Zenith Data Systems Corporation

Zenith Electronics Corporation

Packard Bell Nec

Anthony M Olson

Babu Rajaram

Thomas N Robinson

Mark D Nicol

Todd R Witkowski

Jimmy Dean Smith

Gordon L Helm

Mark J Foster

Richard A Hamersley

William J McAuliffe

A system for upgrading a personal computer includes a motherboard having upgrade sockets for upgrading the CPU and the clock oscillator without the need to remove any components. The system includes sensing circuitry for detecting the type of component plugged into the upgrade socket and circuitry for disabling the upgraded component. In addition, the system includes software for reconfiguring various signals depending on the particular upgrade plugged into the upgrade socket. By providing a system that can be upgraded by merely inserting a newer components, upgrades can be performed rather quickly. In addition, the upgrade system allows the end user a plurality of upgrade options while at the same time allows the end user to take advantage of declining CPU prices.

Processor upgrade system for a personal computer

A computer system includes a memory capable of storing a plurality of data words, and a central processing unit for outputting data words to be stored in the memory. A method and apparatus for facilitating transfer of the data words from the central processing unit to the memory involve accepting and temporarily storing in a storage portion each data word from the central processing unit and then subsequently storing in the memory each temporarily stored data word, the maximum number of data words which can be temporarily stored being selectively set to one of first and second values which are different.

Variable size queue circuit for buffering data transfers from a

A shared bus arbitration system is disclosed which provides logic allowing multiple processors to co-exist on a common bus. In the present invention, the host processor is isolated from the bus by a posted write array or write buffer. The arbitration system accepts bus lock and cycle signals when the processor writes a locked instruction to the posted write array and provides a bus lock signal to the bus when the locked instructions are written to the bus.

CPU lock logic for corrected operation with a posted write array

A computer system includes a control circuit having a request input, and a pair of connectors which are electrically coupled to the control circuit and are each connected to the request input of the control circuit, each connector being capable of removably receiving a circuit card. The control circuit includes a checking arrangement for detecting a condition in which the connectors each have therein a card having a request output controlling the request input of the control circuit.

Dual connector port for bus master card

A system includes a main memory having a plurality of sections which each include a plurality of selectively addressable storage locations, a cache memory, and an accessing arrangement for successively requesting data from respective locations in the main memory. A method and apparatus for controlling the system involve assigning each section of the main memory a changeable status condition which is one of a caching enabled status and a caching disabled status, and inhibiting reading and storing of data by the cache memory when data requested by the accessing unit is in one of the sections of the main memory having the caching disabled status. An alternative method and apparatus for controlling the system involve selective operation in a mode in which data in the cache memory is updated even when reading of data from the cache memory is inhibited.

Programmable cache memory which associates each section of main memory

A system includes a system bus having data lines, an acknowledge line, an enable line, and a control line, a data storage device, a controller circuit, and an arrangement coupling the system bus, controller circuit and data storage device. The system bus can carry out a data transfer cycle in which the acknowledge, enable and control lines are actuated and the controller obtains and checks data from the data storage device and supplies it to data lines of the bus, and a verify cycle in which the acknowledge and enable lines are actuated and the control line is deactuated and the controller obtains and checks data from the storage device but does not supply it to the bus. The controller circuit is capable of operating in different modes, in one of which it forcibly sets a false error indication in response to the verify cycle. In response to the acknowledge line and enable line being simultaneously actuated while the control line remains deactuated throughout a cycle, an arrangement supplies a special signal to the controller circuit so that the controller circuit interprets the cycle as a data transfer cycle and does not set a false error indication.

Floppy disk controller interface for suppressing false verify cycle

A high speed memory component, such as a cache RAM (random access memory), has a bidirectional data line and a control line (CDRAMOE) which, when respectively set to first and second electrical states, causes the memory component to respectively output data onto and accept data from the bidirectional data line. An arrangement is provided to initiate transfers of data to and from the memory component on the bidirectional data line and to produce at a first point in time a valid indication signal (W/R) indicating that one of a data read and a data write is being initiated. A method and apparatus for controlling the memory component involve the control line being set to the second electrical state at a second point in time prior to the first point in time, and the control line being maintained in the second electrical state from the second point in time to the first point in time.

Method and apparatus for pipelining cache accesses using anticipatory

In response to an address decoded in a preselected range, a multiplexer combines translated high order address bits with CPU-generated low order address bits to access random access memory, especially reserved range random access memory. Otherwise, the multiplexer merely combines CPU-generated low order bits with CPU-generated high order address bits to access RAM. An expanded memory specification memory map drives the translator to generate the translated high order address bits. This generates the address for reserved range RAM. RAM contents, normal and reserved range, are available for processing by the CPU.

