>> continued from previous page
Misunderstood MPEGs
MPEG-3 means two different things to people. One is correct; the other is not. MPEG-3 was originally designed to replace MPEG-2 in the HDTV (High-Definition Television) market. MPEG-3 supported coded bit rates up to 40 Mbps at resolutions up to 1,920x 1,080 pixels. With a little tweaking, however, MPEG-2 could handle anything that HDTV or CCIR 601 (720x 480 digital video) could throw at it, so the chances of MPEG-3's being used ever becoming important are slim to none.
Many people call MP3 MPEG-3, and it's an egregious misnomer. MP3 is an audio-only encoding technology that has taken off in the personal music scene. MP3 lets audio be stored in a small space. But MP3 has been around since the early days of MPEG-1. MP3 is actually MPEG-1 Layer III, meaning it is the audio portion of MPEG-1. Drawing from the constant bit rate of MPEG-1, MP3 encodes its audio as a constant bitstream of 128, 160 or 192 Kbps.
To make matters more confusing, MPEG-4's (ISO 14496) file format is sometimes referred to as MP4. For those keeping score, MP3 is not MPEG-3, but MP4 is MPEG-4. To wade through the muddle surrounding MPEG-4, you have to understand where it came from. MPEG-4 uses Apple Computer's QuickTime file format as its basis. So referring to MPEG-4 as QuickTime is almost correct. Envisioned in 1993, version 1 reared its head in 1998 and version 2 arrived in 1999.
MPEG-4 takes up where MPEG-1 and MPEG-2 left off. MPEG-4 is designed to provide a framework for digital content to be delivered to Internet and wireless devices. QuickTime was chosen over Microsoft's ASF (Advanced Streaming Format), adding one last layer of confusion as Microsoft has now added ASF support to its Windows Media Player. But simply supporting ASF doesn't always translate into support for true MPEG-4, so watch out. Other vendors, including DivX ;-), followed Microsoft's lead and added support for ASF to their products. One vendor, besides Apple, that offers true support for MPEG-4 is RealNetworks with its latest RealPlayer.
MPEG-4 operates as its forebears did but adds features like being able to dynamically adjust to network speeds. It also allows multiple layers to be transmitted and played back. Basically, MPEG-4 is designed to be a one-stop standard for delivering rich multimedia content to digital devices. Because it is an open standard, it may not include the tightest codecs, but it can be adopted by anyone that wants it.
The Newest Members
MPEG-7 is defined as Multimedia Content Description Interface, which sounds a little like what MPEG-4 was originally designed for. But again, watch out. MPEG-7 is an odd duck. It's not even a video codec, though it does describe video content. Whereas MPEG-4 defines the file format for delivering interactive multimedia content, MPEG-7 includes tools that describe the content. While most users won't care about MPEG-7, CDNs (content-delivery networks) will find the tools invaluable for cataloging and searching A/V content. MPEG-7 does not depend on how the content was encoded. The final portions of MPEG-7 are due to be formalized late this year.
MPEG-21 is the latest in the MPEG family, with work begun in June 2000. With the previous MPEG standards, what has been missing is the big picture, an overall standard that could describe everything about a stream and what it's used for. Like MPEG-7, MPEG-21 is not a codec. However, MPEG-21 holds the promise of defining a multimedia framework for all other codecs and file formats. The framework might eventually prove useful for describing how different forms of multimedia transmission work and fit together. But it's too soon to tell.
Other Codecs
The digital video world is rife with different codecs to deliver video to the desktop: Cinepak, Sorenson, RealAudio/Video and Windows Media. Each is capable of delivering good quality video at different bit rates. What distinguishes them from MPEG is largely that they are designed for lower data rates. All can deliver data at rates below 50 Kbps and are proprietary. But for the most part, they work like their MPEG counterparts.
The 10 years since MPEG-1 was first used has provided serious advances in compression. The algorithms have allowed for higher compression rates in smaller bit rates. When asked, most of the vendors in this space say there is no limit to how small the video signal can be compressed and retain VHS or higher quality.
Darrin Woods is a technology editor of Network Computing. Darrin has worked as a WAN engineer for a telecom carrier. Send your comments on this article to him at dwoods@nwc.com.
RSS
