Bij HardwareCentral is een stukje tekst verschenen waarin wordt uitgelegd wat pipelining is. Pipelining wordt toegepast bij alle modere CPU's om de kloksnelheid te kunnen verhogen, en tegelijkertijd de clock voor clock performance niet al te ver naar beneden te laten gaan. Voor de die hard tweaker zal het artikel weinig nieuws bevatten, maar voor de 'newbie' is het zeker een leuk stukje leesvoer:
Intel’s Willamette, now called the Pentium 4, is currently one of the most hotly contested PC products in the world. Moreover, its success is of extreme importance to Intel. If Pentium 4 succeeds as a performance leader, it could mark Intel’s return to the top, and make AMD’s Athlon a much less appetizing option than it now seems. Likewise, if Pentium 4 fails either in terms of performance or production, it will only underscore a year almost completely filled with blunders from Intel. So you can see that Willamette’s success is of extreme importance to both AMD and Intel.
[...] There are a lot of technical terms like Trace Cache, Load-Latency, and Pipelining being thrown around, and they may not make sense to a lot of readers. Today’s feature will explore and explain one of those terms: Pipelining. If you’ve ever come across the term ‘pipelining’ in a discussion regarding Intel’s Willamette--or any other processor for that matter--and wondered what exactly it means, then this article is for you. We’ll start from the ground up, and build a general overview of what pipelining is, and why it’s used.
As a side note, this is designed to be a beginning-level tutorial on pipelining and in CPU architecture in general. If you already have a sound grasp on the concepts, or if you’re an engineer at Intel already, there’s probably not much new here.
Onze dank gaat uit naar houte@dolfijn.nl" rel="external">yamha voor deze tip!