On Fri, Jan 24, 2003 at 10:46:17AM +0100, Gabriel Paubert wrote: > On Thu, Jan 23, 2003 at 08:12:24PM +0100, Sven Luther wrote: > > > Any idea if the PPC970 bus is really hypertransport like so many people > > sy, or something else ? > > I don't believe that it is HT like. From what I understand, the chip > has two separate unidirectional 32 bit busses that will be connected > to the chipset. They run at half the core clock and from what it seems > they both transfer data, addresses and control signals. > > Basically it seems to be very similar to a packed based protocol: to read > a cache line the processor sends a request on the outgoing bus and receives > a reply on the incoming bus after some time. But IBM has not released many > details yet.
Thought so also from the information that came out, but since so many web site were speaking about HT, i suppose they are just misinformed. I think the HT rumor sprang from the fact that Apple is a HT member. Friendly, Sven Luther