Method of combining lower order and translated upper order bits to

A memory system includes a memory unit with plural addressable storage locations, a connector arrangement for detachably electrically coupling the memory unit to a computer, and a decoding arrangement for determining whether a computer memory address is within a range of addresses to which the memory unit is to respond, the decoding arrangement including a base address register storing a value representing the first address in the range and a size register defining the number of addresses in the range. A method by which a computer can automatically configure this system includes the steps of determining a base address to be associated with the connector, loading into the base address register a value representing the base address, loading into the size register a value representing the maximum range of addresses, thereafter selectively accessing storage locations to determine the actual number of contiguous locations in the memory unit, and then changing the size register, when the actual number of locations is less than the maximum range of addresses, to a value representing the actual number of locations.

Method and apparatus for automatic memory configuration by a computer

A computer system reads data from selected locations in a memory while each address applied to the memory is temporarily stored in a register. If a data error is detected, two flipflops are set, one of which can generate and interrupt the central processing unit, and the other of which disables an AND gate through which a load signal is applied to the register in order to disable the register. An arrangement for resetting the second flipflop includes a third flipflop which is set and reset respectively at the beginning and end of each input/output cycle and which has an output coupled to one input of an AND gate having a further input to which is applied a signal selectively actuable by the central processing unit, the output of the AND gate and a system reset signal being applied to inputs of an OR gate which has its output connected to a reset input of the second flipflop.

Method and apparatus for error detection and localization

Four parallel, spaced arrays of apertures are provided in a printed circuit (PC) board to accommodate mounting of 16- and 18-pin dynamic random access memories (DRAMs) within a minimum area footprint. First and second 8-aperture linear arrays are provided in the PC board in spaced relation to accommodate 16-pin DRAMs as are third and fourth 9-aperture linear arrays to accommodate 18-pin DRAMs, wherein the third and fourth arrays of apertures are aligned parallel with and respectively positioned intermediate and outside of the first and second arrays of apertures in the PC board. Corresponding apertures in the 16- and 18-aperture arrays for receiving corresponding pins of the 16- and 18- pin DRAMs are electrically coupled, with the two additional apertures of the 18-aperture array extending beyond the 16-aperture array and aligned therewith. The limited footprint area is thus capable of receiving, mounting and electrically coupling in circuit either a 16-pin or an 18-pin DRAM or any integrated circuit having a dual-in-line pinout configuration.

Mounting of DRAMs of different sizes and pinouts within limited footprint

Page mode memory access is enabled despite an immediately previous idle cycle. A row address strobe signal is maintained active during an idle cycle so that if a page hit is detected on a subsequent memory cycle, all that is needed to read or write to memory is a column address strobe signal which can be provided via a page mode access. In this manner, memory speed is enhanced because a conventional access is not required on the first memory cycle following one or more idle cycles.

Method for improving the page hit ratio of a page mode main memory system

A data processing system includes a high speed buffer, or cache, memory for temporarily storing recently executed instructions and a slower main memory in which is stored the system's operating program. Rather than sequentially accessing the cache memory to determine if the next instruction is stored therein and then accessing the main memory if the cache memory does not have the next instruction, system operating speed is increased by simultaneously accessing the cache and main memories. By accessing the main memory during its row address strobe (RAS) precharge time while simultaneously accessing the cache memory, the time necessary for the system's processor unit (PU) to read the next instruction from the main memory when not stored in the cache memory is substantially reduced.

Main memory access in a microprocessor system with a cache memory

A data processing system includes a plurality of memory access devices, each having a characteristic operating speed, for writing data into and reading data from a dynamic random access memory (DRAM) as well as a memory controller for accessing a plurality of addressable storage locations in the DRAM for either storing data in or reading stored data from the various storage locations in the DRAM. The system further includes a dynamic column address strobe (CAS) signal generator responsive to a memory access cycle signal, or READ pulse, and a conventional CAS signal for generating a dynamic CAS signal having a floating trailing edge which extends to the end of the memory access cycle as well as to the trailing edge of the READ pulse irrespective of the length of the memory access cycle signal to allow data to be read from the DRAM by any memory access device regardless of its operating speed during a memory access signal without losing or temporarily storing this data prior to providing it to the memory access device.

Data processing system with extended memory access

A power transition detection and override circuit is coupled to the OUTPUT-ENABLE (OE) pin of a nonvolatile electrically erasable programmable read only memory (EEPROM) for preventing inadvertent write commands thereto from a microprocessor arising from supply voltage (V.sub.cc) transitions, including power up and power down. With the EEPROM provided with internal write protection for V.sub.cc less than approximately 3.0 VDC, the power transition detection and override circuit holds the OE pin of the EEPROM low until V.sub.cc reaches a specified operating voltage, e.g., 4.75 VDC. This masks any WRITE ENABLE (WE) line glitches by the read operation of the active low OE line.

Growing community of inventors

Stevensville, MI, United States of America

Anthony M Olson