Semiconductor Engineering Editor in Chief Ed Sperling recently noted that getting data in and out of memory is just as important as optimizing the speed and efficiency of a processor.
“[Nevertheless], for years design teams managed to skirt the issue because it was quicker, easier and less expensive to boost processor clock frequencies with a brute-force approach,” he explained. “That worked well enough prior to 90nm, and adding more cores at lower clock speeds filled the gap starting at 65nm.”

