Logo   Banner   TopRight
Gigabyte X58A-UD3R: USB 3.0, SATA 6G
Date: Jul 01, 2010
Author: Joel Hruska
Introduction and Specifications

The past few years have been very kind to Gigabyte; the company's shipments have increased to the point that it's now approximately tied with Asus in terms of volume shipments. Part of the reason for the company's good fortune is its decision to aggressively align its higher-end motherboards, both in terms of price and available features. Gigabyte has put a major push behind USB 3.0, and claims to have purchased one million of the three million USB 3.0 controllers NEC has shipped thus far.

The X58A-UD3R we're reviewing today is a good example of Gigabyte's competitive product positioning. The board's feature loadout is excellent given its ~$200 price point—the question we'll be looking to answer is whether or not the company cut any corners to hit its target.

Gigabyte X58A-UD3R Motherboard
Specifications & Features
Processor and Chipset
Based on Intel X58/ICH10R chipset
Supports Intel Core i7 Processors

6 x DIMM, Max. 24 GB, DDR3 2000(O.C.)*/1866(O.C.)*/1800(O.C.)*/1600(O.C.)/1333/1066 ECC,Non-ECC,Un-buffered Memory
Triple channel memory architecture
Support Intel Extreme Memory Profile (XMP)
Expansion Slot
4 x PCIe physical x16
1 x PCI
2 x PCIe x1

SLI / Video Card Configurations*:
Single Card: Slot 1 (x16)
SLI/Crossfire: Slots 1 & 3 (x16, x16)
Tri-SLI/TriFire: Slots 1, 3, & 4: (x16, x8, x8)
QuadFire:  Slots 1-4: (x8, x8, x8, x8)
Storage I/O
Intel ICH10R controller
6x SATA 3 Gb/s ports
Intel® Matrix Storage supporting SATA RAID 0, 1, 10, and 5

Marvell 9128 SATA 6G Controller
2x SATA 6Gb/s Ports (RAID 0 + RAID 1)

Gigabyte SATA2 (JMB362)
1x IDE (ATA133)
2x SATA 3Gb (RAID 0 / RAID 1)

JMicron JMB362
2x eSATA USB/eSATA external ports
Integrated Peripherals
1 x Realtek 8111D  Gigabit LAN
Realtek ALC889, 7.1-Channel High Definition Audio
Front Panel Jack-Retasking
Coaxial / Optical S/PDIF out ports at back I/O
12 USB 2.0 ports
Multi I/O
1 x PS/2 Keyboard
1 x PS/2 Mouse
2 x USB 2.0/eSATA
4 x USB 2.0/1.1
2 x USB 3.0/2.0
2 x IEEE1394 (FireWire 400)
1 x Clear CMOS
1 x S/PDIF Out (Coaxial)
1x S/PDIF In (Coaxial)
1 x LAN(RJ45) port
6 x Audio

Special Features
Xpress BIOS Rescue
Download Center
Xpress Install
Xpress Recovery 2
Dynamic Energy Saver 2
Smart 6
Auto Green
eXtreme Hard Drive

Internal IO
24-pin EATX Power connector
8-pin ATX +12V Power connector
1 x Floppy Drive
1 x IDE Drive
1 x CPU Fan Header
3 x System fan header
1 x fan header
1 x North bridge fan header
1 x front panel header
1 x front panel audio
1 x CD-In
1 x S/PDIF In
1 x S/PDIF Out
2 x USB 2.0 Headers
1 x IEEE1394 / Firewire 400

Anti-Virus Software (OEM version)
Adobe Acrobat Reader
Firefox 3.5

ATX Form Factor
12 inch x 9.6 inch ( 30.5 cm x 24.5 cm )

  • Supports NVIDIA 2-Way and 3-Way SLI technology
  • Supports ATI CrossFire (2-way/3-way listed, quad support implied)

The X58A-UD3R packs all the standard features we'd expect from an LGA1366 motherboard, its six RAM slots, support for multiple video cards, and its plethora of USB 2.0 ports are all checked boxes on the usual list. Where the UD3R stands out is its storage capabilities, the board sports 10 SATA ports, an ATA port, and two combination USB 2.0/eSATA ports. Toss in the USB 3.0 ports and the X58A is capable of handling up to 16 hard drives at (or very near) native interface speeds. That's before we count external enclosures connected via USB 2.0, FireWire, or through the use of add-on cards.

