Gigabyte’s GA-P55M-UD4 and MSI’s P55M-GD45 motherboards compared

Just about every system I’ve built in the last decade has used a full-sized ATX motherboard. Back in the day, you pretty much had to go ATX to get a decent board with the sort of features, BIOS options, and overclocking headroom expected by savvy enthusiasts. MicroATX offerings were largely budget fare, providing little more than the bare minimum required for basic desktops. Plus, they didn’t have enough board real estate for dual CPU sockets. After embracing the creamy smoothness of SMP early on, I wasn’t about to go back to a single core.

These days, you’d be hard-pressed to find a single-core desktop processor for sale. Dual- and quad-core CPUs have become commonplace, and they’ll easily drop into the single sockets available on MicroATX motherboards. Today’s micro mobos are much improved over their ancestors, too. Some models offer the same features as enthusiast-oriented ATX boards, including loads of integrated peripherals, identical BIOS overclocking and tweaking options, and even support for CrossFire and SLI. Plus, you can get versions with LGA1156 sockets primed for Intel’s latest Lynnfield-powered Core i5 and i7 processors.

For the first time, well, ever, I’m actually contemplating a MicroATX build for my next desktop. But should you? We’ve rounded up a couple of MicroATX P55 motherboards from Gigabyte and MSI to help answer that question, and you might be surprised at what we’ve found.

Gigabyte’s GA-P55M-UD4 motherboard
Micro without compromise

Manufacturer Gigabyte
Model GA-P55M-UD4
Price (Street)
Availability Now

Gigabyte’s stable of P55 motherboards currently houses two MicroATX models: the GA-P55M-UD2 and the GA-P55M-UD4. The former is remarkably affordable at just $110 online, but as one might expect from a budget variant, it’s not exactly stacked with the latest and greatest. You’ll need to shell out nearly $35 more for the UD4. In return, you’ll get a board that, at least on the surface, appears to be completely free of compromise. It’s almost as if Gigabyte took one of its enthusiast-oriented ATX boards and chopped a couple of inches off the bottom.

The UD4 even looks the part, with the same turquoise, blue, and white color scheme as the rest of Gigabyte’s recent motherboards. I like the simple palette, particularly since it maintains the company’s trademark turquoisey-blue board color. However, the mismatched “racing stripes” on the heatsinks look a little odd, as does the single orange expansion slot. Old habits die hard, I guess.

Like other members of Gigabyte’s P55 family, the UD4 has a two-ounce copper layer and fancy electrical components, including solid-state capacitors, ferrite-core chokes, and lower RDS(on) MOSFETs. The board itself is four-layer design, just like the ATX-sized GA-P55-UD4P.

Interestingly, the UD4 actually has more power phases dedicated to the processor than its ATX cousin. Gigabyte packs 12 phases around the UD4’s socket, while just eight ring the UD4P’s LGA1156 Land Grid Array.

LGA1156 sockets have come under increasing scrutiny recently, with some suggesting that Foxconn-built units don’t make adequate pin contact with Lynnfield CPUs. The UD4’s socket is indeed made by Foxconn, as is the one on the MSI board. However, I didn’t experience any problems with either. Of course, I also didn’t engage in the sort of extreme, liquid-nitrogen-fueled overclocking that’s apparently scorching the offending sockets.

One might think that the six Serial ATA RAID ports available in the P55 Express chipset would be sufficient for a MicroATX motherboard like the UD4. Or not. Gigabyte adds a GSATA-branded JMicron storage controller that provides two more SATA ports and an IDE connector. Heck, there’s even a floppy port, if you’re stuck in the dark ages.

The edge-mounted SATA ports do a good job of avoiding clearance conflicts with longer graphics cards, but you might run into problems with tighter enclosures that leave little room around the motherboard tray. Serial ATA cables with right-angle ports might not help on that front, either. All the ones I have are oriented so that the cable would end up pointing down, toward the motherboard tray, if plugged into the UD4’s edge-mounted ports.

Gigabyte squeezes four expansion slots onto the UD4, including a pair of physical x16 slots that can be configured in a dual-x8 setup for SLI or CrossFire. The x4 slot is notched to accommodate longer cards, too, making it possible to run three PCI Express graphic cards, provided they all have single-slot coolers.

Having four expansion slots is still a little limiting when you’re looking at losing two to a double-wide graphics card, but then the UD4’s payload of integrated peripherals doesn’t need much embellishment. The board employs Realtek’s ALC889A audio codec—our current favorite for its ability to encode Dolby Digital Live bitstreams on the fly—and offers two flavors of S/PDIF output alongside the usual assortment of analog audio jacks. Gigabyte also includes a hybrid eSATA/USB port capable of powering compatible external Serial ATA devices without the need for auxiliary cables, which is all kinds of awesome.

