Chromotion Decode

Windows Media Video 9 (WMV9) Decoding

The Windows Media Video 9 (WMV9) encoding process allows a higher compression ratio than traditional MPEG-2 encoding processes. This efficient compression algorithm enables consumer-quality video both over the internet and within applications.

As with most encoding processes, WMV9 also produces some inherent visual artifacts. Therefore a decoder must perform a number of advanced processes to be able to produce an image which results in a good viewing experience for a user. These decoding processes can be done either in hardware or software or with a combination of both. When done in software, CPU utilization increases dramatically, performance decreases, and the result is often less than high quality video. For example, on a mainstream platform using a 2.66GHz CPU, the WMV9 decoding process done through software will utilize as much as 40% more CPU resources than when the decode is done utilizing graphics hardware. This frees the CPU to be used for other processes while playing WMV9 files, which will give the viewer a fuller viewing experience.

By efficiently performing WMV9 advanced motion compensation decode in hardware, Chromotion dramatically reduces CPU utilization, increases performance, and improves the viewing of WMV9 videos on high quality HDTV displays.

IDCT and Motion Compensation

During the MPEG-2 encoding process, a mathematical function called a discrete cosine transform (DCT) is applied to all of the MPEG-2 video content, which makes the content much easier to compress. During the decode process, the CPU used to do the inverse, the inverse discrete cosine transform (IDCT). Now, to offload the CPU workload, a high performance graphics processor does the IDCT through its IDCT engine, reversing the entire DCT process during MPEG-2 decoding, and decompressing the image for display on the screen. Now much of the decode workload is off loaded from the CPU to the graphics processor's IDCT engine.

Once S3 Graphics' IDCT engine, which has been optimized for Microsoft's DirectX Video Acceleration standard, completes the decompression of the video data, its motion compensation algorithm comes into play to generate the final fully decoded video images. Motion compensation uses a concept known as predictive coding. Typically, only a fraction of an image changes from frame-to-frame, which makes it quite easy to predict future frames from previous frames. Motion compensation is used as part of the predictive process. If an image sequence includes moving objects, then their motion within the scene can be measured, and this information may be used to predict the content of frames in sequence. Without proper motion compensation hardware support, you are likely to see video artifacts, or banding in areas of gradually changing color. This engine minimizes the extra overhead associated with decoding of MPEG-2 video files and allows maximum efficiency during DVD playback.