Intel DG35EC: G35 mainstream mATX board

We take a look at another G35-based socket-775 m-ATX board, this time directly from Intel. How does it fare against other small boards with integrated graphics?

September 11th, 2008 by Lawrence Lee

An Intel branded motherboard is the safe pick for a stable,
reliable system. Intel motherboards have been traditionally designed for business
and forego the flashy colors and heatsinks, ridiculous feature-sets, and
gimmicky software bundles so popular with retail DIY-oriented products. If it’s going into a PC at a corporate cubicle, it just needs to work, work well, and preferably
work forever.

The Intel DG35EC is a micro-ATX motherboard based on Intel’s G35 chipset and
features support for all current Intel processors. Though the G35 chipset is a year
old, it has been poorly represented in the market. It’s hard to say
why this is the case — though if the Asus
P5E-VM HDMI‘s price-tag is any indicator, high cost may be
a factor. Whatever the reasons, G35’s GMA X3500 graphics processor was (until
recently) the most advanced IGP money could buy for the LGA775 platform, and
until Intel’s new X4500 GMA is proven to be superior, it should not be ignored.

Our criteria for judging mATX boards are quite different than other sites.
As mATX boards typically have integrated graphics, video playback ability is
key when you take into account ever increasing multimedia and high definition content. We also consider
low power consumption to be vital, not only to lessen the toll on the environment,
but also to reduce heat in the system to ease
cooling a system quietly. Finally, fan control by the board is also
taken into account. While it’s not hard slow down and
mod fans to reduce noise, being able to do it with board-integrated controls makes life
that much easier.

PHYSICAL DETAILS

A board’s layout is important in several regards. The positioning of components
can dictate compatibility with other products (third party heatsinks mainly)
and also ease of installation. Poorly placed cable connectors can also disrupt
airflow and make the system more thermally challenging.

The board’s layout is slightly below par.
The 4-pin AUX12V power connector is near the back panel, meaning that cable will
have to be either snaked under the CPU heatsink or behind it, preferably as
tight to the board as possible to keep it out of any airflow paths. The northbridge
cooler is also very close to the PCI-E 16x slot. Many third party graphics card
coolers having mounting hardware on the “back side” of the card, which
may cause interference. The position of the onboard FireWire header between
the PCI and bottom PCI-E 1x slot is also not ideal.

Visually, the most impressive thing about the DG35EC are the heatsinks.
Both the northbridge and southbridge coolers are well-designed with a vast array
of fins to maximize the cooling surface area. They are attached with
wire spring clips to metal hooks on the PCB. The northbridge
heatsink is a bit taller than most, measuring 38mm high from the PCB surface.
From the photo above you can see that most of the capacitors are standard
variety except in the voltage regulation circuitry for the CPU and the memory, which are solid caps.

The back panel supports VGA and DVI outputs as well as FireWire and gigabit
ethernet. HDMI and S/PDIF are not available.

BIOS

BIOS options on mATX boards are typically restricted. The presence
of an IGP and the limited cooling associated with the form factor makes manufacturers
nervous about allowing users the ability to customize frequencies and voltages.
Intel is notorious for locking down their BIOS allowing only
basic options to be changed.

True to Intel’s reputation, the DG35EC’s BIOS has
only slightly more freedom than a detainee at Guantanamo Bay. Only the most
basic of memory options are available.

The Hardware Monitoring section is succinct and the Fan Control
menu is barren. And yes, these screenshots are the BIOS highlights.

TEST METHODOLOGY

Test Setup:

• Intel
  Core 2 Duo E6400 processor – 2.13Ghz, 1066FSB, 2MB L2 cache, 65nm,
  65W
• Intel
  DG35EC motherboard – 128MB allocated to integrated graphics
• Corsair
  XMS2 memory 1GB, DDR2-800
• Seagate Momentus 5400.2
  notebook hard drive – 120 GB, 5400RPM, SATA
• Asus
  BC-1205PT Blu Ray drive – SATA
• Seasonic
  SS-400ET mATX power supply
• Arctic Cooling Alpine
  7 Pro CPU cooler
• Microsoft
  Windows Vista SP1 operating system – Home Premium, 32-bit
• Intel
  15.8 graphics driver
• QuickTime
  Alternative 1.81 codec package

Measurement and Analysis Tools

• Cyberlink
  PowerDVD 7 to play video.
• Prime95
  25.6 processor stress software.
• ATITool
  0.27 Beta 4 artifact scanner to stress the integrated GPU.
• Seasonic
  Power Angel AC power meter, used to measure the power consumption
  of the system.