If you’re familiar with the BIOSes on Gigabyte’s high-end motherboards, you’ll feel right at home with the UD4. The layout and tuning options are all but identical to what’s available on more exotic ATX models, offering enthusiasts plenty of tweaking and overclocking potential. However, Gigabyte continues to have some of the most barren automatic fan speed controls in the business. Users can identify whether they’re using a three- or four-pin CPU fan, but that’s about it. Such a simplistic toggle switch is a stark contrast to the overabundance of fine-grained clock speed, memory timing, and voltage controls available elsewhere in the BIOS. I’ve been harping on this issue for a while now, and have brought it up with Gigabyte on numerous occasions. Unfortunately, the company’s BIOS division seems entirely uninterested in developing more extensive fan speed options.

MSI’s P55M-GD45 motherboard
Black and blue and, er, MicroATX

Manufacturer MSI
Model P55M-GD45
Price (Street)
Availability Now

Unlike Gigabyte, MSI has just one MicroATX motherboard in its P55 lineup: the P55M-GD45.
At close to $125 online, the GD45 costs about $20 less than the UD4. So what do you lose? SLI, for one. Although AMD supports CrossFire on just about every P55-based motherboard with a pair of physical x16 slots, Nvidia requires that individual models be SLI-certified. The GD45 will happily accommodate a couple of GeForce graphics cards as long as one is dedicated to PhysX processing, but you can’t team the pair to share the pixel-processing workload.

Of course, given the performance of AMD’s latest Radeons, SLI support is something most folks can probably do without. You can build a heck of a gaming system with just a single graphics card and still be confident that the latest titles will run fluidly at resolutions up to 2560×1600.

At first glance, the GD45 looks very much like the sort of board you might put inside a fancy gaming rig. The black-and-blue color scheme has a menacing edge, and I quite like the pewter tone of the chipset and voltage circuitry heatsinks.

The VRM cooler is tiny compared to what we’re used to seeing on enthusiast-oriented motherboards, but I’ve often wondered whether massive clumps of copper linked by intricate heatpipe arrays were really necessary. Granted, the GD45 has only four power phases supplying the processor—one third the number available on the Gigabyte board. MSI claims that just one of its so-called DrMOS MOSFETs is better than four “traditional” ones, although Asus and Gigabyte both argue that their boards use higher-quality MOSFETs than what would presumably be classified as “traditional” designs. Comparing CPU power implementations from different manufacturers is more complicated than counting phases, but it is worth noting that the GD45 has two fewer phases than MSI’s mid-range P55-GD65 ATX model.

In addition to purportedly higher-quality MOSFETs, the GD45 uses solid-state capacitors throughout. The board has four layers, just like the UD4, but its copper layer contains just a single ounce of the conductive stuff. One-ounce copper layers are still the industry standard. Although Asus and Gigabyte claim that two-ounce layers lower board-level impedance, we’ve yet to see that translate into tangible benefits for end users.

MSI eschews auxiliary storage controllers for the GD45’s internal SATA connectivity, instead relying on the six 3Gbps ports provided by the P55 chipset. As on the Gigabyte board, the edge-mounted ports leave plenty of clearance for longer graphics cards, but they could present problems in tighter enclosures that put hard drive cages or other obstructions right next to the motherboard tray. In fact, the GD45’s SATA ports are even closer to the edge than on the UD4.

Speaking of clearance issues, I should note that the DIMM slot tabs on both the UD4 and GD45 snug right up next to longer PCI Express graphics cards. This placement can make swapping memory modules difficult when a graphics card is installed, although that’s not the sort of thing most folks would do with any regularity. A little layout crowding is inescapable given the dimensional constraints presented by the MicroATX standard.

The GD45’s lack of SLI certification probably stems from the arrangement of electrical lanes connected to its physical x16 slots. All 16 of the PCI Express 2.0 lanes found in Lynnfield CPUs flow to the first slot, while the second must make do with four lanes of connectivity from the P55 chipset. Worse yet, the P55’s PCIe lanes only signal at 2.5 GT/s, which is the same speed as a first-generation PCI Express. Effectively, you’re looking at a x16/x2 configuration for dual graphics cards. That config doesn’t make the cut for SLI certification, and it may not be ideal for CrossFire, either.

A whopping ten USB ports adorn the GD45’s port cluster. Unfortunately, none of them are cross-bred with the board’s dual eSATA ports. I’d happily trade both non-powered eSATA ports for just one hybrid eSATA/USB connector.

Although the port cluster looks otherwise stacked, a digital S/PDIF audio output is conspicuously missing. There’s an S/PDIF header on the board, and the manual makes mention of an optional PCI back plate connector with an RCA output, but that part doesn’t seem to be included with any of the GD45s currently available at online retailers.

Realtek’s ALC889 codec handles the silicon side of the audio equation, but unlike the ALC889A on the Gigabyte board, real-time Dolby Digital Live encoding isn’t supported. This limitation should only affect gamers looking to output multi-channel audio over a digital connection, since the ALC889 is still capable of passing pre-encoded audio tracks, such as those found with movies, over S/PDIF.