Average Bundle, Excellent Documentation:
The included cables and various items are bog-standard (we dislike the lack of eSATA cables) but we want to take a moment to compliment Gigabyte's manual. When you've been building computers for a decade or more it's easy to overlook the importance of a well-written guide. Happily, Gigabyte didn't. The X58A-UD3R's manual weighs in at 125 pages, nearly all of them useful.

Gigabyte has diagrammed and documented the board's components, BIOS features, and bundled utilities and provided a step-by-step (and screenshot-by-screenshot) how-to on setting up a RAID array or installing an appropriate AHCI driver for Intel's ICH10, the JMicron controllers, or the onboard Marvell 9128 SATA 6G chip. The manual wraps up with a brief tutorial on configuring surround sound speakers; anyone who's curious can take a peek at the digital version

We can't say for certain if Gigabyte has provided this level of documentation for all its current motherboards, but this is precisely the sort of value-added feature that can make a practical difference to anyone whose nervous about building a system. Bravo.
Board Design and Layout

Gigabyte X58A-UD3R
A Closer Look 


The Gigabyte X58A-UD3R in all its glory. The board's layout is clean and uncluttered and we like the white/blue color scheme. Gigabyte rather thoughtfully stuck two PCIe x1 slots at the top of the board, allowing enthusiasts with thicker/non-standard video card coolers to make use of an x1 peripheral.




The board's backplate. Unlike some manufacturers, Gigabyte is still including both PS/2 ports. Optical and coaxial S/PDIF ports are provided along with a CMOS-clearing button.

Let's talk for a second about the group of yellow ports. The top left and right ports are both for FireWire 1394a, aka FireWire 400. Many camcorders and small devices use the smaller, four-pin connector on the right, while HDD enclosures are more likely to use the larger six-pin connector on the left.

Below those we have a pair of standard USB 2 ports. The bottom two ports are nifty, hybridized USB2/eSATA ports.  Both types of cables fit easily and Windows 7 had no trouble recognizing and configuring the attached hard drive. Switching from one interface to the other, even without rebooting, is not a problem. 

Finally, there's the standard group of audio ports and the two new, blue USB 3 ports. These will handle a USB 2 device just fine if you plug one in, but they're the only two ports on the board that support the higher transfer speeds.


Above we've got a larger photo on the mainboard and two different perspectives on the CPU socket. There's plenty of room around the socket—heatsink clearance is not a problem—but the mounting hole near the northbridge-mounted heatpipe is annoyingly hard to reach when attaching standard heatsinks.


Save for that one annoying tight spot, CPU mounting is a cinch. The entire board uses solid-state capacitors—it's been seven years since Taiwanese manufacturing defects led to an epidemic of unexpected motherboard failures, but solid capacitors have become a marketing bullet point meant to convey superior construction.



Above, it's all hot southbridge action. From the left we have the board's Gigabyte-branded SATA controller (actually made by JMicron), the southbridge's modest heatsink, and the single IDE port. Above middle shows how the curved SATA ports allow for longer, higher-end GPUs without blocking or interfering with the SATA cables themselves. Finally, there's the NEC USB 3 controller just to the left of the northbridge and above the first x1 PCIe slot.

With three separate drive controllers, it's important to know which chip controls which ports. The blue ports—ports 0-5—connect to the Intel ICH10R. The first set of white ports from the right (ports 6 and 7) connect to the Marvell 9128 SATA 6G chip. Finally, the last two ports, 8 and 9, are handled by the Gigabyte SATA2 chip.