Our main test procedure is designed to determine the overall system power
consumption at various states (measured using a Seasonic Power Angel), and to
test the integrated graphics’ proficiency at playing back high definition videos.
Standard HD-DVD and Blu Ray discs can be encoded in three different codecs by
design: MPEG-2, H.264/AVC and VC-1. MPEG-2 has been around for a number of years
and is not demanding on modern system resources. H.264 and VC-1 encoded videos, with greater complexity in their compression schemes,
are more challenging and do not play smoothly (or at all) on slower PCs,
especially with antiquated video subsystems.

We use a variety of H.264/VC-1 clips encoded for playback on the PC. The clips
are played with PowerDVD 7 and a CPU usage graph is created by the Windows Task
Manger for analysis to determine the approximate mean and peak CPU usage. High
CPU usage is indicative of poor video decoding ability on the part of the integrated
graphics subsystem. If the video (and/or audio) skips or freezes, we conclude
the board’s IGP (in conjunction with the processor) is adequate to decompress
the clip properly.

SpeedStep was enabled the following features/services were disabled during
testing to prevent spikes in CPU/HDD usage that are typical of fresh Vista installations:

• User Access Control
• Microsoft Defender
• Windows Sidebar
• Indexing
• ReadyBoost
• Superfetch
• Windows Search
• Security Center
• Aero interface

Video Test Suite

1080p | 24fps | ~10mbps

H.264: Rush Hour 3 Trailer 1 is encoded in H.264 with Apple Quicktime.

1080p | 24fps | ~7.5mbps

WMV3: Coral Reef Adventure trailer is encoded in VC-1 using the WMV3 codec (commonly recognized by the moniker, “HD WMV”).

720p | 60fps | ~12mbps

WVC1: Microsoft Flight Simulator X trailer is encoded in VC-1. It is encoded using the Windows Media Video 9 Advanced Profile (aka WVC1) codec — a much more demanding implementation of VC-1.

 

1920×1080 | 24fps | ~19mbps

WVC1: Drag Race is a recording of a scene from network television re-encoded with TMPGEnc using the WVC1 codec. It is the most demanding clip in our test suite.

TEST RESULTS

Our test system is fairly basic, featuring a notebook hard drive and Blu-ray
drive. The CPU is a Core 2 Duo E6400, an older processor with modest power requirements
(65W). It is cooled by an Arctic Cooling Alpine 7 Pro CPU cooler connected to
a variable DC fan controller (so the fan’s power draw does not come into play).

Like the previous G35 board we tested, the Asus P5E-VM HDMI, the DG35EC blew through
our video playback test with flying colors. We experienced no problems during video playback,
and our toughest clip produced a mean CPU usage of only 51%, so the processor
performed this task with one hand tied behind its back. Power consumption was a
bit high, however, idling at 58W, and topping out at well past 100W
on full load.

Compared to the Asus P5E-VM HDMI and the Zotac
NF610i-ITX, the CPU load during high definition playback was almost identical.
The other boards were more energy efficient in every test state (except
the NF610i in sleep mode due to its lack of S3 standby capability). The NF610i-ITX
was the most efficient at low-level load, and the P5E-VM was the most efficient at high load.

The Asus P5E-VM does have a newer, more advanced, and presumably more efficient
southbridge than the Intel (ICH9R vs. ICH8) which may account for some of the difference, but
at heavy load, a 10W difference suggests poorer power regulation.

FAN CONTROL

To test the fan control system, we lowered the CPU cooling fan’s voltage to
5V, stressed the system with Prime95 and monitored the speeds of the fans connected
to the board. Noctua 80mm (1920 RPM at 12V) and Scythe 80mm (1600 RPM at 12V) fans were connected
to the chassis fan headers, while a Scythe 92mm PWM fan (2500 RPM at 12V) was connected
to the CPU fan header.

Throughout testing, the Noctua fan spun at 1140 RPM, while the Scythe 80mm stayed
constant at approximately 800 RPM — indicating that the chassis fan headers
are set to output about a flat 7V by default. The CPU fan’s speed was unreadable
(though it was spinning) until the CPU temperature reached 47°C, at which
point the fan speed reached 700 RPM. It then increased about 200 RPM for every
degree of temperature rise until it maxed out at 2500 RPM at approximately 56°C.

