Chapter One
Choosing a DV Camcorder
There are many fine analog camcorders out there that deliver lovely quality video that translates to high-quality streaming video or DVD output.
On the other hand, if you're buying a DV camcorder, it's easy to get confused by the multiple formats that are available today and the rich feature sets of the individual camcorders. To help you sort through all the options, this chapter starts by identifying the digital video formats available today and their strengths and weaknesses.
Then, looking primarily at DV camcorders, I'll identify the basic components and features of a camcorder, describing which are important and which aren't. Then we'll look at the current crop of Sony camcorders to identify the types of features that boost the price from $699 to $1,499 and when it's worth spending the extra dollars.
The chapter concludes with a look at the types of features that higher-end camcorders deliver as well as a short section on how to test-drive a camcorder before buying.
Reviewing the Digital Formats
A digital camcorder is any camcorder that stores video in digital rather than analog format. Life was simple about five years ago, because camcorders using DV as the storage format were the only digital camcorders, so if it was digital, it was DV. Now the field has expanded into a number of siblings and kissing cousins that are related to DV but different. Starting with DV, let's review the most relevant members of the digital camcorder family.
DV Format
The first and most widely used format for digital camcorders is DV, an interlaced format with a resolution of 720x480 pixels and a combined audio/video data rate of approximately 3.6MB per second. Resolution and interlacing are discussed in more detail later in the chapter.
DV uses JPEG compression on each frame to achieve roughly a 5:1 compression ratio, which is generally unnoticeable on most video. DV uses two kinds of tapes: full-size and Mini-DV tapes, the latter being about the size of a small matchbox. Virtually all camcorders under $5,000 use Mini-DV tapes, and I'll talk exclusively of Mini-DV cameras from here on out.
DV has several professional offshoots, including DVCPRO, DVCAM, and DVCPro-50 that use heavier-duty tapes and slightly different encoding schemes. None of these formats appear in consumer camcorders, so I won't mention them again.
DV camcorders use a serial port transfer mechanism called IEEE 1394 to send video back and forth to computers. This transfer mechanism also includes "machine control," which allows software on the computer to control the camera's playback mechanism, for playback, fast-forwarding, and rewinding operations.
IEEE 1394 was invented by Apple as FireWire, dubbed iLink by Sony and other names by various vendors. As Billy Joel might say, "It's all 1394 to me," and these devices by any name are almost universally interoperable.
Digital8
Digital8 camcorders use the DV format but store video on Hi8 rather than DV tapes. This was important back in the early years of DV when DV tapes cost $25 or more, which was prohibitive for many consumers. Most, but not all, Digital8 camcorders also can play Hi8 tapes, allowing backwards compatibility for those with significant libraries in this format. In addition to sharing the DV format, Digital8 camcorders also use 1394 to communicate with computers.
Overall, Digital8 camcorders are wonderfully inexpensive mechanisms to enter the world of digital video with. On the other hand, one of the key benefits of the DV format is the robustness of the tapes, and in my experience, solely using Hi8 to capture analog video, are fragile as flowers. Now that prices of entry-level DV camcorders and DV tapes have dropped so significantly, I would probably spend the extra money and go DV.
MicroMV
MicroMV camcorders encode their video in MPEG-2 format, which is more highly compressed than DV, allowing tapes and camcorders to be smaller. The only characteristic they share with DV is the transport mechanism to the computer, FireWire. Sony does enough things differently with MicroMV, however, that few video editors support the format, and those that do, like Pinnacle Studio, had significant startup issues. This, and the inherently lower quality of MPEG-2, makes DV a better choice for those who want to edit digitally on their computers and produce top-quality video.
DVD Camcorders
DVD camcorders store captured video in MPEG-2 format on either DVD-RAM or DVD-R disks. The former is ideal for rewriting and instantly accessing individual scenes without the usual capture process; DVD-R disks are targeted for immediate playback of the videos on a DVD player. The MPEG-2 compression used in these cameras is scalable and peaks at 1.125MB per second for DVDRAM and 750K per second for DVD-R.
These camcorders are fairly new and share many of the same issues as MicroMV, including reduced quality and little compatibility with video-editing programs. For these reasons, unless you have a compelling need to capture directly to DVD, you're probably better off with a DV-based camcorder.
