The
BIOS is probably one of the top 10 reasons someone would build
their own PC instead of buying it off of an assembly line. The
assembly line PC’s are fixed; you get what you get, with
minimal ability to modify performance. ECS uses the Award BIOS.
Your
intro screen into the BIOS is something most of us are familiar
with, nothing out of the ordinary here. The same goes for the
“Standard CMOS Features”, “Advanced BIOS Features”,
“Integrated Peripherals” etc. The one we are looking
for in this particular BIOS is the “Advanced Chipset Features”;
here is where we will find the tweaks.
Once
in the submenu of the Advanced Chipset Features we can see this
is where all of the excitement happens. Let’s start with
“CPU Frequency”; we are somewhat limited here in
range from 200.0 to 209.5. Next in the list is “HT Frequency”,
which allows you to choose a multiplier from 1x to 5x as well
as Auto. The HT Frequency option configures the speed of the
HT link through the use of these multipliers. Although we can
not see what the “Auto” selection uses for a multiplier,
if we did select the maximum of 5x on a stock 200MHz, then the
HT link speed would be 1000MHz. The “HT Width” controls
the size of the upload and download transfer pipe available.
Everything I have read says to leave it at the default of 16
up and 16 down.
A
submenu is the DRAM configuration, this menu has memory parameters
that allow you to tweak your timings, and it also allows for “AUTO”
so if you don't feel it necessary, it will do the work for you.
Selecting Manual on Timing Mode allows you to manipulate the Memlock,
Tcl, Tras, Trcd and Trp settings. User Config Mode allows you
to change your 2T timing to 1T :).
Back
up to the previous menu, we finish off with the spread spectrum
choices, to Thermal-Throttle or not, as well as the CPU and
DIMM voltage controls. CPU is allowed to step up to 328mV (.328)
above the CPU selected voltage. Not a massive increase in VCore
to be sure, but enough to get a few extra MHz out of the CPU.
The memory can be driven to 3.10 VDIMM, a pretty decent offering,
only DFI have I seen higher.
Hidden
in the Power Management Setup screens is “Hammer Fid control”,
yes this is the CPU/FSB Ratio, remember, by default with AMD64
CPU's you can step down the multiplier to 4x all day long with
no modifications required, you just can't go higher then its stock
multiplier. Definitely something that would allow you to OC to
a system level performance, more so then a CPU performance. Why
it is hidden in here instead of included in the “Advanced
Chipset Features” is beyond me. I almost thought they had
forgotten it...
SLI
is performed automatically within the ECS motherboard, there are
no jumpers to set, no BIOS settings to manipulate. As much as
this is a nice thing to not have to worry about, with only one
ASUS nVidia 6600GT installed, I was only getting an x8 performance
out of slot 1 PCIe (as labeled on the motherboard). Not that any
board manufactured today could even fill up an x8 slot, with speeds
reaching 4000 MB/s (almost twice that of AGP 8x), I guess it is
the principle that if one card is plugged in, it should be an
x16 slot.
Test System
ECS
KN1 SLI Extreme Motherboard, 1GB (2x512 in Dual DDR Mode) Patriot,
PC3200 / PC4200, AMD Athlon 64 3200+ (Venice Core), ASUS Extreme
N660GT Silencer/HTD, Hitachi Deskstar 80GB, 8MB buffer, 7200 RPM,
SATA Drive, Windows XP SP2
Comparison
System
I
will be comparing the ECS KN1 to the previously reviewed DFI Ultra-D.
That systems Specs are: DFI Ultra-D Motherboard, 1GB (2x512 in
Dual DDR Mode) Patriot, PC3200 / PC4200, AMD Athlon 64 3200+ (Venice
Core), ASUS Extreme N5900, Hitachi Deskstar 80GB, 8MB buffer,
7200 RPM, SATA Drive, Windows XP SP2
This
should indeed be interesting as the only differences are the motherboard
and graphics card, and we are not testing the graphics card in
this scenario.
Testing
Time
for the testing phase, all tests are run 3 times and results are
then averaged (unless otherwise noted). VL’s testing suite
includes the following:
SiSoft
Sandra 2005 - Our standard synthetic benchmark suite,
updated to version 2005. While it doesn't provide real-world information,
it does give us a base for the rest of the tests.
Sysmark
2004 - Sysmark 2004 is BAPCo's latest revision of the
mainstream office productivity and Internet content creation benchmark
used to characterize the performance of the business client. It
uses a number of real-world applications and runs them through
a series of tests. We tested with the office and content creation
benchmarks.
PiFast
- We used a computation of 10000000 digits of Pi, Chudnovsky method,
1024 K FFT, and no disk memory. Note that lower scores are better,
and times are in seconds.
TMPGEnc
2.521 - We used an Animatrix file, titled The Second
Renaissance Part 1, and a WAV created from VirtualDub. The movie
was then converted it into a DVD compliant MPEG-2 file with a
bitrate of 5000. Times are in minutes, seconds, and lower is better.
Unreal
Tournament 2004 - run at 640x480 with minimal detail to test
CPU/Subsystem performance.
HD
Tach – Similar to SiSoft in that it does not necessarily
give us real world indication of performance but does allow for
baseline testing and efficiency reports of CPU utilization at
maximum hard drive transfer rates.
NEXT
