Online Book Reader

Home Category

CompTIA A_ Certification All-In-One Exam Guide, Seventh Edition - Michael Meyers [80]

By Root 1183 0
of PGA that Intel calls organic land grid array (OLGA) (see Figure 5-50).

Figure 5-50 Intel Itanium 2 (photo courtesy of Intel)

Intel made a bold move with the Itanium and the Itanium 2 by not making them backward-compatible to 32-bit programming. In other words, every OS, every application, and every driver of every device has to be rewritten to work on the Itanium and Itanium 2. In theory, developers would create excellent new applications and devices that dump all of the old stuff (and problems) and thus would be more efficient and streamlined. If a company has a lot invested in 32-bit applications and can’t make the jump to 64-bit, Intel continues to offer the Pentium 4 or Pentium Xeon. If you need 64-bit, get an Itanium 2. AMD didn’t agree with Intel and made 64-bit processors that also ran 32-bit when needed. Intel would eventually follow AMD in this decision.

AMD Opteron

Coming in after the Itanium, AMD’s Opteron doesn’t try to take on the Itanium head to head. Instead, AMD presents the Opteron as the lower-end 64-bit CPU. But don’t let the moniker “lower-end” fool you. Although the Opteron borrowed heavily from the Athlon, it included an I/O data path known as HyperTransport. Think of Hyper-Transport as a very high speed link, providing direct connection to other parts of the PC—and to other CPUs for multiprocessing—at a blistering speed of over 6 GB per second! The Opteron comes in a micro-PGA package, looking remarkably like a Pentium 4 (Figure 5-51).

Unlike the Itanium, the Opteron runs both 32-bit and 64-bit code. AMD gives customers the choice to move slowly into 64-bit without purchasing new equipment. This was the crucial difference between AMD and Intel in the early days of 64-bit processing.

Intel and AMD pitch the Itanium 2 and Opteron CPUs at the server market. This means that as a CompTIA A+ tech, you won’t see them unless you go to work for a company that has massive computer needs. Newer CPUs from both companies fight for the desktop dollar.

Figure 5-51 AMD Opteron (photo courtesy of AMD)

Athlon 64

To place the Athlon 64 with the early generation CPUs is hardly fair. The Athlon 64 was the first for-the-desktop 64-bit processor, so in that aspect it is an early 64-bit CPU (Figure 5-52). AMD made two lines of Athlons: the “regular” Athlon 64 and the Athlon FX series. The FX series runs faster than the regular Athlon 64s, uses more wattage, and is marketed to power users who are willing to pay a premium. Underneath those two lines, AMD has almost twenty sub-lines of Athlon 64s in different codenames, making listing all of them here unwieldy.

Figure 5-52 Athlon 64

The AMD 64s have a number of enhancements beyond simply moving into the 64-bit world. The most fascinating is the inclusion of a memory controller into the CPU, eliminating the need for an external MCC and for all intents also eliminating the idea of the frontside bus. The RAM directly connects to the Athlon 64. AMD 64s support Intel’s SSE and SSE2 graphics extensions (later versions support SSE3).

While regular Athlon 64s use the same AMD PR numbers to describe CPUs, Athlon 64 FXs use a two-digit model number that’s just as cryptic as Intel’s current three-digit numbers.

AMD Sempron CPUs

AMD produces various Sempron CPUs for the low end of the market. Semprons come in two socket sizes and have less cache than the Athlon 64, but they offer a reasonable trade-off between price and performance.

Multicore CPUs


CPU clock speeds hit a practical limit of roughly 4 GHz around 2002–2003, motivating the CPU makers to find new ways to get more processing power for CPUs. Although Intel and AMD had different opinions about 64-bit CPUs, both decided at virtually the same time to combine two CPUs into a single chip, creating a dual-core architecture. Dual core isn’t just two CPUs on the same chip. A dual-core CPU has two execution units—two sets of pipelines—but the two sets of pipelines share caches (how they share caches differs between Intel and AMD) and RAM.

* * *

NOTE Putting two or more execution cores onto

Return Main Page Previous Page Next Page

®Online Book Reader