|
 |
|
HotHardware Test System Setup |
|
Geared for performance |
|
|
ASUS P4S800D-E Motherboard
DFI
LANParty Pro875 Motherboard
VIA PT880 Reference Motherboard
Radeon 9800 Pro
Pentium 4 3GHz
1 Gig
Kingston HyperX PC3500 DDR DRAM
Seagate Barracuda V
SATA 150 120 GB Hard Drive
LiteON 52X CD-RW Drive
Windows XP Pro SP-1
DirectX 9.0b
Catalyst 3.9 Drivers
SiS Chipset Drivers,
VIA Hyperion Drivers, Intel Chipset Drivers
|
|
Setup of the P4S800D-E
unfortunately, was not without incident. In fact,
our testing was held up for a few days, as we worked
though what was ultimately a compatibility issue between
the SiS SATA controllers on this board and our stock of
Serial ATA hard drives in the lab. At first we
attempted to use our Maxtor Diamond Max Plus 9 SATA
drives but couldn't get through the first few WinXP
install screens without a crash at the SATA driver
level. Asus sent us a new BIOS and a new set of
SiS SATA drivers but this didn't resolve our issue with
the Maxtor drives. We can't rule out that it may
have been a drive issue that we were experiencing with
these two 80G Maxtor models, since they were early
release units. Regardless, these were not
engineering release drives but production units from
Maxtor, so we were not sure where to point the finger
frankly, until we started working with our Seagate SATA
drives.
The Seagate Barracuda V 120G
SATA drives we ultimately ended up testing with, were
significantly more cooperative with the SiS SATA
controllers but not completely glitchless either.
While we could get one drive up on either the 964
Southbridge channel or the SiS180 controller channel, we
couldn't get the two drives to peacefully coexist on one
controller, in single drive mode. As such, our
hard disk testing was done with the OS drive on one
controller and another blank drive on the second SATA
controller. In an effort to bring you this article
in a timely fashion, we had to forego SATA RAID testing
all together. The SiS SATA drivers seemed to be
the culprit more than anything else, since more often
than not, our issues were manifest during installation
just as the SATA drivers were being installed. So
it seems, Asus and SiS have some work to do, in the area
of SATA drive compatibility, which interestingly enough
was something we also experienced with our recent VIA
PT880 reference motherboard testing.
|
 |
|
SiSoftware SANDRA Professional 4 |
|
Synthetics |
|
As a quick
baseline on performance, we put the P4S800D-E through a
series of tests with SiSoft's SANDRA
CPU @
3GHz.
 |
Multimedia @ 3GHz.
 |
Memory @ DDR 400
 |
SATA
150 HD Performance
 |
CPU @
3.5GHz.
 |
Memory @ DDR 470
 |
At default clock speeds, the
Asus P4S800D-E swept the SANDRA CPU and Memory tests across
the board by a small margin. This is an impressive
initial showing, since the competitive reference 3GHz P4
systems we've pulled from the SANDRA database here, are
driven by an Intel i875 Canterwood based platform. All
three areas, Integer, ALU (Floating Point) and Memory
performance, were slightly faster with the SiS 655TX driving
the benchmark.
Serial ATA 150 hard disk
performance was also fairly robust with our Seagate SATA
drives on this board but we'll also take data-points here
from PCMark 2004 and the Winstones to clarify things more.
 |
|
FutureMark's PCMark 2004 |
|
Desktop Performance Testing |
|
Here we're
introducing
FutureMark's PCMark 2004 benchmark into our testing
suite at HotHardware.com. We've taken three test
sub-routines from PCMark 2004, the CPU, Memory and Hard Disk
performance modules. Here's a brief description of
what is being tested, courtesy of FutureMark.
The CPU test
suite is a collection of tests that are run to isolate
the performance of the CPU. There are nine tests in
all. Two pairs of tests are run multithreaded each test in
the pair is run in its own thread. The remaining five tests
are run single threaded. These tests include such
functions as file encryption, decryption, compression and
decompression, grammar check, audio conversion, WMV
and DivX video compression.
The Memory
test suite is a collection of tests that isolate the
performance of the memory subsystem. The memory subsystem
consists of various devices on the PC. This includes the
main memory, the CPU internal cache (known as the L1 cache)
and the external cache (known as the L2 cache). As it is
difficult to find applications that only stress the memory,
we explicitly developed a set of tests geared for this
purpose. The tests are written in C++ and assembly. They
include: Reading data blocks from memory, Writing data
blocks to memory performing copy operations on data blocks,
random access to data items and latency testing.
The Hard Disk Drive test suite
is a collection of four tests that isolate the performance
of the hard disk. For these tests we use RankDisk, an
application developed and copyrighted by Intel. RankDisk is
used to record a trace of disk activity during usage of
typical applications. These traces can then be replayed to
measure to performance of disk operations for that usage.
The traces used for each test were created from real usage.
The following four input traces are used:
Windows XP Startup: This is the Windows XP start trace,
which contains disk activities occurring at operating
system start-up. This trace contains no user activity.
Application Loading: This is a trace containing disk
activities from loading various applications. It includes
opening and closing of the following applications:
- Microsoft Word
- Adobe Acrobat Reader 5
- Windows Media Player
- 3DMark 2001SE
- Leadtek Winfast DVD
- Mozilla Internet Browser
File Copying: This trace simply contains disk activities
from copying approximately 400 Megabytes of files.
General Hard Disk Drive Usage: This trace contains disk
activities from using several common applications.
Simply put the
SiS 655TX completely dominates the Processor and Memory
tests but falls squarely on its face in the Hard Disk Drive
tests. We're hopeful that with a driver or BIOS
revision or two down the road, that SATA HDD performance
will improve for this motherboard and chipset, since clearly
they're ready to rumble in general system bandwidth and
processing throughput.
Winstones and FutureMark 3DMark 03 |