Argyris
Head-Fi's third most long-winded poster.
Rambus is a tremendously controversial topic. On the one hand, the industry did engage in price fixing in order to undercut them and push them out of the market. On the other hand, they have in many people's minds burned through any goodwill that might have generated for them with their subsequent patent trolling.
I try to stay out of Rambus discussions, since I still don't know what to think about them. Every time I get to hating them, I come across somebody passionately defending them who probably isn't employed by them, and I have to wonder then if maybe I'm missing something.
At any rate, back in the P4 days Rambus was extremely expensive, which is part of the reason the P4 faltered out of the gate. That and the fact that its long pipeline required higher clockspeeds than were feasible at the time in order to stretch its performance legs, the upshot being that the highest end PIII of the time could usually beat anything up to a 2.0 GHz P4 (unless the software was compiled with support for the then-new SSE2 instruction set, which wasn't present in the PIII).
I try to stay out of Rambus discussions, since I still don't know what to think about them. Every time I get to hating them, I come across somebody passionately defending them who probably isn't employed by them, and I have to wonder then if maybe I'm missing something.
At any rate, back in the P4 days Rambus was extremely expensive, which is part of the reason the P4 faltered out of the gate. That and the fact that its long pipeline required higher clockspeeds than were feasible at the time in order to stretch its performance legs, the upshot being that the highest end PIII of the time could usually beat anything up to a 2.0 GHz P4 (unless the software was compiled with support for the then-new SSE2 instruction set, which wasn't present in the PIII).