In the first few years of the 21st century, Intel and AMD were locked into an arms-race of clock-speeds. Each was trying to one up the other with ridiculous clockspeeds till a point where they were just trying to flaunt the numbers, and the improvements didn’t really matter for most users.
In the last few years, they have been locked in a race to get more cores, even if there are very few applications that take advantage of those cores.
But now researchers at the University of California have done what these giants couldn’t: They have developed a 1000 core processor. However, the architecture is not 1000 cores on a chip, and is more like 1000 tiny chips linked together on a large chip. Yes, that means that each chip can be individually controlled, and while that does make it difficult to make programs that leverage this CPU’s power, it helps with power-efficiency. It is so effective that this mammoth 1000 core processor can actually be run on an AA battery, which is beyond impressive.
There aren’t many uses for this processor, since most users won’t need that computing power, and those that do would rather use GPUs, which are better suited for parallel computing. Meanwhile there are very few (none?) applications that can actually leverage this processor’s power.
What do you think?