> On Thu, 15 Feb 2001, Alan Cox wrote: > > > Doesn't nvidia have a patent on this? I thought one of things they sued 3dfx > > > over was mapping a register over an address range, so burst transfers that are > > > designed to write blocks of memory can write to a single register efficiently. > > > > Well if they did them its laughable. The ancient ISA bus Pro Movie Studio has > > a fifo mapped over a memory range. This is an old old idea. I learned it in > > college from people who thought it was standard things you should know > > I did a search, and they do! http://www.delphion.com/details?&pn=US05687357__ > > What is claimed is: > > 1. A method for decoding addresses on a bus which provides a burst transfer > mode which may be utilized for rapidly transferring commands to sequential > addresses, the method including the steps of: > > addressing a sequence of commands intended for the same address to a > sequence of serial addresses, > > transferring the commands on the bus utilizing the burst transfer > mode, and > > decoding all of the serial sequence of commands to the same > destination. > > This sounds exactly like ignoring the low bits of an address so they all > decode to the same register. They just made it sound fancy. This sounds like a good case for http://www.bountyquest.com/ . It is one thing to get a patent these days, another to keep it. If you can show 'prior art' and there is a bounty out on the patented technology, then you may get rewarded for your evidence. -- Peter