Table 1-1 summarizes the similarities and differences between the four formats discussed so far.
Camcorder Basics
Now that we've reviewed the camcorder formats, let's focus our attention on the camcorders themselves, starting with the major components and then working our way down to individual features.
Charge Coupled Devices
Charge Coupled Devices (CCDs) are the electronic chips that sense the image coming from the lens. When comparing camcorders, we care primarily about the number of CCDs employed by the camcorder, their physical size, and the number of pixels each chip can resolve.
THE NUMBER OF CCDS
Briefly, CCDs use light-sensitive regions that create electrical charges based upon the intensity of the light hitting the CCD, distinguishing between bright and dark regions by measuring the electrical charge at each pixel. To discern colors, camcorders with three CCDs split the incoming light with a prism, sending red, green, and blue streams to the respective CCDs and then merging the three signals electronically.
In contrast, single-CCD camcorders use an array of colored filters over the CCD to separate the colors, which is less effective. For this reason, three-CCD camcorders almost always produce video with superior clarity and color accuracy.
As you might expect, most consumer camcorders use one CCD while all prosumer and professional camcorders use three CCDs. In the past, most three-CCD camcorders started at around $2,000, but Panasonic announced a three-CCD camcorder for under $1,000 in mid-2003, and I'm sure other vendors will follow suit.
Probably the most significant determinant of quality and price between camcorders is the number of CCDs used in the camera. Whenever comparing camcorders, be sure to identify this first.
CCD SIZE
CCDs typically range in size between 1/6 and 1/2 inch. Though some experts claim that CCD size doesn't directly relate to quality, higher-end camcorders typically use larger chips than consumer models.
Still, this is a fairly esoteric metric; in several tests at PC Magazine, camcorders with smaller CCDs outperformed those with larger. For this reason, I wouldn't pick one camcorder over another because it has larger CCDs.
CCD RESOLUTION
On the other hand, CCD resolution, or the number of pixels captured by the CCD, is a very key metric, especially if you plan to use the camcorder for capturing still images as well as video. Interestingly, if the camcorders' primary focus is DV video, which has a resolution of 720x480 pixels, the ability to capture about 340,000 pixels is sufficient, and more is a waste. That's why CCDs for prosumer camcorders, like Sony's DCR-VX2000, have fewer pixels than many consumer camcorders.
In contrast, more pixels are better on camcorders designed to capture high-quality stills, and most vendors offer models with 1.5-or 2-megapixel CCDs for still-image capture. When comparing camcorders, be sure to identify both the CCD resolution and maximum picture resolution, typically represented as 640x480, 1600x1200, or similar numbers.
If the number of pixels in the largest image exceeds the number of pixels in the CCD, the camcorder is zooming the image digitally, which delivers less quality than a pixel-for-pixel image capture. For example, the JVC DV3000U camcorder has a 1.33-megapixel CCD with 1,330,000 pixels yet outputs images as large as 1600x1200, which really requires 1.92 megapixels, derived by multiplying 1,600 times 1,200.
In contrast, the Sony DCR-TRV80 shares the same maximum image size of 1600x1200 but creates the picture with a true 2.1-megapixel CCD. All other things being equal, the larger CCD will deliver a higher-quality image.
In short, as with still image digital cameras, if the picture output in pixels is larger than the actual pixels on the CCD, the camera is zooming the image digitally, which you could do in your image editor. Obviously, however, this doesn't add any quality. Accordingly, when assessing a DV camera's still-image capabilities, compare pixels on the CCD, not output size, which can be arbitrary.
SCAN CAPABILITIES (INTERLACED VERSUS PROGRESSIVE)
Scan capabilities refers to the technique used to store the incoming frames. There are two methods, interlaced and progressive.
Televisions in the United States operate under the NTSC (National Television Standards Committee) standard. NTSC video consists of 29.97 frames per second, with each frame made up of two fields, essentially two half frames, the first containing odd lines (1, 3, 5, 7) and the second even lines (2, 4, 6, 8). When shooting NTSC video, the camcorder actually shoots about 60 times a second. The first shot becomes field one and contains the odd lines in the first frame. The second shot becomes field two and contains the even lines in the first frame. Because the two fields are combined to make a frame, this video is considered interlaced.