As Intel does not provide any monitoring software for the DG35EC, we used SpeedFan. System, and AUX temperatures can be ignored as they
were both way beyond reason, but CPU and Core 0/1 rose as the system was stressed.
Core 0/1 was consistantly 10°C higher than the CPU reading.

Speeds from all three fan headers are reported, though none
of them would give a reading if the fan connected was spinning below 650
RPM. The same is true in the BIOS’ hardware monitoring section. All three fan
headers can be controlled in SpeedFan by setting PWM modes 2 and 4 to “Manual
PWM Control” in the Advanced menu. Both chassis fan speeds are tied to
Speed04 and can be controlled fully from 0V to 12V. The CPU fan control was
almost absolute with SpeedFan able to lower our Scythe 92mm PWM fan’s speed
to a paltry 200 RPM. One of the advantages of having a PWM controller is the
ability to keep a fan spinning at very low speeds, making the fan’s starting
voltage irrelevant.

COOLING

To test the cooling on the board, we lowered the CPU cooling fan’s voltage
to 7V to reduce the amount of top-down airflow the nearby components received.
We then stressed the system with Prime95 and ATITool and whipped out our handy
IR thermometer to check the results. After about 20 minutes of load, the hottest
point on the northbridge heatsink read only 52°C and the southbridge 51°C
. Both could be touched for a few seconds before the heat became unbareable.
A couple of MOSFETs near the northbridge were over 75°C, but the rest of
them were in the 60°C range. Cooling is clearly not an issue for this board.

3D PERFORMANCE

As the DG35EC uses the same chipset and integrated graphics as the P5E-VM HDMI,
3DMark results were very similar.

As we have stated before, the GMA X3500 IGP gets trounced by the IGPs available on the AMD side. If you’re a gamer, discrete graphics card is a must-have if you’re using an Intel board with X3500 integrated graphics.

FINAL THOUGHTS

There has been hardly any alternatives to Intel integrated graphics chip for socket 775 boards. nVidia
is the only one with any alternatives, and their efforts are half-hearted
at best. Many of the nVidia-based socket 775 boards are semi-crippled with single channel memory controllers and
only two memory slots. For high definition playback, GMA X3500
performs similarly to nVIdia’s Geforce 7 series IGPs. Its 3D performance is poor in comparison with HD3200 in AMD’s G780 chipset. Still it’s better than Geforce 7, with its dated core architecture.

We expected the DG35EC’s BIOS to be locked down, but we did not anticipate
the poorer power consumption. Though based on the same chipset as the
Asus P5E-VM HDMI, the board’s power consumption was higher all around.
It is, by a small margin, the most power hungry LGA775 mATX board we’ve
tested, though to be fair, we’ve only tested a handful.

The board’s saving grace its fan control capabilities. The automatic fan control
system is simple, keeping the chassis fans at low speeds while ramping up the
CPU fan in response to CPU temperature linearly after a certain target temperature
is reached. However, it cannot be customized to suit particular
needs. A third party software like SpeedFan
is needed to control all three fan headers to your liking.

Overall, the Intel DG35EC is a classic Intel board more suited for
business rather than home use. We were impressed by its fan control ability and the effective cooling. While it does a decent job with high definition
playback, it lacks the outputs that would make it desirable for the multimedia center applications expected of home PCs these days — HDMI, S/PDIF. These omissions, higher power
consumption and lack of undervolting/underclocking options prevent us from giving
the DG335EC a clear-cut recommendation. Intel’s newest integrated chipsets,
G43 and G45, have been recently released. How much real-world improvement these
new chipsets provide is unknown, but if Intel’s promo is accurate, the G35’s sun may have already set.

Our thanks to Intel
for the DG35EC sample.

* * *

Articles of Related Interest
Intel
D945GCLF m-ITX: Atom For The Desktop
Zotac
NF610i-ITX: Compact, low cost, Core 2 solution
Gigabyte
GA-MA74GM-S2: AMD 690G, Take 2
Asus
P5E3 Premium: A Mean, Green Motherboard?
Asus
M3A78 Pro: An ATX 780G Motherboard
Asus
P5E-VM HDMI: A microATX C2D board for gamers?

* * *

Discuss this article in the SPCR forums.