Single page Print

Motherboard performance, power consumption, and overclocking
When they use the same processor and core-logic chipset, motherboards tend to exhibit near-identical application performance. However, the DRAM tuning in a motherboard's BIOS can impact memory subsystem performance, which we've measured with bandwidth and latency tests.

The GA-H55N-USB3 does well but really no better than the competition here. That's what happens when you run these tests on a stack of boards with identical DIMMs and memory timings. The only reason the DH55TC lags behind the rest is because it wasn't able to run our DIMMs at 1333MHz. Instead, its memory is clocked at 1066MHz.

Power consumption is one variable that can be quite different from one motherboard model to the next. We measured system power consumption, sans monitor and speakers, at the wall outlet using a Watts Up Pro power meter. Readings were taken at idle and under a load consisting of a Cinebench 11.5 render alongside the rthdribl HDR lighting demo. We tested with Windows 7's High Performance and Balanced power plans.

Asus and Gigabyte ship their boards with energy-saving software that's supposed to lower power consumption without impeding performance. We've tested the boards from each manufacturer with and without this software installed. The EVO uses an EPU app that must be configured in "auto" mode to avoid performance-sapping clock throttling, while the Gigabyte boards use the company's Dynamic Energy Saver software.

Interesting. The Mini-ITX H55N has nearly identical power draw to the microATX H55M. If we focus on only the Mini-ITX offerings, the Zotac board trumps the Gigabyte at idle, while the reverse is true under load. Gigabyte's power regulation circuitry looks to be more efficient under heavier loads, which at least bodes well for gamers and overclockers. Lower idle power consumption would be preferred for home-theater PC applications.

Speaking of overclocking, I couldn't resist pushing the H55N beyond stock speeds. Dipping into serious CPU overclocking with a stock cooler inside a cramped enclosure wouldn't have been wise, but we can learn much about a motherboard's potential by pushing its base clock speed to the limit.

With our sample, that'd be an even 200MHz, which works out to a 50% increase over the board's default 133MHz base clock speed. The H55N effortlessly sailed up to this speed without the need for extra voltage or additional tweaking, although we did lower the CPU and memory multipliers to take those components out of the equation. We've achieved similar base clock speeds on enthusiast-oriented LGA1156 boards with larger microATX and full ATX form factors, but only hit a 180MHz base clock with Zotac's Mini-ITX model.

Had we not lowered the CPU multiplier, the H55N's 200MHz base clock would have been fast enough to push our Core i5-530 up to an impressive 4.4GHz. As is always the case with overclocking—and doubly true in smaller enclosures with limited cooling options—your mileage may vary.

Integrated peripherals enter the spotlight in the final component of our motherboard test suite. First up: those fancy USB 3.0 ports.

HD Tach USB 3.0 performance
Read burst
speed (MB/s)
Average read
speed (MB/s)
Average write
speed (MB/s)
CPU utilization
(%)
Asus P7H55D-M EVO 174.6 118.7 126.3 8
Gigabyte GA-H55M-USB3 152.5 119.1 123.7 7
Gigabyte GA-H55N-USB3 147.1 118.9 123.9 8

The GA-H55N-USB3 looks competitive, matching the sustained read and write speeds of the other SuperSpeed-equipped H55 boards. These scores were obtained with no graphics card in the system and the USB 3.0 controller hanging off the processor's full-speed PCIe lanes, so what happens when we pop our GeForce GTX 470 into the H55N and force the USB controller over to a half-speed lane linked to the PCH? Performance drops substantially. Burst rates fell to 104MB/s, sustained reads to 99MB/s, and sustained writes to only 75MB/s. That's still a lot faster than a gen-two USB port and more than quick enough for most external hard drives, but it's a disappointment nonetheless.

 
HD Tach USB 2.0 performance
Read burst
speed (MB/s)
Average read
speed (MB/s)
Average write
speed (MB/s)
CPU utilization
(%)
Asus P7H55D-M EVO 33.8 32.5 28.5 5
Gigabyte GA-H55M-USB3 34.2 31.1 21.4 8
Gigabyte GA-H55N-USB3 34.4 31.1 26.6 7
Intel DH55TC 33.4 29.5 20.4 7
Zotac H55 ITX 33.4 29.9 29.7 8

The H55N has USB 2.0 ports, too, and they're about as fast as those found on other H55 boards.

HD Tach Serial ATA performance
Read burst
speed (MB/s)
Average read
speed (MB/s)
Average write
speed (MB/s)
Random access time (ms) CPU utilization
(%)
Asus P7H55D-M EVO 231.2 110.5 110.7 8.4 4
Gigabyte GA-H55M-USB3 (Intel) 218.8 109.4 110.2 7.4 7
Gigabyte GA-H55M-USB3 (GSATA) 179.5 110.5 80.0 7.0 3
Gigabyte GA-H55N-USB3 211.4 109.0 110.0 7.3 8
Intel DH55TC 256.7 110.5 109.5 8.6 5
Zotac H55 ITX 212.7 108.1 109.9 7.3 9

The same holds true for the Serial ATA ports.

NTttcp Ethernet performance
Throughput (Mbps) CPU utilization (%)
Asus P7H55D-M EVO 935.2 10.3
Gigabyte GA-H55M-USB3 926.8 9.6
Gigabyte GA-H55M-USB3 933.0 8.8
Intel DH55TC 931.4 6.9
Zotac H55 ITX 927.7 6.3

I can't say for certain why Taiwanese motherboard makers favor discrete networking controllers over the GigE MACs built into the last several generations of Intel core-logic chipsets. The Realtek Gigabit Ethernet controller on the GA-H55N-USB3 works just fine, though, offering plenty of throughput with low CPU utilization.

RightMark Audio Analyzer audio quality
Frequency response Noise level Dynamic range THD THD + Noise IMD + Noise Stereo Crosstalk IMD at 10kHz Overall score
Asus P7H55D-M EVO 5 5 5 4 3 5 5 5 5
Gigabyte GA-H55M-USB3 5 5 5 4 3 5 5 5 5
Gigabyte GA-H55N-USB3 5 5 5 4 3 5 5 5 5
Intel DH55TC 5 4 4 5 3 5 4 5 4
Zotac H55 ITX 5 5 5 5 3 5 5 5 5

Unless you're willing to live with integrated graphics or rely on USB audio devices, you'll be stuck with the H55N's onboard audio. Fortunately, its analog signal quality is reasonably good, at least when compared to the competition. The H55N's score was unchanged when we ran this test again with the SG07 fully loaded, which is a good sign.