Terarecon, Inc.

William R Peet

James M Knittel

Hanspeter Pfister

Hugh C Lauer

Takahide Ohkami

Kenneth W Correll

Christopher J Kappler

Jan C Hardenbergh

Stephen R Burgess

Jay C Wilkinson

Stephen F Mason

Beverly J Schultz

A plurality of identical rendering pipelines are connected in parallel to read an array of voxels and to write an array of pixels. Each pipeline processes one voxel in one processing cycle of the pipelines. Each pipeline includes a plurality of serially connected different stages. The stages can include interpolation, classification, gradient estimation, illumination, and compositing stages. Interfaces connect identical stages in adjacent pipelines as one-way rings to communicate information associated with spatially adjacent voxels, and delay buffers connected parallel to particular stages communicate information associated with temporally adjacent voxels.

Volume rendering pipeline

A volume rendering integrated circuit includes a plurality of interconnected pipelines having stages operating in parallel. The stages of the pipelines are interconnected in a ring, with data being passed in only one direction around the ring. The volume integrated circuit also includes a render controller for controlling the flow of volume data to and from the pipelines and for controlling rendering operations of the pipelines. The integrated circuit may further include interfaces for coupling the integrated circuit to various storage devices and to a host computer.

Volume rendering integrated circuit

A two-level skewing architecture is imposed on the memory subsystem of a volume rendering system in which voxel data is stored in mini-blocks assigned to a set of DRAM memory modules, thereby permitting data transfer at the maximum burst rate of the DRAM memory and enabling real-time volume rendering. Within each DRAM module, mini-blocks are assigned to the memory banks so that consecutively accessed mini-blocks are assigned to different banks, thereby avoiding idle cycles during data transfer and increasing DRAM transfer efficiency to nearly 100%. In one embodiment, read-out of voxel data from banks of the DRAM memory proceeds from left to right unless there is a conflict of banks, in which case the read-out order is reversed. A specialized de-skewing network is provided to re-order the voxel data read out from DRAM memory so that the voxels can be processed in the order which they are arranged in the volume data set rather than the order in which they are stored in memory.

Two-level mini-block storage system for volume data sets

Growing community of inventors

Billerica, MA, United States of America

William R Peet