Tech Preview 2012: The PC's Bright Future

Posted 12/12/2011 at 2:31pm | by The Maximum PC Staff and Loyd Case

10

Comments

Print

AMD Bulldozer

AMD Finally Catches Up, Sort of

AMD’s newest CPU is perhaps the worst-kept secret in the industry. It seems like years ago that the company telegraphed the microarchitecture and garnered much attention. That’s no surprise, as the chip code-named Bulldozer is considered AMD’s first true CPU redesign since the original Athlon 64. Truth be told, there’s also a lot hanging on the new chip, as many are wondering if AMD still has the mojo to go toe-to-toe with Intel’s processors.

JUST WHAT IS A CORE?

The last year has seen a blurring of the lines regarding the definition of a core. Is it strictly x86? Do you count the integrated graphics portions? Adding to that Jack Daniel’s blurred-and-slurred line is AMD’s new Bulldozer. Officially named FX (in a throwback to the glory days of the Athlon 64 FX-51), the chip makes you wonder if what you thought was a core is still a core.

FX isn’t made up of cores, but rather modules. Each module is built using two monolithic “cores.” Each core has its own set of integer schedulers, pipelines, and L1 data cache. AMD says that compared to Intel’s Hyper-Threading, which splits up the resources of a single core into two virtual cores, FX’s design won’t get as bogged down when it has to deal with multithreaded workloads. On an Intel chip with Hyper-Threading, the core really only has resources for one core, and multithreaded loads must take turns running if the code calls on the same portion of the chip. That’s not the case with FX.

The FX is the first consumer eight-core CPU, and may even work in some AM3 boards with the proper BIOS.

But AMD didn’t completely duplicate all the resources of a dual core in its module—a single floating-point unit services both of the cores in integer workloads. Why just integer workloads? AMD says it believes that’s where most of the performance is to be made today.

AMD also says the modular design lends itself to higher performance when, say, a single-threaded workload is thrown at a single module. That’s because the cores are so interconnected that if only one core is working, some of the second core’s resources can be put toward that single-threaded workload.

AMD will launch four FX chips (two eight-core, one six-core, and a quad-core) ranging from $115 to $245. The company’s top-end part is the FX-8150, which is made up of four dual-core modules on a single die. One potential performance issue AMD has already admitted could crop up on Windows 7 and older OSes is the scheduler inefficiencies. The scheduler should know to throw four threads at four different modules, instead of four cores on two modules for the highest performance. Unfortunately, Windows 7 and anything older isn’t capable of determining how to load an FX for the utmost performance returns, AMD says. That may not happen until Windows 8 is released. Intel faced similar teething pains when Hyper-Threading was first released.

NEW MAX TURBO AND NEW INSTRUCTIONS

AMD first introduced Turbo modes with its hexa-core Thuban chips, aka Phenom II X6. With FX, the company has refined its Turbo even more with a new Max Turbo mode that, well, maxes out the overclock. On workloads that hit all cores, each one can be overclocked by 300MHz, to 3.9GHz. On lightly threaded workloads, half of the cores can go to sleep while the other half can clock up to 4.2GHz.

Elsewhere in the chip, AMD has brought the new CPU to instruction parity with Intel. The FX processors will have AES instructions to support acceleration of encryption and decryption workloads, and Advanced Vector Extensions (AVX), which Intel introduced with its Sandy Bridge lineup, is now also present. The old instruction-set wars still run hot, though, as the FX will support the Fused Multiply Add 4, or FMA4, instruction set. Intel, unfortunately, is only supporting FMA3 in its upcoming Ivy Bridge CPUs and has apparently canceled plans to support FMA4. This little standoff could cause problems with developers as to which instruction set they support and how they support it. Who is at fault? Most observers say both companies are playing games. Our standard guidance is to not sweat over new instruction sets, because by the time software support is there, the first chips to support it are usually so old that it’s just easier to upgrade instead.

NEW CHIPS, SAME OLD SOCKET

There’s one thing AMD gets right year after year—the same old socket. While Intel has shuffled through five sockets, AMD has pretty much stuck with just one. The only change has been to the electrical underpinnings in the AM3+ spec, which could render some AM3 boards incompatible. Still, it’s expected that the AM3+ FX chips will drop into most late-model AM3 boards without incident (check with your motherboard vendor first, of course) and even better, the mounts for the coolers have remained the same, so you can reuse your exotic cooler.

In a nod to enthusiasts, AMD says all FX chips will be fully unlocked, giving overclockers an all-access backstage pass to faster speeds. With Intel’s chips, only the K versions and Extreme Editions are fully unlocked. AMD has made overclocking a bragging point, too, and helped fund a team of overclockers to push an FX to 8.429GHz using liquid helium.

So where does Bulldozer stand? On paper, it looks like AMD has finally caught up—to some of Intel’s Sandy Bridge chips, anyway. But what does it look like in benchmarks? For the answer to that, you’ll have to read on.

Trinity and Beyond

AMD continues down the APU path

AMD’s E-350, aka Zacate, was the sleeper hit of the year and completely sold out several times. The company’s second Fusion part, Llano, is also forecast to sell well, with some reports saying the chip will equal 40 percent of AMD’s total sales.

The message? While AMD has been unable to outdo Intel in the performance category for some time, its APU approach to the mainstream seems to be gaining traction. Next year, AMD will introduce a new CPU‑cum‑GPU, code-named Trinity, which it hopes will continue that trend.

The top-end Trinity will use a derivative of the FX core modules to create an eight-core chip with a new DirectX 11 GPU. Like FX, the chip will be fabbed on a 32nm process by Global Foundries. What’s unfortunate is that AMD may pull an Intel and switch to a new socket called FM2, and it’s not clear whether Trinity will be compatible with existing FM1 motherboards.

Despite the excitement over Trinity, AMD won’t be giving up on the high end. The company has a very ambitious plan to increase the performance of FX chips over the next few years. In 2012, the company expects to release Piledriver. The year after that we’ll see Steamroller, and the year after that Excavator.

Each iteration is expected to bring at least a 10–15 percent performance increase, the company says. AMD won’t say what the microarchitecture changes are exactly, but that most of the performance should come from clock-speed bumps and other changes under the hood.