This blog post references my cover story, 'Instigating a Platform Tug Of War: Graphics Vendors Hunger for CPU Suppliers' Turf' in the October 13, 2005 edition of EDN.

On June 10th, Nvidia invited a group of analysts and journalists, including myself, to attend a day-long tour of its Santa Clara, CA campus (subsequent to a day-long briefing on its soon-to-be-launched GeForce 6800 architecture). The first facility we visited was the massive server farm that the company uses, among other things, to compile and simulate its chip designs. Somehow (I don't remember the exact sequence of events) the topic of using the computational muscle of GPUs to supplement (or replace) servers came up in the discussion, and IT manager Ed Yee cryptically quipped, with a smile on his face, "Wouldn't it be fabulous for Nvidia chips to be designed on Nvidia chips?". Unfortunately, we weren't able to wring any more details out of Ed, and the rumoured public launch of Nvidia's GPUs-on-GPUs program at SIGGRAPH two months later didn't materialize. Something to keep an eye on....

Here's another journalist's take on the scene that day (scroll to the bottom of the writeup).