Although the MSI and Gigabyte BIOSes differ a little in their presentation and the specific ranges and granularity available with various overclocking and tweaking options, enthusiasts engaged in the usual performance tuning should find them functionally equivalent. Fan speed controls are the obvious exception, and MSI deserves props for allowing users not only to set a target CPU temperature between 40 and 70°C in 5° increments, but also to choose a minimum fan speed between 0 and 87.5% in 12.5% increments. Even though temperature-based fan speed control isn’t available for the board’s two system fan headers, each can be set to run at 50, 75, or 100% of full speed.

The devil’s in the details

This wouldn’t be TR motherboard coverage without a painstakingly detailed assessment of each board’s BIOS options and specifications. These details don’t exactly lend themselves to eloquent prose, but you should be able to find what you need in the tables below. Note that BIOS variables preceded by a * can be keyed in directly rather than selected from what can sometimes be quite a long list of available values.

Gigabyte GA-P55M-UD4 MSI P55M-GD45
Clock speeds *Base: 100-1200MHz in 1MHz increments
PCIe: 90-150MHz in 1MHz increments
*Base: 100-600MHz in 1MHz increments

*PCIe: 90-150MHz in 1MHz increments
Multipliers
*CPU: 9X-20X in 1X increments

DRAM 6, 8, 10X

QPI: X23, X36

CPU: 9X-20X in 1X increments

DRAM 3, 4, 5X

QPI: X23, X36

Voltages *CPU: 0.5-1.9V in 0.00625V increments

*CPU PLL: 1.6-2.24V in 0.02V increments

*QPI/VTT: 1.05-1.99V in 0.02V increments

*DRAM: 1.3-2.6V in 0.02V increments

*PCH: 0.95-2.0V in 0.02V increments

*DRAM data A/B: 0.72-1.59V in 0.01V increments

*DRAM address A/B: 0.72-1.59V in 0.01V increments

CPU: +0.006-0.389V in 0.006V increments

*CPU PLL: 1.0-2.43V in 0.01-0.05V increments

*CPU VTT:
0.451-2.018V in 0.004-0.005V increments

*DRAM:
0.906-2.405V in 0.006-0.007V increments

*PCH:
0.451-1.953V in 0.0005-0.006V increments

*DRAM data A/B: 0.825-1.225V in 0.025V increments

*DRAM address A/B: 0.825-1.225V in 0.025V increments

Fan speed control CPU CPU,
System 1-3
Gigabyte GA-P55M-UD4 MSI P55M-GD45
CPU support LGA1156-based Core i5, i7 series processors LGA1156-based Core i5, i7 series processors
Chipset Intel P55 Express Intel P55 Express
Interconnect DMI (2GB/s) DMI (2GB/s)
Expansion slots 2 PCI Express x16 (x16/x0 or x8/x8)

1 PCI Express x4

1 32-bit/33MHz PCI

2 PCI Express x16 (x16/x4*)

1 PCI Express x1

1 32-bit/33MHz PCI

Memory 4 240-pin DIMM sockets

Maximum of 16GB of DDR3-800/1066/1333 SDRAM

4 240-pin DIMM sockets

Maximum of 16GB of DDR3-800/1066/1333 SDRAM

Storage I/O Floppy disk

1 channel ATA/133 via Gigabyte GSATA2

5 channels 300MB/s Serial ATA with RAID 0, 1, 10, 5 support

2 channel 300MB/s Serial ATA via Gigabyte GSATA2

Floppy disk

1 channel ATA/133 via JMicron JMB363

6 channels 300MB/s Serial ATA with RAID 0, 1, 10, 5 support

Audio 8-channel HD audio via Realtek ALC889A codec 8-channel HD audio via Realtek ALC889 codec
Ports 1 PS/2 keyboard/mouse

9 USB 2.0 with headers for
4 more

1 RJ45 10/100/1000 via Realtek RTL8111D
1
hybrid eSATA/USB
1 1394a FireWire via
Texas Instruments TSB43AB23 with header for 1 more

1 analog front out
1 analog bass/center out

1 analog rear out

1 analog surround out

1 analog line in

1 analog mic in
1 digital S/PDIF out (TOS-Link)

1 digital S/PDIF out (RCA)

1 PS/2 keyboard
1

PS/2 mouse

10 USB 2.0 with headers for 4 more

1 RJ45 10/100/1000 via Realtek RTL8111DL

2 eSATA via
JMicron JMB363 with RAID 0, 1 support

1 1394a FireWire via Via VT6315N with header for 1 more

1 analog front out

1 analog bass/center out

1 analog rear out

1 analog surround out

1 analog line in

1 analog mic in

Our testing methods

We’ve narrowed our focus on motherboards today, and the testing conducted reflects that approach. If you’d like to see Lynnfield’s application and gaming performance against competing processor platforms, hit up our coverage of the CPU. For a closer inspection of the P55’s performance implications, we have an in-depth chipset review.

