Abstract:
A general purpose, programmable media processor (12) for processing and transmitting a media data streams. The media processor (12) incorporates an execution unit (100) that maintains substantially peak data through out of media data streams. The execution unit (100) includes a dynamically partionable multi-precision arithmetic unit (102), programmable switch (104) and programmable extended mathematical element (106). A high bandwidth external interface (124) supplies media data streams at substantially peak rates to a general purpose register file (110) and the execution unit. A memory management unit, and instruction and data cache/buffers (118, 120) are provided. The general purpose, programmable media processor (12) is disposed in a network fabric consisting of fiber optic cable, coaxial cable and twisted pair wires to transmit, process and receive single or unified media data streams.
Abstract:
A general purpose, programmable media processor (12) for processing and transmitting a media data streams. The media processor (12) incorporates an execution unit (100) that maintains substantially peak data through out of media data streams. The execution unit (100) includes a dynamically partionable multi-precision arithmetic unit (102), programmable switch (104) and programmable extended mathematical element (106). A high bandwidth external interface (124) supplies media data streams at substantially peak rates to a general purpose register file (110) and the execution unit. A memory management unit, and instruction and data cache/buffers (118, 120) are provided. The general purpose, programmable media processor (12) is disposed in a network fabric consisting of fiber optic cable, coaxial cable and twisted pair wires to transmit, process and receive single or unified media data streams.
Abstract:
The present invention provides a cross-bar circuit (100) that implements a switch (115) of a broadband processor. The cross-bar circuit (100) includes: a switch circuit (115) which includes 2m.2n:1 multiplexor circuits (202-204) where each of the 2n:1 multiplexor circuits (202-204) has a unique n-bit index input, one disable input, and a 2n-bit wide source input receives an n-bit index at the n-bit index input, a disable bit at the disable input, and the 2n-bit input source word at the 2n-bit wide source input, and decodes the n-bit index either to select and output as an output destination bit one bit from the 2n-bit input source word if the disable bit has a logic low value; a cache memory (110) that has 2m cache datapath inputs; and 2m cache index input; and a control circuit (105) that has a plurality of control inputs receives the partially decoded instruction information on the plurality of control inputs, provides a second set of the n-bit indexes for the switch circuit (115), and provides the disable bits for the switch circuit (115) where the control circuit (105) is logically coupled to the switch circuit (115) and to the cache memory (110).
Abstract:
A system for processing geometry which reduces the amount of memory spaces while improving the processing speed. The system delivers vertices in sequence to a vertex queue (70) so that data in the vertex queue is freed as it is delivered and only minimal intermediate results are stored. By this incremental evaluation, less memory space is needed. In another aspect of the invention the vertices are maintained in the proper sequence so that sorting operation can be eliminated. A sorted vertex queue (70) and an unsorted vertex list (72) are utilized so that resorting of the entire vertex list may be prevented. In addition, a compressed format (34) for storing geometry is utilized based on the fact that much information can be rederived from a sorted and reduced vertex queue.
Abstract:
Circuit logique BICMOS amélioré (70) utilisant une paire de transistors bipolaires (21, 22) à couplage par émetteurs pour comparer de manière différentielle un signal d'entrée (Vin) et un niveau de référence logique (VBIAS). Chaque transistor bipolaire est chargé de manière résistive par un réseau de transistors (26, 27) à semiconducteur à grille isolée par oxyde métallique à canal P (PMOS) couplés en parallèle. La grille d'au moins une des combinaisons parallèles de transistors est couplée à un signal de commande (VREF2) fournissant une résistance de charge variable. De préférence, le signal de commande est fourni par un réseau de retour (52, 53) servant à maintenir à une valeur constante l'excursion de tension constante dans le réseau, malgré les variations de température.
Abstract:
Transistor bipolaire possédant un émetteur (25), une base (31) et un collecteur (30). Il comprend une zone de base intrinsèque (33) à zones latérales étroites (p-) et à zone centrale plus large (37). Les zones latérales sont contiguës à la zone dopée (31), et la zone centrale (37) se trouve sous l'émetteur (25). Le profil de dopage latéral de la base est tel que les concentrations de dopage dans la zone dopée (31) et la zone centrale (37) sont relativement élevées par rapport à celles dans les zones latérales étroites (p-) de la base intrinsèque (33). L'association des zones latérales étroites (p-) au profil de dopage latéral de la base limite l'étendue de la zone de déplétion à l'intérieur de la base, ce qui assure une réduction de la tension de claquage du transistor sans entraîner une perte du gain en courant.
Abstract:
Expandably wide operations are disclosed in which operands wider than the data path between a processor and memory are used in executing instructions. The expandably wide operands reduce the influence of the characteristics of the associated processor in the design of functional units performing calculations, including the width of the register file, the processor clock rate, the exception subsystem of the processor, and the sequence of operations in loading and use of the operand in a wide cache memory.
Abstract:
A general purpose, programmable media processor (12) for processing and transmitting a media data streams. The media processor (12) incorporates an execution unit (100) that maintains substantially peak data through out of media data streams. The execution unit (100) includes a dynamically partionable multi-precision arithmetic unit (102), programmable switch (104) and programmable extended mathematical element (106). A high bandwidth external interface (124) supplies media data streams at substantially peak rates to a general purpose register file (110) and the execution unit. A memory management unit, and instruction and data cache/buffers (118, 120) are provided. The general purpose, programmable media processor (12) is disposed in a network fabric consisting of fiber optic cable, coaxial cable and twisted pair wires to transmit, process and receive single or unified media data streams.
Abstract:
A general purpose, programmable media processor (12) for processing and transmitting a media data streams. The media processor (12) incorporates an execution unit (100) that maintains substantially peak data through out of media data streams. The execution unit (100) includes a dynamically partionable multi-precision arithmetic unit (102), programmable switch (104) and programmable extended mathematical element (106). A high bandwidth external interface (124) supplies media data streams at substantially peak rates to a general purpose register file (110) and the execution unit. A memory management unit, and instruction and data cache/buffers (118, 120) are provided. The general purpose, programmable media processor (12) is disposed in a network fabric consisting of fiber optic cable, coaxial cable and twisted pair wires to transmit, process and receive single or unified media data streams.
Abstract:
A general purpose processor with four copies of an access unit, with an access instruction fetch queue A-queue (101-104). Each A-queue (101-104) is coupled to an access register file AR (105-108) which is coupled to two access functional units A (109-116). In a typical embodiment, each thread of the processor may have on the order of sixty-four general purpose registers. The access unit functions independently by four simultaneous threads of execution, and each compute control flow by performing arithmetic and branch instructions and access memory by performing load and store instructions. These access units also provide wide specifiers for wide operand instructions. These eight access functional units A (109-116) produce results for access register files (105-108) and memory addresses to a shared memory system (117-120).