Abstract: A non-blocking always visible display application is provided to copy and save first pixel values corresponding to a first display screen area, blend the copied first pixel values with second pixel values corresponding to a non-blocking always visible display to generate third pixel values, and replace the original first pixel values with the third pixel values to effectuate display of the non-blocking always visible display. In one embodiment, the application further monitors for display operations that impact the first display screen area, and re-blend accordingly. In one embodiment, the re-blending and replacement are advantageously delayed. In another embodiment, the application further intercepts and causes cursor events to be handled properly, based at least in part on a current blending bias between the non-blocking always visible display, and the underlying display windows.

Abstract: A computing device is provided with a number of programming instructions to cause display of first execution results of a first set of applications in a first plane of a metaphoric desktop, and display of second execution results of a second set of applications in a second plane of the metaphoric desktop. In one embodiment, the programming instructions are further designed to morph the metaphoric desktop from one plane to another. In one embodiment, the second set of applications are on-line applications, and the programming instructions are designed to cause the metaphoric desktop to morph from the first plane to the second plane when the computing device is being connected on line; and cause the metaphoric desktop to morph back to the first plane in response to a user request to return to the first plane.

Abstract: In many computer systems, the rate of data from an input source may become significantly faster than a processor's ability to respond. This situation occurs when a faster and more modern system writes to an older and slower legacy system. The result of this interaction between disparate capacity systems may be system failure or data corruption, particularly due to overloading the slower system's input queue to the point where it can no longer catch up to its workload. The invention provides a method, system, and computer-readable medium having computer-executable instructions for transferring data between disparate capacity systems. The method comprises the steps of receiving transactions from a first computer system at a first rate and providing the transactions to a second computer system, wherein the second computer system receives the transactions at a second rate.

Abstract: A technique for providing PCI bus mastering compatibility for legacy PCI bus devices which may not support PCI bus mastering RETRY protocols. DLDMM provider software in a device driver for a target device may provide a callback signal at a callback address to DLDMM client software in a device driver for a bus mastering PCI device. The callback address may be used by the DLDMM client software to signal the bus mastering PCI device to suspend operation in the event of an interrupting event. The bus mastering PCI device may then generate in driver software a RETRY signal to the device driver of the target device. If the interrupting event is over, a signal may be sent by the DLDMM provider software in response to the RETRY signal to the DLDMM client software indicating that the bus mastering device may resume operation. The bus mastering device may then resume operation where left off.

Abstract: A multimedia system including an audio/video decoder/decompresser for decoding/decompressing a compressed encoded audio/video data stream to generate video images for display and audio signals for audible reproduction. The multimedia system includes an integrated system and video decoder with an audio/video synchronization circuit for substantially synchronizing the display of video images with audio playback. In addition, a step control is provided to allow for viewing of video images on a frame-by frame basis or to freeze or play video in slow motion. When step control is activated, audio output is muted. Audio data corresponding to displayed video is transmitted to the muted audio decoder. An internal system clock may be suppressed to the system clock counter. An external CPU may provide system clock start times corresponding to video frames to be displayed. The external CPU may increment the system clock counter by a an amount corresponding to the difference between a successive frame or number of frames.

Abstract: A multimedia system including an audio/video decoder/decompresser for decoding/decompressing a compressed encoded audio/video data stream to generate video images for display and audio signals for audible reproduction. The multimedia system includes an integrated system and video decoder with an audio/video synchronization circuit for substantially synchronizing the display of video images with audio playback. An audio presentation time stamp (APTS) is detected in the compressed/coded audio data stream in the integrated system and video decoder and stored in a data latch. The compressed/coded audio data stream is fed to an audio decoder which decodes/decompresses the audio data and outputs and audio signal. The audio decoder detects when audio data corresponding to an APTS had been output and sets a corresponding flag. The flag indicates to the integrated system and video decoder that a corresponding audio segment had been decoded/decompressed and output.

Abstract: A multimedia system including an audio/video decoder/decompresser for decoding/decompressing a compressed encoded audio/video data stream to generate video images for display and audio signals for audible reproduction. The multimedia system includes an integrated system and video decoder with an audio/video synchronization circuit for substancially synchronizing the display of video images with audio playback. A method is described for detecting when the playback of audio and the display of video images are out of synchronization. The circuitry includes three programmable registers, a finite state machine and one full adder using an audio presentation time stamp and the video presentation time stamp. The method uses a rounded programmable bias value which is compared with the difference between the video presentation time stamp and the audio presentation time stamps.