Taken as a whole, the layout works quite well for us. There aren't any surprises, but there aren't any random oddities from overenthusiastic northbridge coolers or ports jammed in odd places, either.
Controller Conundrum: USB 3.0 & SATA 6G

Now that we've discussed the board's layout and design, let's take a closer look at two of its defining features. Gigabyte has put a major push behind USB 3.0 and SATA 6G, but first-generation controllers are often plagued with compatibility issues or simply aren't as fast as later iterations.

Below is a block diagram of the UD3R's Marvell 9128 SATA 6 controller. Of the four SATA 6 controllers Marvell currently has available, the 9128 is the nicest; it supports RAID, 128-bit/256-bit AES encryption, offers two SATA 6G ports, and an additional PATA port (though Gigabyte isn't using it). Some of you may recall the last-minute discovery of compatibility issues that forced board manufacturers to pull SATA 6G support right before the P55 chipset launched. The problem turned out to be with the 9123's PATA port; Marvell has respun the chip into the 9128, but Gigabyte opted to use a Jmicron controller for PATA support.


The only theoretical shortfall is the controller's available bandwidth. The Marvell 9128 connects to the system via a single PCIe x1 link capable of transferring 400MB/s. That's only 67 percent of the theoretical peak of the SATA 6G interface, but it's far more bandwidth than any conventional hard drive can saturate, even in RAID 0. SSD owners don't need to worry, either. No current single SSD can sustain read/write speeds much above 250MB/s, it's far from certain that even a pair of top-end drives in RAID 0 would actually encounter a significant bottleneck.

Next up we've got USB 3.0, courtesy of NEC's PD720200 controller. Here's the simplified block diagram (if you want to see the more detailed version, you can read about it here):

What we see above is that the NEC chip contains both a USB 3 and a USB 2 controller with both controllers interfaced to both ports. What the rather confusing intersection of arrows shows is that any combination of USB 2 and USB 3 devices can be attached without the need to worry about interface speeds. Need to combine a USB 2 and a USB 3 device on those two ports? No problem.

This sort of flexibility is often something that early adopters sacrifice in exchange for new technology, and it's to NEC's credit that their implementation avoids compatibility headaches. Those of you planning to use a pair of high-performance external enclosures, however, should be aware that there's a caveat—available bandwidth is shared between all devices. Slower, 2.5" drives may not be affected and may not be capable of saturating USB 3's bandwidth even in tandem.

Power Provisioning Muddies Backwards Compatibility Of USB 3 Devices In USB 2 Ports:

Seagate's BlackArmor 110PS

One other caveat about USB 3 devices in general is the question of how much power modern USB 2 ports can provide. We talked this over with Intel, which indicated that newer, higher-power USB 2 spots are probably those that conform to the USB-IF Battery Charging 1.1 Specification. This update was released in 2007 and refreshed about a year ago.

Unfortunately, it's hard to know which motherboards support it. The modern Core i7 boards we tested can power the Seagate BlackArmor PS110 from any USB 2 port, but one of our own X48 motherboard from Asus can only run the external enclosure from the back ports. Based on this, we don't recommend buying a BlackArmor with the plan to upgrade to USB 3 via controller card at a future date unless you know your motherboard can provide the necessary juice.

With that said, if you know your motherboard can charge an iPad while the device is running, it should also be able to run the external hard drive.

BIOS Options

A Look At The BIOS
Lots of Options and Tweakability

Gigabyte's BIOS options are extremely thorough in every area but one.


The first screenshot is the top-level BIOS navigation page, with PC Health and Advanced BIOS features on the far right. Compared to boards from certain competitors, Gigabyte's board-level fan speed controls leave nearly everything to be desired. You get exactly two settings—Smart Fan and Smart Fan Mode. 

The Advanced BIOS options (and really, all of the other areas the BIOS covers) are quite detailed, as shown below.


Integrated Peripheral options are on the far left, the control panel for various CPU and motherboard settings on the right. There's a full range of settings to experiment with—a feature that can come in exceedingly handy if you're trying to find the exact point at which your system is stable while overclocked.


