The chief of server business for Advanced Micro Devices is looking forward rather optimistically, as he expects AMD’s forthcoming quad-core microprocessors to be significantly faster compared to arch-rival Intel’s offerings. Nevertheless, Randy Allen admits that Intel’s approach to create quad-core chips did make sense, as customers do not really care how the chip is made.

“We expect across a wide variety of workloads for Barcelona to outperform Clovertown by 40%. The quad-core chip also will outperform AMD’s current dual-core Opterons on ‘floating point’ math1ematical calculations by a factor of 3.6 at the same clock rate,” said Rick Allen, AMD’s corporate vice president for server and workstation products, in an interview with Cnet News.com web-site.

Quad-core processors clocked at moderate speeds and consisting of two dual-core dice on a single slice of substrate – Intel Xeon 5300-series as well as desktop Core 2 chips – allowed Intel Corp. to steal performance crown from Advanced Micro Devices, as such chips, thanks to the most recent “Core 2” micro-architecture by the company, provide decent performance in single-thread and dual-thread applications and lift-off in multi-thread apps.

AMD, however, believes, that a microprocessor with four cores with “monolithic” or “native” quad-core design with the x86 micro-architecture known as K8L provides better performance since processing engines do not have to use processor system bus for communication, but may use internal interconnection.

AMD agrees that clients are interested in performance achieved on quad-core processors, rather than their relative efficiency or the way they were made.

“Customers don’t care whether chips are monolithic or combine separate processors, but they do care about performance. We came to the conclusion that, given the capabilities and performance with the monolithic design, it was clearly the right answer,” Mr. Allen is reported to have said.

According to AMD, K8L includes a quad-core design for servers, workstations and high-end desktops, and a dual-core design intended for mainstream desktop markets. These next-generation processors will be built using AMD’s 65nm Silicon-on-Insulator (SOI) fabrication process, and include a broad range of functionality and micro-architectural improvements, including a new ability to dynamically alter the frequency of each core on the chip to match application workloads and thereby reduce overall power consumption.

“The magnitude of the transition is about halfway between the small tweaks AMD has made to Opteron over the years and the clean-sheet redesign Intel employed in moving from NetBurst to its current Core design,” Mr Allen is reported to have said.

AMD’s quad-core K8L processors for servers will have 128KB level-one cache (64KB for instructions and 64KB for data), 512KB level-two cache and 2MB level-three cache with the possibility to expand L3 cache. Due to higher amount of execution engines and micro-architectural improvements, clock-speeds of the initial K8L quad-core chips are projected to be lower than those of dual-core K8 processors.