<Intel has stated that the P7 is at least two years away (late 2000, early 2001). That's a mighty big window of opportunity. Maybe even big enough for AMD. Quite frankly, Intel screwed up bigtime.>
Man, why can't people get it? The P7 isn't going to be a K7-equivalent. If anything, K7 is more of the P6-equivalent, perhaps even surpassing the P6 clock-for-clock. Yet I haven't seen anything in the K7 core design that is vastly different from the P6 core. In other words, the K7 doesn't introduce anything new; it just features more of what made the P6 so powerful. Like I said, this is the "Ford Mustang" approach to CPU design.
The only window of opportunity for AMD is the short time between the K7 introduction and the introduction of Coppermine and Cascades, Intel's first 0.18-micron products. It is unlikely that AMD will have 0.18 in full swing before early 2000, and by then, there will be yet another small window of opportunity before Merced, Willamette, and Foster.
Tenchusatsu |