What we'd really like to see Gigabyte offer is a BIOS that's just as detailed as the X58A-UD3R's, but includes comprehensive fan setting options. Sure, it's fair to argue that those features only matter to a small number of enthusiasts, but if you check some of the screenshots above, it's clear that the company provides much more overclocking and voltage control than anyone but the highest of the high-end OCers would actually need.

Since this isn't a zero-sum game, we're hoping Gigabyte might listen on this one and add the one feature that sticks out as weak in a virtual garden of options.

Test Systems, Storage Controller Tests

Test System Configuration Notes: All of our benchmarks were run with RAM SPDs set to Auto and a 1066MHz memory clock. AHCI or IDE mode was configured as noted in specific test scenarios. Because the X58A-UD3R focuses on storage and interface performance, we've chosen to highlight performance in this area rather than strictly relying on a generic test suite.

The fact is, motherboard performance within a given chipset family varies very little these days. When such variances occur, they're either typically within a 1-3 percent margin of error or are the result of subtle manufacturer overclocking. The situation has changed dramatically from what it was 8-10 years ago, when a motherboard roundup could include boards with chipsets designed by ALI, AMD, ATI, Intel, NVIDIA, SiS, or VIA. As if that wasn't difference enough, there were three RAM standards in play over the same time period (SDR, DDR, and RDRAM), and different memory channel configurations within the same socket family. 

Today, instead of seven chipset manufacturers, there's just Intel, AMD, and NVIDIA's fading spectre. With CPU memory controllers now on-die, there are only two "northbridge" functions that meaningfully impact performance:  PCI-E x16 lane configuration and, on an AMD board, the clockspeed and dedicated RAM for the onboard GPU. Of these two, AMD's Fusion products will eliminate the latter in the not too distant future.

We've included some general performance benchmarks that pit the Gigabyte X58A-UD3R against an EVGA X58 SLI Classified, but these results are meant to confirm performance parity between the two products. Our primary interest this time out is in the comparative performance of the X58A-UD3R's peripheral controllers and explore how the board performs under stress. For our USB 3 performance comparisons, we used the Seagate BlackArmor 110PS. When comparing AHCI and IDE performance across controllers, we opted for the highest-performing HDD we have on hand; a WD VelociRaptor 300.

 HotHardware's Test System
 Gigabyte X58A-UD3R
Gigabyte X58A-UD3R
EVGA X58 SLI Classified


Core i7 920 (2.66GHz - Quad-Core - 2.8GHz Intel Turbo Mode Enabled)

3x2GB Elpida DDR3-133
CL 7-7-7-20 - DDR3-1066
CL 7-7-7-20 - DDR3-1333
CL 7-7-7-20 - DDR3-1333 (12GB)
CL 7-7-7-20 - DDR3-1600 (12GB)

Hard Drive:
WD 1TB Caviar Black (32MB Cache)
WD VelociRaptor 300
Seagate BlackArmor 110PS

Onboard Controllers:
Intel ICH10
Gigabyte GSATA (JMicron 632)
JMicron 632 (eSATA ports)
Marvell 9128 (SATA 6G)

Radeon HD 5970
On-Board Ethernet
On-board Audio

Windows 7 Home Premium 64-bit
Catalyst 10.2 drivers.


eSATA Remains Relevant, Despite USB 3.0:

eSATA has never been all that popular, but it continues to offer certain benefits that even USB 3 can't match, particularly for high-performance consumer storage arrays. Because eSATA is a native interface—your motherboard has no idea if the attached drives are in a case or in the bathtub—it supports SMART monitoring, RAID, AHCI, and NCQ without the need for a multi-drive enclosure and a controller chip. It's also potentially faster than USB 3 in RAID scenarios; connecting multiple eSATA drives to the same motherboard doesn't split available bandwidth the way a USB controller would.

eSATA enclosures typically require their own power cables, but the standard's benefits can clearly outweigh the drawbacks depending on the sort of external storage solution you're attempting to build.

