Thread View: comp.dsp
1 messages
1 total messages
Started by Chris J. Wein
Wed, 12 Dec 1990 00:45
Fast Matrix multiplication in hardware
Author: Chris J. Wein
Date: Wed, 12 Dec 1990 00:45
Date: Wed, 12 Dec 1990 00:45
23 lines
1507 bytes
1507 bytes
I am faced with the task of builing a video board to perform certain pixel transformations. The problem essentially boils down to performing one matrix multiplication per 100 ns (10 MHz) where the matrices are integer 3xN and Nx1 where N is in the range of 4 to 8. (ie 3 dot products to be done where the vector length is variable). The value of N must be software selectable and not hardwired into the design. Array processors have been suggested as a possible building block but I know very little about them much less any manufacturer/part number. Could someone help me out with the array processors or with any other info they might consider useful? Keep in mind that this design will end up being wire wrapped so keeping the chip count low is desirable. Mail to this address is preferred so not to waste bandwidth but if the results are interesting enough I will post a summary. Thanks! -- ============================================================================== Chris Wein | cjwein@watcgl.waterloo.edu Computer Graphics Lab, CS Dept. | cjwein@watcgl.uwaterloo.ca University of Waterloo | (519) 888-4548
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads