|
TESTING METHODOLOGY:
We have seen quite a
variation in benchmark scores from one site to the
next, so we feel it is necessary to explain
exactly how this system was configured before
we ran any benchmarks. The first thing we
did was enter the system BIOS and set the board
to it's default settings. We
then set the CAS Latency to 2 in "nForce Turbo
Mode". The IGP was assigned 32MB of RAM and
the AGP clock was set to 100MHz. The hard drive
was then
connected and formatted, and Windows XP
Professional was installed. After XP was
completely installed, we hit the Windows Update
site and downloaded all of the available
updates.
We then installed all of the necessary drivers, disabled Windows Messenger, disabled
Auto-Updates and disabled System Restore.
Lastly we set the Visual Effects to "best
performance", installed the benchmarking software,
defragged the hard drive and ran the tests at the
CPU's default clockspeed.
 |
|
The Hot Hardware Test Systems |
|
Some Bad
Mamma-Jammas! |
|
|
HARDWARE USED:
NVIDIA nForce
Reference Motherboard
nForce Drivers
v22.90
GeForce 3
Ti500 (23.11 Drivers)
COMPARISON SYSTEM:
ABIT KR7A-RAID
GeForce 3
Ti500 (23.11 Drivers)
Sound Blaster
Audigy
3Com NIC
VIA 4-in-1's v4.37
COMMON HARDWARE:
AMD Athlon XP
1800+ (1533MHz.)
256MB Corsair
PC2400 (CAS 2) @ 133MHz.
IBM 7200RPM
30GB HD
Creative Labs
52X CD-Rom
Standard Floppy
Drive
Windows XP Pro
(With all current updates)
DirectX 8.1
|
 |
|
Performance Comparisons |
|
Time for
some numbers... |
|
|
The first benchmarks
we ran were using the very popular subsystem tests
that are part of SiSoftware's Sandra Benchmark
suite.
SiSOFT SANDRA:
CPU @ 1533MHZ.
M.M. @ 1533MHZ
In the CPU and
Multimedia tests, the nForce system performs on par with
similarly configured reference systems.
These test's rarely show any differences when the
same CPU is used.
MEM @ 1533MHZ
HARD DRIVE
The memory and hard
drive tests are a bit more interesting though. The
memory throughput scores are good, but they don't
come close to dominating the competition. It
seems like the Athlon XP can't utilize the
theoretical 4.2GB of bandwidth that is available.
This is similar to what you'd see when you couple DDR
RAM with a Pentium III. Yes, the performance
is slightly better, but the CPU's architecture
doesn't utilize all the bandwidth that is available.
The Hard Drive
performance is also slightly lower than we
expected. This particular IBM hard drive
consistently
performs in the 23-24000 ranges in this test on
other boards. The difference is minimal
though, and considering this is NVIDIA's first
attempt at writing IDE drivers, we won't complain
too much. There is probably some "headroom"
left in the drivers.
Next up we'll focus a
bit on the GeForce 2 MX core that is part of the
IGP. MadOnion's Video 2000 played a
double-role in this article.
VIDEO 2000 - Digital Video Decode and Playback
Performance:
In the complete Video
2000 test, the
nForce performed quite well. In a similarly
configured system, a GeForce 3 usually scores
between around 2100 Video Marks. Optimized
drivers and the 100MHz AGP clock speed helped bump up
the nForce's IGP to almost the 2300 mark.
We also ran Video
2000's MPEG encoding test to spotlight CPU
performance. We compared the performance to
a similarly configured system using VIA's KT266A
chipset.
VIDEO 2000 MPEG ENCODE:
The nForce just
slightly edges out the KT266A in the encoding
test. These scores are well within the
"margin of error" though. Neither chipset
holds a clear performance advantage here.
Another of MadOnion's
benchmarks is up next. 3D Mark 2001 is used
to test DirectX 8 capabilities and performance.
3D
MARK 2001:
The nForce IGP doesn't exactly
shine in this test, but it wasn't meant to. From a
gamer's standpoint, at 800X600 the IGP's performance was
decent. When compared to other integrated parts, the
performance is very good.
We also threw a GeForce 3 into the AGP slot and ran the default
benchmark again, and as you can see, performance shot way
up (obviously). As expected from NVIDIA, the AGP
implementation on the nForce is very good. If you're
looking for an nForce, but won't be using the IGP, you can
be sure your video card will operate properly.
Unfortunately, at 1600x1200, we were not able to
complete the benchmark using the IGP.
|
|
More Gaming
and the Stones...
 |
|