To determine whether these MicroATX boards can hang with the big boys, we’ve tested them against a collection of full-size ATX models. We’ve also thrown in an X58-based rig for reference.

Note that our X58-based system is only running its memory at 1066MHz—the fastest speed allowed by our Core i7-920 engineering sample. The platform’s third memory channel should easily make up the difference, but the results are only there for reference, anyway.

All tests were run at least three times, and their results were averaged.

Processor Intel Core i5-750 2.67GHz Intel Core i7-920 ES 2.67GHz
CPU/chipset link DMI QPI
Motherboard Asus P7P55D Asus P7P55D Deluxe Gigabyte GA-P55-UD4P

Gigabyte GA-P55M-UD4
Gigabyte GA-P55-UD6 MSI P55-GD65

MSI P55M-GD45
Gigabyte GA-EX58-UD5
Bios revision 0606 0504 F3 F3 F3 100 1.2 F3
North bridge Intel P55 Express Intel P55 Express Intel P55 Express Intel P55 Express Intel P55 Express Intel P55 Express Intel P55 Express Intel X58 Express
South bridge Intel ICH10R
Chipset drivers Chipset: 9.1.1.1015

AHCI: 8.9.0.1023

Chipset: 9.1.1.1015

AHCI: 8.9.0.1023

Chipset: 9.1.1.1015

AHCI: 8.9.0.1023

Chipset: 9.1.1.1015

AHCI: 8.9.0.1023

Chipset: 9.1.1.1015

AHCI: 8.9.0.1023

Chipset: 9.1.1.1015

AHCI: 8.9.0.1023

Chipset: 9.1.1.1015

AHCI: 8.9.0.1023

Chipset: 9.1.1.1015

AHCI: 8.9.0.1023

Memory size 4GB (2 DIMMs) 4GB (2 DIMMs) 4GB (2 DIMMs) 4GB (2 DIMMs) 4GB (2 DIMMs) 4GB (2 DIMMs) 4GB (2 DIMMs) 6GB (3 DIMMs)
Memory type OCZ OCZ3P1866LV4GK DDR3 SDRAM at 1333MHz OCZ OCZ3P1866LV4GK DDR3 SDRAM at 1333MHz OCZ OCZ3P1866LV4GK DDR3 SDRAM at 1333MHz OCZ OCZ3P1866LV4GK DDR3 SDRAM at 1333MHz OCZ OCZ3P1866LV4GK DDR3 SDRAM at 1333MHz OCZ OCZ3P1866LV4GK DDR3 SDRAM at 1333MHz OCZ OCZ3P1866LV4GK DDR3 SDRAM at 1333MHz OCZ OCZ3G1600LV6GK DDR3 SDRAM at 1066MHz
CAS latency (CL) 7 7 7 7 7 7 7 7
RAS to CAS delay (tRCD) 7 7 7 7 7 7 7 7
RAS precharge (tRP) 7 7 7 7 7 7 7 7
Cycle time (tRAS) 20 20 20 20 20 20 20 20
Command rate 1T 1T 1T 1T 1T 1T 1T 1T
Audio codec Via VT1828S with 6.0.1.7300 drivers Via VT2020 with 6.0.1.7300 drivers Realtek ALC889A with 2.31 drivers Realtek ALC889A with 2.31 drivers Realtek ALC889A with 2.31 drivers Realtek ALC889 with 2.31 drivers Realtek ALC889 with 2.31 drivers Realtek ALC889A with 2.31 drivers
Graphics Nvidia GeForce GTX 260 896MB with ForceWare 190.62 drivers
Hard drive Western Digital Raptor X 150GB
OS Windows 7 Ultimate RTM x64

Our test systems were powered by a PC Power & Cooling Silencer 750W power supply unit.

We’d like to thank Western Digital for sending Raptor WD1500ADFD hard drives for our test rigs.

We used the following versions of our test applications:

The test systems’ Windows desktop was set at 1280×1024 in 32-bit color at an 85Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests.

All the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.

Memory performance

Lynnfield’s on-die memory controller largely evens the playing field for P55 boards. However, each manufacturer has the freedom to tune the CPU’s integrated memory controller, and that can lead to small but consistent bandwidth and latency differences.

There’s very little difference in memory performance between the P55 boards, and the MicroATX models are right in the thick of things. In fact, the GD45 takes the top spot in CPU-Z’s memory latency test, albeit just barely.

Power consumption

Of course, performance is only one part of the equation for a modern PC. Power consumption counts for a lot these days, and that’s an area where motherboards can really set themselves apart. We’ve measured total system power consumption, sans monitors and speakers, at the wall outlet with a Watts Up Pro power meter. Readings were taken at idle and under a load consisting of a Cinebench render alongside the rthdribl HDR lighting demo. Windows 7’s “Balanced” performance profile, which uses CPU power management features, was enabled for this and all of our other tests.