eSATA has never been all that popular, but it continues to offer certain benefits that even USB 3 can't match, particularly for high-performance consumer storage arrays. Because eSATA is a native interface—your motherboard has no idea if the attached drives are in a case or in the bathtub—it supports SMART monitoring, RAID, AHCI, and NCQ without the need for a multi-drive enclosure and a controller chip. It's also potentially faster than USB 3 in RAID scenarios; connecting multiple eSATA drives to the same motherboard doesn't split available bandwidth the way a USB controller would. eSATA enclosures typically require their own power cables, but the standard's benefits can clearly outweigh the drawbacks depending on the sort of external storage solution you're attempting to build.


SiSoft Sandra
 Preliminary Testing with SiSoft SANDRA 2009
 Synthetic Benchmarks

We began our testing with SiSoftware's SANDRA 2009 SP1, the System ANalyzer, Diagnostic and Reporting Assistant. We ran three of the built-in subsystem tests that partially comprise the SANDRA 2010 SP1 suite with the Gigabyte X58A-UD3R and the EVGA X58 SLI Classified
. All of the scores reported below were taken with the processor running at its default clock speeds of 2.66GHz but with Intel Turbo Boost enabled for a final clock speed of 2.8GHz and  3GB of DDR3-1066 RAM running in triple-channel mode.

Equal processors, equal platforms, equal performance. The Gigabyte board posts a tiny lead in several benchmarks, but nothing that escapes a standard margin of error.
PCMark Vantage

Futuremark PCMark Vantage
Simulated Application Performance

We then ran our test motherboards through PCMark Vantage, Futuremark’s latest system performance metric built especially for Windows Vista. PCMark Vantage runs through a host of different usage scenarios to simulate different types of workloads, including High Definition TV and movie playback and manipulation, gaming, image editing and manipulation, music compression, communications, and productivity. Most of the tests are multi-threaded as well, so they can exploit the additional resources offered by multi-core CPUs.


We checked Vantage's subtest scores to verify that there were no differences that merited demonstration, but again, the two boards were in step right down the line. The Gigabyte X58A-UD3R is consistently just a hair faster than the EVGA X58 SLI Classified, but the difference is so small we'd scarcely call it significant.

The real reason to opt for one over the other are features like USB 3 and SATA 6G; we'll explore the former's performance later on in the review.


LAME MT and Low-Res Gaming

LAME MT MP3 Encoding Test
Audio Encoding with Up To Two Threads of Processing

In this test, we created our own 223MB WAV file (a hallucinogenically-induced Grateful Dead jam) and converted it to the MP3 format using the multi-thread capable LAME MT application in single and multi-thread modes. Processing times are recorded below, listed in seconds. Shorter times equate to better performance.

Each motherboard wins and loses a test by a second with the Gigabyte board being a tad faster in multithreading and the EVGA a bit faster single-threaded.


Low-Resolution Gaming: Crysis - CPU-Bound Test
Taking the GPU out of the Equation

Gaming has been primarily GPU-bound since 3dfx introduced the original Voodoo, but cranking game resolution down and turning detail levels to minimum can reveal just how quickly and effectively the CPU is crunching data. We tasked both motherboards with running the first CPU benchmark in Crysis and measured the results.

We didn't expect to see a difference here, really--and we don't. Either motherboard is capable of taking full advantage of the Core i7-920 we've installed.

USB 2, 3, & eSATA: HDTach, SiSoft Sandra


HD Tach 3.04.0
Simulated Application Performance

We've started with HD Tach—while the program is almost certainly on its way out, it's still useful for comparing read performance between the various drives we're testing. Since HD Tach can only compare performance between two drives at a time, the top graph is USB 3 vs. USB 2, while the bottom graph is USB 3 vs. eSATA.

USB 2 vs. USB 3 Read Performance. 

