Laying The Foundation
It's easy to sit back and say "he was a good boy -- he took it on the chin like a man", or so to speak, as the good folks at Intel would like to say about their rivals at Advanced Micro Devices. Conversely, you might quickly discover the error of your ways, if you've counted AMD out. As we're sure you're aware, the lumps don't go unanswered in this high-stakes game we call the semiconductor industry. Retribution is often sharp and swift. Capiche?
Recently we spent some time in private conference with AMD down in the lovely city of New York and learned of a few more details with respect to their upcoming "4X4" platform offering and a two-chip, dual-core CPU bundle affectionately dubbed the "Quad Father". Though perhaps from a different Pacino era, might we offer, "say hello to my lil' friend".
During our session, AMD was quick to point out the inherent benefits of their "Direct Connect" architecture, which is brought to the system via serial HyperTransport links.
In the architecture Memory, CPUs and certain IO subsystems are all directly connected to each other via HT and it affords AMD a huge advantage in scalability. So much so that they're already speaking of octal core implementations. In addition, it's our early estimation that this is the single most important advantage that a quad-core AMD architecture has over Intel's Kentsfield CPU that is soon to be launched.
While Kentsfield in all likelihood will scale nicely in mutli-threaded applications and under heavy multi-tasking scenarios, the chip still shares a single front side bus, which unlike AMD's dedicated HT links for each dual core CPU, is a shared-bus architecture and potentially not as efficient. Time and benchmark data will tell the real story here of course.
AMD officials also spoke to their effort with 4x4 as a platform "inspired by enthusiasts". Multiple 4x4 processor bundles will be brought to market and although not officially confirmed, we've since learned that speeds ranging from 2.6GHz to 3GHz will be brought to the table.