Intel's new 'Core' could gore AMD
Magical knob tweaks
IDF Intel today worked hard to convince anyone who would listen that AMD's performance advantage has come to an end. A new processor architecture stretching across its mobile, desktop and server lines will deliver better overall performance and better performance per watt than AMD's rival products. And this performance edge is coming "sooner than you think."
Intel has been on the defensive since 2004 due to a series of chip cancellations, delays and failures to advance its processor technology at the same rate as AMD. The chip giant, however, would never admit to such failings in public, insisting instead that all was fine.
This week's Intel Developer Forum in San Francisco cracked Intel's vow of silence as officials conceded that past products did not address power consumption issues as well as they should have. In addition, Intel confessed to falling behind AMD's Opteron on numerous benchmarks. (Intel has never discussed its miscues with 64-bit extensions and dual-core server processor delivery, but two out of four ain't bad.)
Intel plans to eradicate any failings with its new "Core" microarchitecture. This chip design builds on the devices created by Intel's mobile processor team in Israel. Intel will extend the performance per watt advantages seen in its mobile chips (Pentium M) to the desktop and server markets where AMD currently holds a clear lead.
For the desktop, Intel will ship a processor code-named Conroe in the third quarter that brings a 40 percent performance boost while reducing power consumption by 40 per cent when compared to the Intel Pentium D processor 950.
Server customers will see the new architecture appear with a chip called Woodcrest in the second half of this year. Woodcrest should deliver even more impressive gains, upping performance by 80 per cent while cutting power by 35 per cent when compared to a dual-core Xeon running at 2.8 GHz with 2x2MB of cache.
A little bit later, Intel will produce the Merom processor for notebooks with similar performance stats.
"Simply put, what you saw today is just a better platform," said Pat Gelnsinger, an EVP at Intel.
Basically, AMD appears to have about six months to hawk Opteron as hard as it can before Intel reaches performance parity or even stretches past AMD. Intel officials believe that Woodcrest will consistently outperform the new 2.6GHz Opteron from AMD and that similar scenarios will hold true on the desktop and with laptops.
A lot of attention had been paid here and elsewhere to AMD's use of an integrated memory controller and Hyptertransport to take a large performance lead over Intel. Instead of copying AMD, Intel has worked to tweak various knobs in its current designs to catch up with its rival.
Such advances include Wide Dynamic Execution that allows Intel chips to crunch through more instructions per clock cycle, improved power management tools, larger caches, faster front side buses, and Advanced Digital Media Boost, which allows 128-bit SSE, SSE2 and SSE instructions to be executed within one clock cycle.
Industry analysts believe that this collection of advances has put Intel back on the performance map.
"They have cranked a lot of knobs," said Kevin Krewell, editor in chief of the Microprocessor Report. "They have compensated for not having an on-chip memory controller.
"I think the new architecture will present a big challenge for Opteron."
Krewell noted that AMD will likely retain its edge with four-socket servers and above that make use of AMD's more scalable memory architecture. Intel, however, will put enormous pressure on AMD in the two-socket server market, which makes up the vast majority of system sales.
The only downside for Intel appears to be possibly higher costs for the future products.
"The Intel chipset is much more complex," Krewell said. "What they don't talk about is that the cost of the chipset will go up significantly."
Quite frankly, we're surprised to see Intel catch up to AMD so quickly. Lacking an existing server business, AMD had the luxury of moving forward with a brand new chip architecture. Meanwhile, Intel tried to wring as much life out of its NetBurst architecture as possible and seemed to think it could keep boosting GHz to remain competitive for a bit longer.
Intel's reliance on the massively successful NetBurst design allowed AMD to take a lead with 64-bit extensions, dual-core and performance per watt. It seemed AMD would keep this lead for the next couple of years, as Intel worked to mimic the integrated memory controller and Hypertransport links of AMD.
Intel still has much to prove given that the "Core" designs remain locked on PowerPoint slides. Benchmarks won't arrive en masse for several months.
In addition, Intel's failure to adjust quickly enough to AMD's Opteron allowed its rival a new place in the server market.
"Intel let a competitor sneak into a market where they weren't before," said Gordon Haff, an analyst at Illuminata. "Still, it was hard to believe that AMD would be able to maintain the kind of performance lead that they have enjoyed over the past one to two years."
The "Core" architecture will put AMD to the test in a massive, massive way. AMD cannot afford any miscues moving forward and really needs to show something spectacular and surprising at this point to counter Intel's product and marketing efforts.
"Intel has shown preliminary benchmarks that backup their general statements about performance," Haff said. "Based on what I have seen, there is not reason to think the products will not live up to their billing." ®