USB 3 vs. eSATA

These performance results are as encouraging as anyone could ask for. The 7200.4 Momentus hard drive inside the BlackArmor isn't a high performance drive by desktop standards, but its read performance triples when we move to USB 3. If we toss out the random dip  in read performance, eSATA and USB 3 are almost identical, with the edge apparently going to USB 3. eSATA's burst read speeds are about 15 percent higher, however, probably due to the native interface's lower latency. We don't know for certain if USB 3 can keep pace with eSATA when comparing faster drives yet, but as far as mass market external enclosures are concerned, it's a non-issue—and a fabulous performance improvement over its predecessor.

SiSoft Sandra 2009
Simulated Application Performance

Our next stop is SiSoft Sandra 2009 and that application's suite of hard drive tests. Sandra also reports the random access times and full stroke times in milliseconds. 

Again, we see an advantage for eSATA when it comes to full stroke and random access speeds but read speeds are identical for both of the two high-speed standards.

Based on what we've seen here, it's time to wave goodbye to USB 2. Depending on what you need from an external storage solution, it may even be worthwhile to replace a USB 2 enclosure with a USB 3-capable one and add USB 3 support via a PCI Express card. USB 2's bandwidth is capable of handling simple desktop tasks, including streaming 1080P video smoothly, but you'll definitely notice the difference if you spend any amount of time shifting data from Point A to B.

USB 2, 3, and eSATA Continued: PCMark Vantage

PCMark Vantage
Simulated Application Performance

We used the 64-bit version of the Vantage benchmark here, and defragmented the hard drive immediately prior to running it. The test was looped 3x. One thing to keep in mind when comparing PCMark Vantage results is that the benchmark's margin of error is fairly wide—we'd estimate a few percent points at least. Relevant factors include  whether or not the hard drive was defragmented immediately prior to the run and whether Vantage was run immediately following OS+driver installation, or only after a full suite of tests and other benchmarks had been run. We ran full iterations of the benchmark in all situations, but have limited the sub-test breakdown to those individual suites that actually show variation.  We benchmarked the Seagate 110PS using both USB 3 and USB 2, then attached the bare drive to the Gigabyte's eSATA ports and tested it with AHCI enabled and disabled.

The aggregate scores are all close together; there's just a 4.8 percent difference between the USB 2 and eSATA AHCI tests. We saw more significant variation in the "Music" test suite, where the AHCI-enabled eSATA port easily outpaced the other connections. We have to note that we saw quite a bit of variation in the Music suite when testing eSATA with the JMicron AHCI driver; scores ranged from 6792 to the listed 8176. The higher number is given here because it's what was mostly reported, but the test continued to periodically throw low numbers.

We already take multiple steps to ensure PCMark Vantage runs in a clean environment and we loop the test 3x each time. The 20 percent variation we saw simply shouldn't be occurring, but since it was unique to the JMicron-provided driver, we're assuming it's not PCMark Vantage causing the problem—at least not entirely. In the drive-specific benchmarks, USB 2 is blown away by both of the other two interfaces. While that's no surprise, the fact that modest 2.5" mobile hard drive is nearly 50 percent as fast is a strong indication that even older/lower-end enclosures could benefit significantly from USB 3.0, assuming you take the time to upgrade them.

USB 3.0 Performance Evaluation

If you've been eyeing USB 3 and wondering if its worth the cost, we'd say yes. First generation or not, NEC's controller performed flawlessly in our tests, and Windows 7 had no problem recognizing and installing the proper USB 3 driver. The difference between USB 2 and 3 is quite noticeable, even with modest hardware; consumers who work with external drives on a regular basis may find this to be an extremely practical upgrade.

While USB 3 probably will shove eSATA out of the market, the older external SATA standard is more than capable of keeping up with its new rival and still offers several features that USB 3 can't match. If you like the performance of USB 3 but aren't ready to jump for a new motherboard just yet, eSATA is a good alternative if you've got the ports. Alternatively, you can grab an expansion card; there are several PCI-Express solutions that combine both NEC's chip and Marvell's 6G controller on a single PCB.

