We already know Intel is planning to build a better GPU for Ivy Bridge than what's currently available in Sandy Bridge, but just how much better it will be is the question. Intel provided a partial answer at this year's IDF event by detailing parts of the next-generation GPU, which the Santa Clara chip maker says will support up to a 4Kx4K (Quad HD) screen resolution.

One of the many awesome things coming out of this year's Intel Developer Forum (IDF) is a new DRAM concept Intel claims will deliver a 7-fold improvement in energy-efficiency over today's DDR3 modules. It's called Hybrid Memory Cube and Intel is working closely with Micron to turn this concept into a shipping product. So what exactly is a Hybrid Memory Cube?

Things are about to heat up in a big way in the handheld mobile space, a sector that's currently dominated by ARM. Intel has long said it plans to push its platforms into smartphones and tablets, and the Santa Clara chip maker took a gigantic step towards that goal by getting Google to agree to optimize future versions of Android for Atom processors. Should ARM be worried?

AMD, apparently eager to get its Bulldozer core out the door, might be shipping its next generation architecture sooner than we thought. Mike Magee over at TechEye.net says he recently met up with John Fruehe, who said that AMD plans to start sampling Bulldozer cores in the fourth quarter of this year.

Fruehe went on to say that Bulldozer, which has been built from the ground up, will continue to use the same sockets and the same power envelopes to alleviate potential headaches and ease the process of migrating to the newer platforms.

AMD's Bulldozer architecture marks the first major redesign of the company's processor line since the launch of the original Athlon 64 chips way back in 2003. Every CPU since then has basically been a tweaked version rather than a brand new architecture. Bulldozer is different.

"There will be enhancements to our memory controllers, things we cannot talk about just yet, that we expect to help reduce the time to access memory, both locally and remotely," Fruehe said in a blog post.