Broadwell-E is here, giving Intel’s high-end desktop platform an impressive shot in the arm. This refresh bumped the top-end X-series part to ten cores—each now sporting Intel’s Broadwell microarchitecture—all while fitting into the same LGA2011-v3 socket that we welcomed with Haswell-E. Part of Broadwell-E’s appeal is the fact that the chip is a drop-in replacement for its Haswell-based predecessors. Existing X99 boards should require no more than a firmware update to work with the latest processors.
That said, motherboard manufacturers are taking this opportunity to release refreshed X99 boards. This new breed has been spruced up with all the latest hotness in storage and peripheral connectivity that their predecessors lack. In particular, we’re seeing support for standards like USB 3.1 Type-C, Thunderbolt, and U.2 ports on some boards. One example of this refresh wave is Asus’ X99-A II, the Broadwell-E refresh to Asus’ X99-A. That board won our coveted Editor’s Choice award when it was the new thing on the market. Let’s dig in to see if the sequel is as good as the original.
Asus has kept the same monochromatic visual theme going with the X99-A II as we’ve seen from its previous X99 and Z170 boards. The company has made a few cosmetic tweaks compared to its first-generation X99 boards, though. Instead of heatsinks clad in matte black and white, Asus has gone with silver and white to match is current crop of Z170 “pro” boards. The printed circuit board itself is still completely black, however.
The X99-A II makes full use of every square inch of board real estate. Alongside the huge LGA2011-v3 CPU socket, Asus’ engineers managed to squeeze in the full allotment of eight DDR4 DIMM slots. This also lets Asus make use of all nine ATX mounting holes. If builders need to apply pressure to the board when it’s mounted in a case, the X99-A II’s PCB won’t flex like some narrower ATX motherboards might.
Flipping the board over reveals several clusters of surface-mounted components, a second Asus TPU chip, and a row of LEDs for the under-board lighting. The back of the board also clearly shows that all the heatsinks are held in place with screws. There’s even a large backplate on the underside of the main VRM heatsink. This should ensure that the electrical components make adequate contact with the metal above.
On the topic of voltage regulation, it’s worth pointing out that Broadwell-E, like the previous Haswell-E, makes use of Intel’s fully-integrated voltage regulator (FIVR) technology, although it’s presumably a more advanced iteration than what we saw with Haswell. The motherboard-based VRM components are found underneath the two VRM heatsinks, the second of which is hidden underneath a large plastic shroud that extends all the way to the rear port cluster. This shroud is purely cosmetic, and it’s secured with just three screws on the underside of the board. Removing it could improve airflow to that left VRM heatsink. It also gives the X99-A II a more blacked-out look.
One difference compared to Asus’ original X99-A is that the board is fitted with two CPU power connectors: one eight-pin EPS12V and one four-pin ATX12V. Both of these connectors need to be occupied for proper operation of the board. When we stop to consider the potential power draw when overclocking Intel’s top-end ten core Broadwell-E part, this move certainly makes sense.
This power-connector requirement caught me off guard as I opened the X99-A II’s retail box. Thankfully, a quick trip to my box of cables, connectors, and miscellany rectified the situation with a handy Molex to four-pin ATX12V adapter, since my Cooler Master V750 is only fitted with a single eight-pin ATX12V connector. Be sure to check your power supply, and your cables box, before opening your wallet for this motherboard.
Taking a stroll around the socket
Eight DDR4 DIMM slots are one of the more impressive features of high-end boards built on Intel’s X99 platform. That’s even more true when we consider that boards like the X99-A II can support up to 128GB of RAM when fully populated. These slots take up a great deal of real estate on the board, which can push slots and sockets closer together than on mainstream platforms. Take the top-most PCIe slot on the X99-A II, for instance. It’s close enough to the CPU socket that there’s potential for clearance issues with large aftermarket CPU coolers. To help you figure out which components can safely fit together on the board, we’ve provided the measurements below:
Five fan headers are within easy reach of the CPU socket: two for CPU fans, one for use with a CPU liquid-cooler pump, one for a system fan, and one high-amperage header that can supply 3A of current.
Unlike many X99 boards, the X99-A II doesn’t have the DIMM slots butting right up against the top edge of the motherboard. This can make installing and removing DIMMs easier since it grants better access to the locking mechanisms on each slot. This breathing room comes at the cost of the top-most expansion slot, however, leaving the board with six PCI Express expansion slots.