Storage Benchmarks: Intel vs. Everyone

If you've been around a few years, you can probably remember when hard drive manufacturers began introducing features like native command queuing (NCQ) and Intel created the advanced host controller interface (AHCI) standard. The purpose of AHCI was to expose the more advanced capabilities of SATA as compared to the old parallel ATA (PATA) standard, while maintaining backwards compatibility with the older IDE standard. Even now, years (and several OS releases) after AHCI first debuted, it's still a feature most motherboards disable by default. Enabling it isn't as easy as flipping a BIOS switch; the OS almost always has to be reinstalled to enable or disable AHCI support. Even once that's done, there's still the issue of which driver one should use, Microsoft provides native AHCI drivers for Vista and Windows 7, but third party controller manufacturers still provide their own custom solutions.

The reason anyone bothers in the first place is because without AHCI, it's impossible to hot-swap SATA hard drives and NCQ doesn't function (some utility programs will report NCQ support is available, but this is based on whether or not the hard drive supports it, not whether or not it's actually running). Since the X58A-UD3R is such a storage-centric board, we opted to focus on comparative AHCI performance between the three separate controllers Gigabyte included. We tested PCMark Vantage in the following scenarios, all tests were conducted using a WD300 VelociRaptor hard drive.  PCMark Vantage was looped 3x in each instance. 

We've broken eSATA, IDE, and AHCI performance out into three separate charts and included the aggregate PCMark Vantage score as well as the results of any suite subtests that showed potentially meaningful variation.

The Contestants:


  • Gigabyte GSATA2:  This is a rebadged JMicron JMB362. Tested using both the Gigabyte-provided drivers and a newer JMicron driver from the company website.
  • eSATA Ports:  These are also controlled by a JMicron JMB362. Tested using the same set of drivers as the GSATA chip.
  • Marvell 9128:  The only 6G controller chip around at the moment. Tested using the latest company driver. 
  • Intel ICH10: Tested using Intel's own driver.

PCMark Vantage   
Synthetic Application Performance

Performance scarcely budges when all of the controllers in question are run in IDE mode. The Intel controller is about 7 percent faster than the GSATA2/JBM362, with the Marvell halfway between the other two. The HDD test was the only suite that demonstrated any variation; all three controllers were within 2-3 percent of each other in all other tests.

AHCI performance is much more varied. Intel's ICH10 leads the aggregate score by a whisker, but decisively wins out over the JMicron controllers in the HDD test suite. Given Vantage's noted tendency to throw wild scores, the six percent gap between ICH10 and Marvell's 9128 is too close to comfortably call, but the Intel chip is a full 30 percent faster than the JMicron 362 using Gigabyte's driver and ~25 percent faster than the reference driver from JMicron. JMicron's reference driver is an interesting case. Flip back a page to our eSATA comparisons using the Seagate BlackArmor P110, and you'll note that PCMark Vantage again reported a much higher result in the "Music" suite when using this particular driver than when using anything else. With the exception of that particular series of tests, however, the driver's performance is consistent. It's typically a few percent faster than the Gigabyte-provided driver, but falls behind the Intel and Marvell controllers.

If you check the HDD Test suite scores, you'll note that while testing in AHCI mode increased aggregate PCMark Vantage results across the board, HDD Test Suite results moved in the opposite direction. Intel and Marvell performance rose by six percent and four percent, but the JMicron controller's performance fell 9-13 percent depending on which driver we tested. One of the good things about the X58A-UD3R is that it allows you to manually select which controllers you want to operate using AHCI and which you want to use as IDE. Given the results we've seen, it may be best to keep the JMicron ports on IDE, particularly if you plan to hook a CD drive up to them. We've not had problems personally, but some CD-ROM drives don't play nice with AHCI drivers and third-party controllers.

Before we wrap this up, let's check eSATA performance again, with the faster VelociRaptor instead of our mobile Seagate.

eSATA performance follows the same trend we saw when testing the board's three controllers, although the baseline scores aren't as high. Again, JMicron's own driver outperforms Gigabyte's included version.