Power consumption has become a hot topic among motherboard makers of late, and each board has a few different configuration options tied to BIOS switches and additional Windows software. MSI takes the simplest approach: either enable active power-phase switching (APS in the graphs below) or leave it off. With Gigabyte, you can run with or without Dynamic Energy Saver software installed. The Dynamic Energy Saver software didn’t work right with the UD5 BIOS revision we used for testing, though. Asus offers three options: an Xtreme Phase mode that’s enabled by default, an Xtreme-less normal operation mode, and a third mode calibrated by EPU software for Windows.

The MicroATX boards are among the most power-efficient at idle, with the GD45 drawing about a watt less than the UD4. Interestingly, the power-phase-switching capabilities of both micro models do little to lower system power consumption here.

MSI’s APS scheme doesn’t have much impact under load, either, but Gigabyte’s Dynamic Energy Saver software shaves more than 10W off the UD4’s socket draw. Having to run the DES Windows app to enjoy these power savings is a little annoying, though. The BIOS switch used to toggle APS on the MSI board is much less obtrusive.

The UD4 and GD45 both have admirably low power consumption overall. However, neither board is substantially more power-efficient than comparable ATX models.

Motherboard peripheral performance

To provide a closer look at the peripheral performance you can expect from these motherboards, we’ve compiled Ethernet, Serial ATA, USB, FireWire, and audio performance results below.

HD Tach FireWire performance
Read burst

speed (MB/s)

Average read

speed (MB/s)

Average write

speed (MB/s)

CPU utilization

(%)

Asus P7P55D
33.0

30.0

21.8

1.0
Asus P7P55D Deluxe 41.6 37.3 28.5 2.0
Gigabyte GA-EX58-UD5 30.5 28.8 17.2 1.0

Gigabyte GA-P55M-UD4

33.2

28.9

21.4

2.0
Gigabyte GA-P55-UD4P
40.9

35.5

25.4

2.0
Gigabyte GA-P55-UD6 40.9 35.6 25.6 2.0
MSI P55-GD65 31.8 28.6 18.7 3.3
MSI P55M-GD45
40.4

34.8

21.6

3.0

The GD45’s FireWire read speeds are quicker than those of the UD4, but write speeds are nearly identical.

HD Tach USB performance
Read burst

speed (MB/s)

Average read

speed (MB/s)

Average write

speed (MB/s)

CPU utilization

(%)

Asus P7P55D
35.1

34.4

33.8

3.3
Asus P7P55D Deluxe 35.1 34.5 33.7 2.7
Gigabyte GA-EX58-UD5 33.9 32.5 28.7 1.3

Gigabyte GA-P55M-UD4

35.6

34.0

32.1

5.7
Gigabyte GA-P55-UD4P
35.6

34.2

32.2

5.0
Gigabyte GA-P55-UD6 35.7 34.1 32.3 4.3
MSI P55-GD65 32.0 29.0 24.2 9.3

MSI P55M-GD45

32.0

29.0

24.4

4.7

Even though the MSI and Gigabyte boards use the exact same P55 USB controller, Gigabyte’s implementation is a little quicker overall, and notably faster in HD Tach’s write speed test.

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 P7P55D (P55)
252.7

110.5

110.1

7.1

2.3
Asus P7P55D (JMB363)
182.0

110.4

81.0

7.1

3.3
Asus P7P55D Deluxe (P55) 254.0 110.5 109.6 7.2 3.0
Asus P7P55D Deluxe (JMB363) 176.3 110.5 80.5 7.1 6.0
Asus P7P55D Deluxe (JMB322) 146.9 110.5 81.1 7.1 3.7
Gigabyte GA-EX58-UD5 (ICH10R) 253.6 110.5 111.0 7.1 1.3
Gigabyte GA-EX58-UD5 (GSATA) 145.6 110.5 81.5 7.1 2.0
Gigabyte GA-P55M-UD4 (P55)
207.8

107.1

107.9

7.5

4.7
Gigabyte GA-P55M-UD4 (GSATA)
181.8

110.5

80.9

7.1

2.7
Gigabyte GA-P55-UD4P (P55)
198.4

109.3

111.3

7.4

6.3
Gigabyte GA-P55-UD4P (GSATA)
191.1

110.5

81.0

7.1

4.0
Gigabyte GA-P55-UD6 (P55) 212.8 108.0 108.3 7.3 5.0
Gigabyte GA-P55-UD6 (GSATA) 181.9 110.5 81.2 7.0 4.3
MSI P55-GD65 (P55) 205.1 106.7 109.6 7.4 6.3
MSI P55-GD65 (JMB363) 163.8 108.5 76.2 7.2 7.0
MSI P55M-GD45
215.5

108.7

111.1

7.4

5.7

Neither Gigabyte nor MSI seems to have figured out how to get the P55’s SATA burst speeds up to par. Thus far, only Asus has the VelociRaptor we use for these tests bursting at its full potential.

