摘要:
A switching circuit which comprises a digital circuit (11) formed of a superconductor, a transmission line (14) connected to the digital circuit (11) through magnetic coupling (12,13), and a resistor element (17) disposed in the transmission line (14) for differentiating the output of the digital circuit (11).
摘要:
An input buffer circuit having a source follower circuit composed of a first FET (Q 1 ) whose gate electrode has an input (V in ) connected thereto, and a second FET (Q 2 ) of the same conductivity type as that of the first FET, whose drain electrode is connected to a source electrode of the first FET (Qi) directly or through at least one level-shifting diode (D i ) and whose gate electrode is supplied with a control voltage (V cont ); and a FET inverter circuit (Q 3 , 0 4 ) which is connected to the drain electrode of the second FET directly or through at least one level-shifting diode (D 2 ); an output signal being derived from the FET inverter circuit (Q 3 , Q 4 ).
摘要:
An input buffer circuit having a source follower circuit composed of a first FET (Q 1 ) whose gate electrode has an input (V in ) connected thereto, and a second FET (Q 2 ) of the same conductivity type as that of the first FET, whose drain electrode is connected to a source electrode of the first FET (Qi) directly or through at least one level-shifting diode (D i ) and whose gate electrode is supplied with a control voltage (V cont ); and a FET inverter circuit (Q 3 , 0 4 ) which is connected to the drain electrode of the second FET directly or through at least one level-shifting diode (D 2 ); an output signal being derived from the FET inverter circuit (Q 3 , Q 4 ).
摘要:
An information processing system includes a plurality of functional blocks (neurons) (100) and a data bus (300) for transmitting in common the outputs of the individual functional blocks (neurons). Data transaction among the functional blocks (neurons) is performed through the data bus on the time-division basis. For preventing the outputs from conflicting or competition, addresses are assigned to the individual blocks (neurons), respectively, so that only the functional block (neuron) having the own address designated by the address signal supplied through an address bus (302) outputs data signal onto the data bus, while the other functional blocks (neurons) receive the information on the data bus as the signal originating in the functional block whose address is designated at that time point. The addresses are sequentially changed. During a round of the addresses signals, data are transmitted from given functional blocks (neurons) to other given functional blocks (neurons).
摘要:
@ Disclosed is a cooling structure cooling a multichip module for effectively removing heat generated from integrated circuit chips. A suction plate (30) formed with minute grooves (31) on one of its surfaces is disposed between each of the integrated circuit chip (2) and an associated cooling block (20) and is brought into contact at the grooved surface with the integrated circuit chip through a layer of a liquid (200) such as silicone oil interposed therebetween, thereby producing negative hydrostatic pressure by the capillary action of the grooves. The suction plate (30) has a thickness small enough to be bent under influence of the negative hydrostatic pressure to follow up warping of the integrated circuit chip, so that the clearance between the opposing surfaces of the suction plate and the integrated circuit chip can be minimized. An integrated circuit chip cooling device of surface-to-surface contact type operable with a low thermal resistance is provided, which can improve the maintainability and reliability without sacrificing the cooling efficiency.
摘要:
An information processing system includes a plurality of functional blocks (neurons) (100) and a data bus (300) for transmitting in common the outputs of the individual functional blocks (neurons). Data transaction among the functional blocks (neurons) is performed through the data bus on the time-division basis. For preventing the outputs from conflicting or competition, addresses are assigned to the individual blocks (neurons), respectively, so that only the functional block (neuron) having the own address designated by the address signal supplied through an address bus (302) outputs data signal onto the data bus, while the other functional blocks (neurons) receive the information on the data bus as the signal originating in the functional block whose address is designated at that time point. The addresses are sequentially changed. During a round of the addresses signals, data are transmitted from given functional blocks (neurons) to other given functional blocks (neurons).
摘要:
@ Disclosed is a cooling structure cooling a multichip module for effectively removing heat generated from integrated circuit chips. A suction plate (30) formed with minute grooves (31) on one of its surfaces is disposed between each of the integrated circuit chip (2) and an associated cooling block (20) and is brought into contact at the grooved surface with the integrated circuit chip through a layer of a liquid (200) such as silicone oil interposed therebetween, thereby producing negative hydrostatic pressure by the capillary action of the grooves. The suction plate (30) has a thickness small enough to be bent under influence of the negative hydrostatic pressure to follow up warping of the integrated circuit chip, so that the clearance between the opposing surfaces of the suction plate and the integrated circuit chip can be minimized. An integrated circuit chip cooling device of surface-to-surface contact type operable with a low thermal resistance is provided, which can improve the maintainability and reliability without sacrificing the cooling efficiency.