Esos aparatos del demonio

• Cuotas de mercado de los ordenadores personales en los últimos 30 años


• Conceptos básicos de placas base


• Estrategia a la largo plazo de AMD


• El gusano que borra los ficheros el 3 de febrero


• La cuota de mercado de AMD llega al 20%


• ¿Va a ofrecer pronto Amazon también descargas de vídeo?

For the present moment, and indeed for the next five or six years, I think that we should use the term "multicore" to refer to ICs that contain multiple general-purpose processor cores, where each core is a peer (as opposed to one master and one or more slaves) [más arriba decía: should have some kind of on-die connection, e.g. a shared L2, a snoop bus, or some type of real die-level integration that truly distinguishes it from an MCM]. Note that I didn't say that the cores must be identical—only that they're capable of running general-purpose code (fixed-point scalar arithmetic at a minimum) and that they're peers. For all other arrangements, including heterogeneous chip multiprocessors, the term system on a chip (SoC) is perfectly appropriate. By this definition, the Cell would be an SoC and not a multicore chip, while the Xbox 360's Xenon processor would be a multicore chip.

here's a certain subset of pointer skills that students just need to know cold. Consider this code:


int buf1[10];
int *buf2 = new int[10];

for (int i = 0; i < 10; i++)
{
buf1[i] = i;
buf2[i] = i;
}


What's the difference between these two buffers? Where is each allocated? Why could you safely return buf2 from a function but not buf1? Any CS student should be able to answer these offhand.