NTttcp Ethernet performance
Throughput (Mbps) CPU utilization (%)
Asus P7P55D
944.8

5.9
Asus P7P55D Deluxe (RTL8111D) 941.3 6.1
Asus P7P55D Deluxe (RTL8110) 712.5 6.2
Gigabyte GA-EX58-UD5 (1) 940.8 2.4
Gigabyte GA-EX58-UD5 (2) 941.5 2.3
Gigabyte GA-P55M-UD4
944.5

4.8
Gigabyte GA-P55-UD4P (1)
947.9

5.3
Gigabyte GA-P55-UD4P (2)
945.8

5.4
Gigabyte GA-P55-UD6 (1) 935.3 4.7
Gigabyte GA-P55-UD6 (2) 941.2 4.7
MSI P55-GD65 (1) 938.9 9.0
MSI P55-GD65 (2) 941.5 5.3
MSI P55M-GD45
932.4

5.1

Thankfully, the days of PCI-based Gigabit Ethernet controllers are largely behind us. The PCIe GigE chips on the MicroATX boards deliver nearly equivalent throughput and CPU utilization.

RightMark Audio Analyzer audio quality: 24-bit 192kHz
Overall score Frequency response Noise level Dynamic range THD THD + Noise IMD + Noise Stereo Crosstalk IMD at 10kHz
Asus P7P55D
4

5

4

4

5

3

4

5

4
Asus P7P55D Deluxe 4 5 4 4 5 3 4 4 4
Gigabyte GA-EX58-UD5 5 5 5 5 5 4 5 4 5

Gigabyte GA-P55M-UD4

5

5

5

5

5

3

5

5

5
Gigabyte GA-P55-UD4P
5

5

5

5

5

4

5

5

5
Gigabyte GA-P55-UD6 5 5 5 5 5 4 5 5 5
MSI P55-GD65 5 5 5 5 5 3 5 5 5
MSI P55M-GD45
5

5

5

5

5

3

5

5

5

The UD4 and GD45 score identically in a RightMark Audio Analyzer loopback test between their respective front-channel outputs and line inputs.

Overclocking

Although it’s still early days, Lynnfield appears to be a willing overclocker. You have two options with the boards we’re looking at today: kick it old-school by cranking base clock speeds or use trendy new auto-overclocking features that seem to be all the rage. We’ve tested both, and I’ll start with the retro approach.

To probe each board’s base clock speed limits, we dialed back the CPU and memory multipliers to take them out of the equation. Next, we ramped up the base clock speed, testing for stability along the way with a combined Prime95/rthdribl load.

The UD4 effortlessly cruised up to a 210MHz base clock speed without so much as a bump in voltage. 220MHz wouldn’t post, though, and tweaking processor and chipset voltages didn’t help. I even tried relaxing memory timings, but the UD4 stubbornly refused to cooperate.

Like the UD4, the GD45 reached its peak stable base clock speed without extra voltage, and upping the voltage didn’t help. The board only made it up to 200MHz, but that’s still pretty respectable, and more than enough headroom for air-cooled processor overclocking.

What’s most interesting about these results is that they don’t stray from the peak base clock speeds we’ve achieved with full-size ATX boards from Gigabyte and MSI. That’s promising news for anyone leery of overclocking with a MicroATX mobo—well, these ones, anyway.

Next, we turned our attention to the auto-overclocking schemes offered by each board, starting with the UD4. Gigabyte’s Smart QuickBoost software selects a CPU clock speed based on the company’s own experience overclocking the processor in question. With our Core i5-750, QuickBoost settled on a 160MHz base clock speed. With Turbo Boost, that took the CPU up to 3.36GHz on just 1.216V.

This is the same clock speed that QuickBoost chose for the processor when we ran it on the UD4’s ATX siblings. Our i5-750 seems to be perfectly comfortable at 3.36GHz, too, even under a demanding four-way Prime95 load.

MSI’s approach to automatic CPU overclocking revolves around an OC Genie BIOS switch. Enabling OC Genie automagically overclocks the CPU at the next reboot, and it arrived at a somewhat different conclusion than QuickBoost.

OC Genie actually lowered the system’s CPU multiplier to 17X and then cranked the Base clock speed past 190MHz, pushing the CPU to 3.27GHz. That’s a little slower than what QuickBoost achieved, and it required a smidgen more voltage—1.312V, according to CPU-Z.

While I prefer overclocking the old-fashioned way, I do appreciate the fact that OC Genie doesn’t require Windows software. Of course, if the point of an automated approach is to bring overclocking to users who would otherwise be intimidated by poking around in the BIOS, Windows software might actually be preferable.

Conclusions

There was a time when buying a MicroATX motherboard meant making a lot of concessions. Today, however, one need not lose more than a couple of inches off the width of the board. The GA-P55M-UD4 and P55M-GD45 are both solid P55 implementations that retain the enthusiast-oriented sensibilities and apparent overclocking prowess of their full-size ATX brethren within the confines of a smaller form factor.

