Adding 1.75 lbs. of glass to the front of your camera comes at a cost, which is often the ability to balance your tripod. I was unable to get my old Manfrotto 503 head to balance with the added and front-heavy weight. It was also a bit much for my Vinten Vision 3 tripod with a No. 1 spring, so I shelled out $49.95 for a heavier-duty No. 3 counterweight spring, and it balances nicely.
A telephoto lens also reduces your wide shot as the barrel is visible when you are wider than 50%, so it may not be the best solution when you need to quickly change between a wide shot and a tight shot. The street price for the telephoto lens is $799.95.
In addition to the option of going more telephoto, Century Optics also offers four options for going wider. I tested the 0.75 wide-angle, the 0.6 wide-angle, and the 0.55 fisheye; there is even a 0.3 ultra fisheye. I found the 0.75 to be the most useful; with the lens mounted, I could still focus manually, and I had full zoom-through. The 0.6 was limited to 75% of telephoto, but it worked only with autofocus and added some barrel distortion (curvature).
The 0.55 fisheye is designed for increased barrel distortion and had no zoom-through. Like the 0.6, it could only be used with the autofocus. Of all the conversion lenses I tested, the fisheye was the most fun, as it exaggerates depth; it pulls the foreground closer and pushes the background farther back.
Ultimately, your choice of camera is going to be determined by your preferences, your requirements, and the fit with your video business. For my business, the pairing of the HVR-Z7U and HVR-Z5U is the clear winner in this latest face-off of high-definition video cameras.
Shawn Lam (video at shawnlam.ca) runs Shawn Lam Video, a Vancouver video production studio. He specializes in stage event and corporate video production and has presented seminars at WEVA Expo 2005-7 and Video 07. He won a Silver Creative Excellence Award at WEVA Expo 2008 and an Emerald Artistic Achievement Award at Video 08.