The X99-A II gives us four PCIe x16 slots. Three of these are fed with Gen3 lanes from the CPU: the left-most and two right-most x16 slots. SLI and CrossFire configurations with up to three video cards are supported with both 40-lane and 28-lane CPUs from the Haswell-E and Broadwell-E families alike.
For dual-card setups, the first and third x16 slots from the left should be used, and each of those will get the full 16 lanes with 40-lane CPUs. This arrangement gives breathing room not only for a pair only double-wide cards, but also for two triple-slot behemoths. In the unlikely event that you can find a reason to use two triple-slot coolers, you won’t be left with any remaining PCIe slots, but you’d be the talk of the town. The right-most x16 slot comes in to play for three-card configs.
The second PCIe x16 slot from the left is fed with Gen2 lanes from the chipset. Normally, this slot gets by with a single lane of bandwidth, but if you’re willing to disable the board’s right-most PCIe x1 slot and the USB 3.1 controller, it can be fed with four lanes of Gen2 PCIe goodness. My take is that it would have been preferable to have this lane-sharing happen with the left-most PCIe x1 instead, because it’s far more likely to be covered up by a graphics card.
The silver cladding on the left-most PCI Express x16 slots is more than just some bling. Similar to other boards that we’ve looked at lately, Asus has reinforced this primary PCIe slot with a metal shroud that is soldered to the board at multiple points. This setup, which Asus calls SafeSlot, should reduce the chance of damage to the PCIe slot if you’re transporting a system that has a massive video card, but we’d still recommend removing any expansion cards rather than chancing damage to the board with a bump or jostle. It’s somewhat surprising to see this reinforcement on the first PCIe slot only. It would have been nice to have all three of the PCIe slots that are connected to the CPU protected in this manner.
On the subject of bling, the PCIe slots don’t leave us wanting. The translucent locking mechanisms on each of the x16 PCIe slots actually have embedded RGB LEDs in them. This lighting works in conjunction with the audio trace path lighting as well as any LED strips connected to the onboard 4-pin RGB headers. More on this later.
The number of lanes going to each slot for the different possible multi-GPU configurations will depend on whether your little slice of Broadwell-E silicon has 40 or 28 PCIe lanes enabled. Rather than attempting to paint you a picture with prose, we’ve instead mapped out how PCIe lanes are assigned to slots in diagrams. Click the buttons to toggle between the two possible processor options:
All four x16 slots are usable, no matter how many PCIe lanes your CPU provides. The processor choice just governs how many lanes each x16 slot gets: for models with 40 lanes, the three CPU-fed slots will have lane arrangements of x8/16/x8, while those running Core i7-6800Ks are looking at x8/x8/x8.
Between the third and fourth x16 slots we find the tail end of the board’s M.2 socket. We’ll look at it in more detail on the next page when we examine the X99-A II’s storage features.
Storage, audio, and the goodies
The X99-A II’s SATA-based storage connectivity can be found running up the right-hand side of the board. All of the ports are right-angled to face the edge of the board, which can make for easier cable insertion when longer graphics cards are installed.
Here we can see one of the X99 platform’s trademarks. The X99A-II has an abundance of SATA-based storage with ten 6Gbps SATA ports in total. Within the gray SATA Express connector are two standard Serial ATA ports atop a backward-compatible SATA Express port. The four SATA ports in the center and the two on the right edge of the picture above—those clad in gray—can be combined into RAID arrays using Intel’s drivers. The leftmost four SATA ports, clad in black and labeled SATA6G 7-10, operate in either IDE or AHCI mode only. This arrangement is a fundamental limitation of the X99 chipset. That said, it doesn’t preclude you from using OS-managed or third-party software RAID if you’d like.
Hanging out to the left of all the SATA-based storage, we can see a right-angled U.2 connector. Four Gen3 lanes from the processor are shared between this U.2 connector and the M.2 port just behind it. As as result, it’s not possible to use both the U.2 and the M.2 connectors at the same time.
The X99-A II’s M.2 port is located south of the low-profile chipset heatsink. This lets the board support devices up to 110 mm long, and it could allow mini-SSDs installed here to run cooler than on boards with M.2 connectors situated underneath PCIe x16 slots.