These Micro ATX models are a little cheaper than their ATX counterparts, and they consume less power, too. The only real drawback: a two or three fewer expansion slots. I don’t expect the extra slots will be missed by many. In fact, I’d wager that even most enthusiasts would be happy slapping in a graphics card and calling it a day. The UD4 certainly doesn’t need much more in the way of accompaniment. Sure, you could plug in a sound card, but those with a digital receiver or speakers won’t even need to go that far thanks to the ALC889A’s Dolby encoding capabilities. And don’t forget the hybrid eSATA/USB connectivity or the balanced dual-x8 PCI Express configuration that supports SLI in addition to CrossFire.

The only fly in the ointment is Gigabyte’s apparent disinterest in providing users with meaningful BIOS-level fan speed options. The UD4’s dearth of fan speed controls is embarrassingly dated for a cutting-edge mobo with an otherwise well-stocked BIOS, and Gigabyte would do well to study MSI’s significantly superior approach.

Unfortunately, the MSI board has issues elsewhere. The lack of hybrid eSATA/USB connectivity is disappointing, especially since it’s a feature available on MSI’s mid-range P55-GD65. However, that doesn’t irk me as much as the optional—and absent—digital S/PDIF audio output. I’m not crazy about the GD45’s inexplicably slower USB performance, either.

To MSI’s credit, the GD45 does cost a good $20 less than the UD4. The question you have to ask yourself is whether a better audio implementation, hybrid eSATA/USB, and a dual-x8 PCI Express layout is worth the extra scratch—and the lousy fan speed controls. For me, the answer is yes. While the GD45 feels like it’s been compromised to hit a lower price point, the UD4 really is a full-featured enthusiast board in a MicroATX package. The higher price tag is justified, I think, and I’m happy to spend a little more on what’s ultimately the better motherboard. Besides, if I were on a really tight budget, I’d be eying Gigabyte’s GA-P55M-UD2. The UD2 is similar to the GD45 but runs about $15 cheaper, and you get two digital audio outputs.

Comments closed
    • oldDummy
    • 11 years ago

    Nice article.

    smaller is better.

    Knew it all along. 😉

    • AxMi-24
    • 11 years ago

    Any chance of getting intel MBOs tested? Just the fact that they include intel network HW is good enough reason to pick them over most others.

      • MadManOriginal
      • 11 years ago

      The Intel mATX boards are kind of intriguing especially the top tier one although it’s price is quite silly. It’s got all PCIe slots and no legacy ports whatsoever.

    • FuturePastNow
    • 11 years ago

    I’ve bought a number of Gigabyte boards over the years and been happy with them, but I’m very impressed with what I’ve seen of recent MSI boards. Particularly the way they don’t go overboard with the power phases.

    Actual fan control sounds nice, too. Right now I use hardware fan controllers to quiet things down.

    • anotherengineer
    • 11 years ago

    I think it would be cool if these reviews offered Underclocking/Undervolting section as well.

    I like gigabyte, my current board is a 790gp-ud4h, the 2 MAJOR peeves I have with it is the high default voltages, and not being able to lower my dram voltage, and fan control. The lowest I can get for vdimm is 1.87V, however having ocz LV 1066 platinum says only up to 1.85V for voltage protection warranty doh, hopefully it lasts a long time.

    • OneArmedScissor
    • 11 years ago

    <rant>

    Considering the 40w idle power difference that increases almost linearly across the different boards, I hope TR will be a bit pickier and more consistent in the future with the motherboards used for testing.

    It’s one thing if they’re all clustered in one place and then one is way off, but that makes it readily apparent that motherboards vary more than CPUs do. So what good are CPU power comparisons where the motherboards vary?

    It drives me nuts when I look at the test setup and there’s like 4 different boards of one socket type that were used. It drives me even crazier when only one is used that’s a “high end” board, inundated with “bells and whistles” no one is going to ever touch.

    </rant>

    But really, despite being a bunch of similar boards, that was enlightening, and thanks for bothering to put that together.

    • derFunkenstein
    • 11 years ago

    “motherboardsy”

    Top of page 6

      • UberGerbil
      • 11 years ago

      That’s a word. I’ve stuck my head in a server and said “hmm, looking a little motherboardsy in here, isn’t it?” Not to mention the motherboardsy smell of my junk closet….

        • derFunkenstein
        • 11 years ago

        ah, yes. Good one sir. Then he was just using it out of context.

    • Jambe
    • 11 years ago

    Why does the UD2 have ALC888B – is 889A *[

      • Kurotetsu
      • 11 years ago

      Just how good is the 889A chipset? I’ve heard its a massive improvement, but how much? I’m particularly interested in its analog output.

