DCAWD.com

AMD announced that the initial production of the world’s first 16-core x86 processor, codenamed “Interlagos” (Opteron Series) began in August, with shipments now underway to customers (OEM customers). The Opteron is compatible with existing AMD Opteron 6100 Series platforms and will be available to partner systems by Q4 this year. The Opteron processors are built on 32nm process technology, utilizing either 8, 12, or 16 cores (model dependent), quad-channel DDR3 memory controller, up to 16MB of L3 cache, and four HyperTransport links.

On August 31, a team of AMD-sponsored overclockers cranked a Bulldozer-based AMD FX processor up to an unearthly 8.429GHz, setting a new world record. This week, The Reg sat down with the leader of that processor-torturing team to find out exactly how they did it.

Model Cores Threads Frequency Turbo Frequency L2 cache L3 cache TDP Price
FX-Series FX-6100 6 6 3.3 GHz 3.9 GHz 6 MB 8 MB 95 Watt $188.32
FX-Series FX-8120 8 8 3.1 GHz 4 GHz 8 MB 8 MB 125 Watt $221.73
FX-Series FX-8150 8 8 3.6 GHz 4.2 GHz 8 MB 8 MB 125 Watt $266.28

From what he told us, he wouldn't be as happy working at Intel. In addition to the fact that he's encouraged to do "neat, risky things" at AMD, he told us that "It's fairly common knowledge that when we did this beyond–liquid nitrogen stuff with Phenom II, that other processors would cold-bug out. So that's one of the things that's cool and distinguished: that we can actually do these experiments with these CPUs."

More-recent Intel chips were beginning to exhibit fewer cold bugs, he told us. But he says Sandy Bridge does not scale particularly well with cold. "You'll see the top frequencies in Sandy Bridge in the fives – it does not have this remarkable ability to scale with cold as an attribute of its design."

AMD has finally delivered what it has promised for years, Bulldozer. It was four long years ago that we first talked to you about Bulldozer. What AMD brought to market, and what it talked about four years ago are two very different things. There is no denying that somewhere in the last four years the train fully left the tracks. I can't lie, I have thought more than once while going through all the data that AMD might have been better off concentrating on a successful die shrink and clock scaling of its current flagship desktop processors rather than spending the money on Bulldozer development.
...
AMD has delivered what will be a disappointment to many. The Intel fanboys have won this round. There just isn't any way around it. AMD fans, get ready to eat crow. If you expected something to outshine Sandy Bridge in terms of performance overall, it is just not there.
...
When you look at how much power the enthusiast's overclocked Bulldozer system will chew through under those multithreaded workloads, you have to realize just how much wattage you are giving up to not have an Intel processor. (Yes, we know some of you are brand loyalists.)
...
Is there any burning reason as to why you should buy AMD's $200 chip rather than Intel's $200 chip? I really can't give you one. The Bulldozer is harder to overclock for the most part and surely more picky about its voltages than Intel's Sandy Bridge. Bulldozer is going to take a heavier cooling solution than Sandy Bridge.

For the die-hard AMD fans that have been waiting for this day since the company first started hinting at Bulldozer, the performance exhibited by this first batch of FX series processors is probably somewhat puzzling. This was supposed to be the architecture that propelled AMD back into a strong, competitive position versus Intel’s desktop processors. Alas, that is obviously not the case. The FX-8150 is very competitive with Intel’s upper-mainstream Core i5 processors, but the Core i7 remains the ultimate performance champion. No if, ands, or buts about it.

This announcement will unveil the board’s and Rory Read’s new business strategy for AMD and reveal how the company will be competing in a quickly reshaping processor market. The content of that announcement will have to be at least as shocking as the clean sweep through its executive ranks that we have been witnessing for almost a year.
...
This may be very subjective, but I sense that AMD has, on some levels, a growing credibility issue. Sure, Fusion has had considerable success, AMD has captured market share from Intel in the mobile segment, and I am hearing very confident notes from AMD staff that Trinity is something to anticipate. However, if we are honest, AMD’s technology is way too average today on too many levels. What AMD needs is a true flagship product in its core CPU space – a technology that carries the industry mindshare for an entire company.

Intel has those flagship products, which may not bring in revenues on a product level, but they pay for themselves 1,000 times over in perception creation and marketing. Nvidia is building those products within its Tegra platform and has clearly stated its intentions to include in each product generation at least one feature that comes as a complete surprise to its rivals – such as the fifth power saving core in Kal-El. Such surprises, if executed correctly, can create product and market leadership.

AMD made an attempt to devise such a product with the most recent FX-8150 but largely failed to capture the perception of a flagship product. The pitch greatly represents a similar approach as the 2006 Quad FX, which consisted of two dual-core CPUs that created one quad-core system. It did not offer any performance advantages over a dual-core system and was extremely power hungry.

AMD will need an unquestioned flagship product to support its overall product credibility.

After the poor desktop performance, the possibility still existed that the Bulldozer architecture would start to make sense once we could see the server performance. Now the benchmarks have arrived, AMD's perseverance with Bulldozer is bordering on the incomprehensible. There's just no upside to the decisions AMD has made. All of which raises a question: why did AMD go this route? The company must have known about the weak single-threaded performance and the detrimental effect this would have in real-world applications long before the product actually shipped, so why stick with it? Perhaps AMD's anticipation of high clock speeds caused the company to stick with the design, and there's still a possibility that it might one day attain those clock speeds—but we've seen AMD's arch-competitor, Intel, make a similar gamble with the Pentium 4, and for Intel, it never really paid off.

Currently, the performance of AMD Bulldozer CPUs is slower than expected. This behavior occurs because the threading logic in Windows 7 and in Windows Server 2008 R2 is not optimized to use the Simultaneous Multithreading (SMT) scheduling feature. This feature was introduced in the Bulldozer family of AMD CPUs," Microsoft explained in a support article.