Abstract: A circuit comprising a programmable routing network, a logic array configured to generate a plurality of product terms in response to one or more of a plurality of input signals from said programmable routing network, a plurality of look-up tables each configured to receive a logical combination of at least two of said product terms and a plurality of macrocells each configured to generate an output in response to one or more of said look-up tables.

Abstract: A programmable device architecture that may improve functionality over look-up table based or product-term based programmable logic devices and that may provide for the efficient implementation of user-programmable logic designs resulting in implementations that may require less area and may provide increased performance. A product-term array (either fully or partially populated) may be placed in front of a number of LUT-based macrocells, utilizing the available routing wires as wordlines to form the product terms. The present invention takes advantage of existing routing to do more than just route signals from one point to another by allowing logic to be implemented in the same die area. The result is logic implementations that may require fewer total macrocells, fewer levels of macrocells, and fewer point-to-point nets (because logic density increases). The present invention may apply to FPGAs comprising an array of macrocells and to FPGAs comprising an array of clustered macrocells.

Abstract: A programmable device architecture that may improve functionality over look-up table based or product-term based programmable logic devices and that may provide for the efficient implementation of user-programmable logic designs resulting in implementations that may require less area and may provide increased performance. A product-term array (either fully or partially populated) may be placed in front of a number of LUT-based macrocells, utilizing the available routing wires as wordlines to form the product terms. The present invention takes advantage of existing routing to do more than just route signals from one point to another by allowing logic to be implemented in the same die area. The result is logic implementations that may require fewer total macrocells, fewer levels of macrocells, and fewer point-to-point nets (because logic density increases). The present invention may apply to FPGAs comprising an array of macrocells and to FPGAs comprising an array of clustered macrocells.

Abstract: A product-term array that may allow for the implementation of product terms requiring less silicon area than conventional designs. The product terms may also have a shorter propagation delay when compared with conventional designs. A multiplexer, which may be programmed with a configuration bit or signal, may select the polarity of an input signal to the product-term array. Duplicating a number of the initial inputs to the array may accommodate particular design constraints that may require both polarities (i.e., both positive and negative) of a given signal or set of signals. Even with the duplication of certain inputs, the total number of product-term inputs to the array will generally be reduced when compared with conventional designs, that duplicate the polarity of every input internally to the array.