        • Jambe
        • 11 years ago

        I’m just irked. UD2 ($110) has 1 GPU slot, ALC888B audio and, er, its eSATA isn’t also a USB. The UD4 ($140) has 2 GPU slots, ALC889A audio, and a dual-function eSATA/USB port. Why couldn’t they just put 889A on the UD2? How much of that thirty dollars difference would it cost?

        *shrug*

          • Skrying
          • 11 years ago

          They could, easily. It wouldn’t cost them hardly anything except that huge margin they’re making by charging $30 for the next motherboard up. This is a very rough guess but I’m going to say those changes in the $140 motherboard cost Gigabyte somewhere around $5.

          • UberGerbil
          • 11 years ago

          Taking a step up in trim level gets you the one feature you want and a bunch of features you don’t care about. Like that’s something new — in motherboards, or cars, or, well, anything.

            • Jambe
            • 11 years ago

            I didn’t say it was anything new, only that I was annoyed. I don’t see any differences between the two boards other than what I mentioned – does anybody else?

    • flip-mode
    • 11 years ago

    q[

      • derFunkenstein
      • 11 years ago

      case fans. I have Cooler Master R4’s and they’re OK but still louder than I’d like. It’d be nice if the BIOS let me crank them down based on CPU temps.

      Nitpicky? Maybe, but it’s the single solitary drawback to Gigabyte’s boards these days.

        • Skrying
        • 11 years ago

        That can be fixed with better case fans.

        On the other hand having to install a Windows program to enable power saving features is a major annoyance to me. Nearly all the programs motherboard manufacturers include are horrendous. I much rather Gigabyte develop a method similar to MSI’s, allowing you to enable such a common feature in the BIOS and not have to deal with bad applications.

          • flip-mode
          • 11 years ago

          ^ Good post. As you said, I’d rather have a case fan that was quiet without adjustment. I just plain don’t want to fool with that stuff anymore. I buy quiet fans so I don’t have to adjust them.

            • Kurotetsu
            • 11 years ago

            This is why Silent PC Review is awesome. I wanted a pair of 120mm case fans that would stay quiet even when running 100% from Molex. It introduced to Yate Loon. I got what I wanted.

            • MadManOriginal
            • 11 years ago

            Even besides that there are a few very simple and cheap options for fan control. I’m not a fan of using PSU rails not as intended for the 7V or 5V trick. Series diodes, with the number depending upon the desired voltage, or Zener diodes are great for getting a set voltage to a fan and they cost almost nothing. Of course there are the usual fan controllers too.

            If mobo makers added some more set percentage voltage options that would go a long way. Fancy temperature controlled variable speed settings might make some people happy but fans spinning up and down annoy me. I just set mine up for a speed that is quiet and make sure it’s enough to keep temps within reason, the fans I have atm are too loud on full speed for normal use but it gives me the option of cranking them if I feel like it.

            • OneArmedScissor
            • 11 years ago

            Amen. I want fan control so that I don’t EVER hear the fans, not hear them off and on. That’s even worse than a persistent sound.

            • derFunkenstein
            • 11 years ago

            I already have reasonably decent case fans. They were the quietest thing in my system before I replaced the CPU heatsink/fan combo. I don’t want to spend $60 on three fans.

            Air flow is such I could probably get away with just unplugging two of them and have a single exhaust fan, though. That would probably quiet it down enough without much rise in temps.

            • flip-mode
            • 11 years ago

            I’ve gone to a single 120mm case fan myself; it is quite adequate. I don’t ever want to need more than that, either.

        • Voldenuit
        • 11 years ago

        I wouldn’t call that their only drawback. Noisy audio, chipset/VRM coolers that block expansion slots, DES needing a software app to launch, lack of fan headers come to mind. I have an EP45-UD3P, and the audio output was very susceptible to electrical noise, and I was irked that it only had 3 fan headers (including CPU) for a mid-high end P45 board.

        At least it got me away from ASUS and their buggy Atheros networking chips…

      • OneArmedScissor
      • 11 years ago

      On top of fans being quieter, most stuff runs so cool and disperses heat so well now, most people should only need an exhaust case fan, anyways.

      Some “good” cases don’t even come with more than that, which should be a pretty clear indicator of what’s actually necessary.

    • UberGerbil
    • 11 years ago

    Those SATA burst results are a little disappointing. A year ago I’d be blowing that off, but given that these boards may be going into systems that could get an SSD added to them before they’re retired, you could be looking at a bottleneck eventually.

    Also, wrt to right-angle SATA cables, I’m pretty sure you can find ones that bend the other way. It’s a little hard to see in the pics, but I think eg this one does:
    §[<http://www.monoprice.com/products/product.asp?c_id=102&cp_id=10226&cs_id=1022601&p_id=1250&seq=1&format=1#largeimage<]§ I don't know what you do about the lower tier of connectors, however, if the top tier is already filled.

    • UberGerbil
    • 11 years ago

    Yes! I’ve been looking forward to an mATX comparison for a while. Good job TR.

    Now we just need to do something about the state of mATX cases….

Pin It on Pinterest

Share This