Let’s just face it. AMD processors still can’t catch up with Intel’s in terms of maximum performance and performance/watt ratio (with few exceptions) since the Core 2 Duo was introduced. The year 1999 was a historical year for AMD, as it could finally defeat Intel’s advantage in terms of performance with cheaper CPU when the Athlon crushed Pentium III in all benchmarks. This triggered an interesting performance race and AMD could hold on its own by outperforming Intel with its early Athlon family. But now, AMD seems to be hopelessly lagged. Of course, AMD’s Phenom II chips are by no means slow, as it can still compete with some of Intel’s Sandy Bridge processors. Things are even more disappointing on the notebooks front as AMD mobile processors (Turion) have significantly lower battery life than Intel mobile models.
AMD has successfully survived a few major upheavals in recent years. The acquisition of ATI could be a huge strategic advantage in the long run, but the move was challenging and exhausting to say the least. A few top executives left the company in the past few years and the wafer manufacturing process was transferred to the GlobalFoundries. Certainly AMD doesn’t have the resource to match Intel’s production, distribution and marketing capacity, but can AMD develop a hands-down better processor again? Well, if all the stars align just right, perhaps it might.
Intel has all it needs to maintain dominance in the processor industry, as it is at least one year ahead of its competitors in manufacturing process technology. AMD is just introducing its first 32nm processors, while Intel has been mass-producing them since early 2010. In addition, its first 22nm processors would be available in early 2012. Processors made using 22nm process technology will be cooler and smaller, however they require better manufacturing process as more transistors must be packed in each chip. Intel is a major technology company with some serious manufacturing chops. It is natural for the best micro-technology engineers to work for Intel and this has always been the case for many years, in addition, Intel has expanded its human resources by hiring software, video and graphics engineers.
Does AMD still have the chance? Perhaps, if:
- AMD can pick its goals wisely. It did a wonderful job with its low-cost Fusion processors, which is intended for the netbook market. For example, while its C-50 processor is slower then Intel’s Atom N550, the embedded graphic solution Radeon 6250 simply blows the Intel GMA 3150 out of the water. Another AMD’s netbook processor, the E-350, is not only faster than the latest Intel Atom processor, with the embedded Radeon 6310, it can support graphic intensive games, such as Call of Duty: Modern Warfare 2 and Crysis 2 (at low detail) as have been proven by many videos in the Internet, something that is unheard of previously in the netbook market. AMD understands how Atom disappoints many netbook users as the sluggish CPU and lackluster graphic solution limit Intel-powered netbook usability in some areas. As Intel is planning to make Atom more energy-efficient and smaller, so it can one day end up in smartphones and tablets industry, AMD are trying to make faster processors for sub-$400 netbook. It’s a prudent move, as AMD can avoid confronting Intel too much on its strongest area.
- AMD can focus more on its graphic advantage. Graphics and video performance starts to define the way we use a computer and the complexity of the Internet, require better graphic hardware. Intel has put some effort on improving its graphic solution with the HD Graphics 3000 that comes with Sandy Bridge CPUs, but it is still far from what AMD currently has. The upcoming Fusion A-series processors (code-named Llano) will be developed for mid-priced laptops and can push the already impressive integrated graphic solutions on the C- and E-series processors forward. AMD should be able to convince buyers that they can provide much better solution than the weak Intel’s integrated graphics at the same pricing level. AMD currently has all the resources to do just that.
- AMD doesn’t spend too much effort on the smartphone and tablet industry. Struggling to gain a market share in this field may require enormous engineering resources due to stiff competition. Qualcomm, Samsung, PowerVR, Nvidia, ARM and Intel are already well entrenched in this industry. AMD doesn’t have unlimited resources and its R&D efforts should be focused on making better processors for desktop, notebook and netbook. Sure, tablets and smartphones are growing industry; and may lure some people to abandon their netbooks or even notebooks, but traditional computers won’t go away anytime soon. AMD isn’t strong enough to wage a war in two fronts right now.
Will the A-series Fusion processors put AMD right back on the top? It is still uncertain. The architectural modification of Phenom-II cores, new power-saving cores, powerful integrated graphics and 32nm manufacturing process may make for an interesting combo. But unfortunately for AMD, Intel is just still too far ahead, and although Llano will offer better integrated graphic performance than Intel’s HD Graphics by a wide margin, it still can’t beat the performance level offered by Intel’s high-end processor models, such as Core i7 995X.
It’s important for AMD to stay relevant, for example by focusing its marketing effort on people who need the best “bang-for-the-buck”. Many people, especially on undeveloped countries, still refuse to buy a processor higher than $100. At this moment, Athlon II family offers the best performance level at under $100. It means, AMD can benefit from intensive marketing efforts on highly populated countries with many low income families. The company should be able to convince them that choosing Athlon II X3 or Athlon II X4 processors is a more sensible solution than Intel’s Core2Duo or Core i3 processors.
Another way that come in my mind is that AMD should focus on mobile processor but it seems already too late since Intel has join the mobile processor business.