Overclocking and Stability

Overclocking is, by its very nature, unpredictable. Even if you buy a CPU according to make, model, and week of production, there's no telling how much extra performance you might or might not get through overclocking. This is definitely true in in the case of the Core i7 920 we used when testing this board—the CPU runs just fine at stock speeds but overclocks with the grace and agility of a one-winged epileptic duck.

OverclockingThe X58A-UD3R
Reasonably Robust

This is less than fun, and it makes it difficult to push a motherboard given the limited range of multipliers that keep the CPU's clock sufficiently low. In this case, however, a recalcitrant processor doesn't seem to have been the issue. The X58A-UD3R was nominally stable at a Base Clock of 180MHz, but increasingly unstable from the 180-200MHz range. Higher voltages would, we think, have stabilized the board at the 200MHz mark, but we'd already hit the maximum recommended voltages for the CPU and QPI links.

The QPI speeds were one factor in the system's instability above 180MHz. Gigabyte offers just four options—Slow, x36, x44, and x48. System stability improved if we used the "Slow" setting, but this slowed the interconnects to just 100MHz. The X58A-UD3R is certified for the 6.4GB/s QPI speeds higher-end Core i7 processors use, but any setting that pushed the transfer rate above 7GB/s noticeably increased the likelihood of a crash.

Overall, we'd rate the X58A-UD3R as a fairly good overclocking board, if not an exceptional one. All of the right BIOS options are baked in, voltage control is extremely granular, and the board recovers easily from incorrect voltage/frequency settings.

Stability Testing:
As we noted in our Test System information, we wanted to do more than simply put the X58A-UD3R through a standard set of paces. After the normal benchmarks were complete, we loaded the board with six DIMMs of dual-bank Elpida 2GB DDR3-1333 modules.

The more DIMMs (and by extension, the more banks) of RAM that are present, the harder it becomes for the memory controller and motherboard design layout to handle the load. Loading all six DIMM slots with dual-bank RAM let's us see if the board is stable under stressful conditions. We looped PCMark and 3DMark Vantage several times each to ensure system stability.

The X58A-UD3R passed our tests with flying colors. It had no trouble running at 1333MHz with all six DIMMs installed and took the RAM up to 1667MHz without a hitch or the need to raise CAS/RAS latencies. When it comes to memory stability, this board is a rock.


Performance Summary: In a word, we'd categorize the Gigabyte X58A-UD3R's performance as "excellent." USB 3 is great, the combo USB2/eSATA ports work well, and both the Marvell and Intel ICH10R controllers turned in solid numbers. Admittedly, the JMicron chips aren't so great—we'd prefer to see Gigabyte use other solutions on the market—but the board allows each controller to be configured for IDE or AHCI individually. That makes it much easier to fine-tune devices for optimum performance.

Stability was also excellent. We had no problems running 12GB of RAM across all six slots, even when overclocking, and had no crashes.

We like this Gigabyte motherboard and we like it quite a bit. For $200, Gigabyte has put together a package that overclocks well, hits all the major features, and offers both USB 3 (immediately useful) and SATA 6G. The storage options are also great; you could technically hook up 28TB of storage to this thing and have your own massive storage server.

SATA 6G is currently nothing more than a marketing bullet point since even high-end SSDs can't saturate it, but the Marvell controller actually keeps up with Intel's ICH10R reasonably well. As for the JMicron controllers, they perform adequately in IDE mode, but AHCI performance leaves something to be desired and throws back some odd results.

Given that the cheapest LGA1366 boards are around $150, the X58A-UD3R's price is quite reasonable, particularly for what you're getting. Unlike some of the other X58 boards in the sub-$200 market, it also supports two or more video cards in SLI without hanging one of the x16 physical connectors off the southbridge. The only option missing that we wished we had is better fan control in BIOS. For as good as the board is, that's a very small niggle indeed.


  • Excellent documentation
  • Tons of storage options
  • USB 3.0. It's sexy.
  • No fan/noise controls to speak of.
  • JMicron SATA controllers don't keep up with Marvell or Intel.

Content Property of HotHardware.com