quickstart THE BEGINNING OF THE MAGAZINE, WHERE ARTICLES ARE SMALL
I
nstalling a water-cooling kit on your
CPU is no big deal because there’s a
heat spreader on the core to prevent it
from being cracked during installation.
Videocards, on the other hand, have
exposed cores, so ratcheting a copper
block down onto your $600 GPU takes
courage—or at least it used to. The lat-
est trend in VGA cooling is cards that
have preinstalled water blocks, making
the process of adding your VGA card to a
pre-existing cooling circuit relatively risk-free. So far BFG, Gainward, and EVGA
have announced water-cooled video-
cards, and hopefully more vendors will
embrace the trends.Water-Cooled Videocards Splash
onto the Scene
Preinstalled water blocks take the risk out of extreme VGA cooling
A
ccording to a story published by TheInquirer.com, Intel is about to
change its Byzantine chip-numbering system to a new system that is
only slightly less convoluted. Purportedly, the new system is similar to Intel’s
current naming scheme; a combination of numbers and letters will identify
each part’s basic features. The fi rst symbol will be a letter representing the
CPU’s class (high-power, low-power, etc.), and the second
symbol will be a number identifying the family to
which the CPU belongs (workstation, desk-
top, mobile, etc.). There will be one more
letter or number, designating speed,
followed by two zeros that appar-
ently have no meaning at all.
Intel’s upcoming dual-core
notebook processor, code-
named Yonah, will be the
fi rst to use the system.
Under this scheme, a
standard Yonah would
be labeled T1x00, while
low-voltage parts will be
identifi ed as L1x00 and
ultra-low-voltage chips
will be dubbed U1x00.Intel to Change Chip-
Labeling System–Again
If what we hear is true, the new CPU names won’t be any clearer
Rumor Mill
free. So far BFG, Gainward, and EVGA
have announced water-cooled video-
cards, and hopefully more vendors willDECEMBER 2005 MA XIMUMPC 11
I
f two processor cores are better than one, can
we evaluate multicore chips simply by counting
cores? Unfortunately, we can’t. That would be as
misleading as judging single-core processors solely
by their clock speeds, and everybody knows how that
turned out.
All multicore processors are not created equal.
They have profound differences that greatly affect
their performance. Before the marketing schemers at
AMD and Intel corrupt your mind with propaganda,
consider some important aspects of multicore designs.
One factor is the microarchitecture of the cores
replicated on the chip. For now, AMD and Intel are
using processor cores originally designed for single-
core chips, because that’s all they’ve got. Multicore
PC processors are so new that neither company has
had time to create entirely new cores. I think Intel
has a slight edge, because its latest multicore pro-
cessor (Yonah, expected in January) has two cores
derived from the low-power Pentium M (Banias)
microarchitecture. AMD’s cores aren’t quite as
power-optimized.
Another factor is whether the multiple cores are
homogenous or heterogeneous. Homogenous means
all the cores are the same, and so far, that describes
all the multicore chips announced by AMD and
Intel. Someday we may see heterogeneous designs
with different cores optimized for different software
workloads.
Another very important factor is how the cores
communicate with each other. There are numerous
options. AMD’s first dual-core processor, the Opteron
800, has a sophisticated on-chip link called a cross-
bar switch. In comparison, Intel’s first dual-core
processor, the Pentium EE 840, looks like a duct-tape
job. Under pressure to match AMD’s multicore intro-
duction, Intel crammed two Prescott Pentium 4 cores
on the chip and routed their communications through
the external chipset. Intel’s Yonah is a much better
multicore design.
Cache matters, too. Because AMD and Intel
derived their first multicore processors from single-
core chips, each core has its own L2 cache. Like
jealous siblings, they don’t share. Yonah takes a step
forward by using a shared cache, which is more
efficient when one program needs more cache than
another does.
There are many more differences among multi-
core processors, and they always reflect engineer-
ing trade-offs—there’s no absolute best design.
Weighing those trade-offs will require better bench-
marks and knowledgeable reviewers.Tom Halfhill was formerly a senior editor for Byte magazine
and is now an analyst for Microprocessor Report.Sizing Up
Multicore
Processors
FAST FORWARD
TOM
HALFHILLGot an industry rumor? Send it to
[email protected]i.867.