Interlaced video works well at 60 fields per second but causes problems when shooting for still-image capture, especially if there's high motion in the video. That's because a surprising amount of motion can occur during the 1/60 of a second between the two fields, resulting in two pictures of slightly different things. This is shown on the left in Figure 1-1, where the two fields don't combine into a matched frame.
In contrast, progressive scan CCDs store images from top to bottom (lines 1, 2, 3, 4, and so on), just like digital still-image cameras. When shooting in progressive scan mode, DV camcorders shoot only 30 times a second, capturing a complete frame from top to bottom, then dividing each frame into the two fields required for NTSC compatibility, one with even lines and one with odd lines. Similarly, when capturing still images only, the progressive scan camcorder shoots only once, just like a digital still-image camera. As you can see on the right in Figure 1-1, a progressive image of the same motion is very sharp, because the two fields precisely match.
In practice, the image on the left in Figure 1-1 is an extreme example, because camcorders use deinterlacing techniques to minimize the artifacts, and we disabled this for our tests. Still, a camcorder shooting a still image in interlaced mode must combine two disparate fields to create a single frame, generally resulting in a slightly blurry image like that shown in Figure 1-2. Clearly, if you're shooting for still-image capture, the progressive scan technique provides better quality than interlaced, even after de-interlacing.
Interestingly, progressive scan capabilities first appeared to produce video that could easily be converted to film, which consists of 24 discrete frames per second. That's why newer DV camcorders like Panasonic's DVX100 can capture 24 frames per second in progressive scan mode. Progressive scan capture is also useful, but not essential, when converting to frame-based digital formats like MPEG-1 or streaming formats like Real, QuickTime, or Windows Media Technologies.
As I'll discuss in more detail below, if you already have a high-resolution still-image camera and just want to capture video for video's sake, progressive scan probably isn't a feature worth paying extra for. If you're looking for a high-resolution CCD to capture high-quality still images and video, you should verify that the camcorder captures in progressive scan mode, but virtually all camcorders with high-resolution CCDs do.
If you're looking to produce a movie that ultimately will be displayed on film, as in a theater, progressive scan is essential. Be sure to identify the number of progressive frames per second (fps) the camcorder outputs, however, because some, like the VX2000, produce only 15 fps, which is inadequate for film. Most DV camcorders produce 30 fps, which can be converted to 24 fps for film, though native 24 fps, or the ability to capture directly at 24 fps, is preferable. (Check out Table 1-2 for a comparison of CCD types.)
Analyzing Lens Capabilities
Most major camcorder vendors source their lenses from other vendors, with Sony buying from legendary vendor, Carl Zeiss, for many cameras, while Panasonic and Canon employ DICOMAR lenses from equally legendary vendor, Leica. Both brands have their advocates, some quite vocal, and they're both excellent companies, which complicates a brand-based buying decision.
Instead, when I compare camcorders, I focus on comparative performance, particularly how they perform in low-light conditions. Figures 1-3 and 1-4 show why. Though low-light performance is a function of CCDs and camcorder electronics as well as lenses, I'll discuss it here.
Briefly, I took these videos using Sony's three-CCD DCR-VX2000 on the left and a one-CCD consumer camcorder on the right. (I won't name the camcorder because it's simply not a fair comparison; no consumer camcorder can compete with the VX2000.) I placed both camcorders in fully automatic mode and then shot the videos, shutting lights and shades during the shoot to darken the scene.
On the CD-ROM
An MPEG-1 file containing side-by-side shots of the videos is on the CD-ROM, titled Figure 1-3.MPG.
In Figure 1-3, the scene is generally well lit, and the consumer camcorder holds its own. If you were seeing the image in color, you would probably notice that Rosie's sweater (she's on the left) isn't quite the right color and that the contrast and detail overall isn't quite as good. Still, you probably wouldn't pay $1,500 extra for the camcorder on the left based upon this image.
However, in Figure 1-4, when lighting conditions are poor, the VX2000 still produces a usable image while the one-CCD camcorder is pretty much worthless.
Continues...
Excerpted from PC Magazine Guide to Digital Video by Jan Ozer Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