The four Gen3 PCIe lanes grant storage devices connected to either port an impressive amount of potential bandwidth—up to 32 Gb/s (4 GB/s). This is bandwidth straight from the processor, as well, so builders need not worry about their next-gen storage getting bottlenecked by the DMI link connecting the chipset to the CPU. SATA-based M.2 devices need not apply on the X99-A II, though. Only PCIe-based NVMe storage devices are supported.
The X99-A II’s rear port cluster looks a little Spartan, but it should have most users covered. The first thing one might notice is the button all the way to the left. This is used to kick off the firmware flash procedure when using Asus’ USB BIOS Flashback. To use this handy feature, simply put your USB thumb drive loaded with your desired firmware release into the bottom right-most USB 3.0 Type-A port. Then with your computer off, press the button for three seconds, and the board will start the flash procedure. No need to install a supported processor to update the firmware. Although it might not be used every day, USB BIOS Flashback can come in very handy if you need to flash the firmware to support your processor of choice—not an inconsequential consideration with Broadwell-E.
The remaining three blue USB 3.0 ports are connected to an ASMedia ASM1074 USB 3.0 hub chip, which is fed from one of the X99 chipset’s USB 3.0 ports. For USB 3.0 ports directly linked to the chipset, you’ll need to look toward the port that can be used for USB BIOS Flashback, or one of the four ports that are made available through two USB 3.0 internal headers.
For the latest USB 3.1 hotness, Asus outfits the X99-A II with both a Type A and a Type C port, courtesy of an ASMedia ASM1142 controller. This controller is connected to two Gen2 PCIe lanes from the chipset. As mentioned on the previous page, these two lanes are also shared with the second x16 PCIe slot. Rounding out the board’s USB support, we have four USB 2.0 ports on the rear port cluster, with four more available via two internal headers.
A single PS/2 port capable of supporting either a keyboard or a mouse is the only legacy interface to be found on the port cluster. This port, along with the serial port header found on the bottom of the board come thanks to the X99-A II’s Super I/O controller, a Nuvoton NCT6791D.
The Gigabit Ethernet port is powered by Intel’s I218-V controller. Asus bundles its own traffic prioritization software with the X99-A II, called Turbo LAN. This utility aims to improve ping times when online games are competing with other applications for bandwidth. While packet prioritization is nice in theory, it doesn’t help if the network congestion is occurring at some point outside of the PC.
After all those words, here’s a colorful diagram to show diagram to tie it all together:
The X99-A II’s onboard audio implementation goes under Asus’ “Crystal Sound 3” moniker. The actual hardware underneath is the familiar ALC1150 codec from Realtek, backed by a TI R4580 amplifier and high-end Nichicon audio capacitors.
Asus’ audio hardware choices have given the X99-A II a good foundation. But, that’s only half the picture. Those sensitive analog audio signals still have to make their way to the rear cluster audio outputs. Thankfully, Asus has taken measures to keep these signals as noise-free as possible. A pre-regulator is placed in front of the audio codec to minimize power supply noise. Left and right output channels are split between different PCB layers. Shielding is applied to both the codec and the analog traces. The TI amplifier chip’s output can be routed to either the front or rear outputs in software. A special “de-pop” circuit should prevent popping noises through speakers and headphones during startup.
The X99-A II’s analog audio output produces sound that made my ears happy. I didn’t notice any untoward noise under a variety of load and idle conditions. For those who want digital output from the board, DTS Connect can encode multi-channel digital output in real time. Another DTS component, Studio Sound, enables surround sound virtualization for stereo speakers and headphones.
This mobo is equipped with a number of builder-friendly features. The first is a CPU installation tool that’s actually fairly nifty. Simply pop your processor into the tool, then insert that whole combo into the specially designed insertion point in the socket, lower the CPU down, close the socket latches and you’re done. The installation tool remains in the socket with the chip for easy removal. This tool makes it much less scary to mate $1,650 of CPU with the LGA 2011v3 socket.
A front-panel wiring block—Asus calls it a Q-Connector—is included with the board. This makes the finicky job of wiring up these headers so much more pleasant. It sure beats fumbling with a flashlight in a dimly-lit case.
Just above the front-panel header, we can see headers for clearing the CMOS, as well as the DirectKey firmware shortcut. DirectKey offers a convenient way to enter the UEFI by default without having to resort to furious mashing of the Del key.
For those users needing more than five fan headers, the X99-A II fully supports Asus’ fan extension card. This extension module connects to the EXT_FAN header shown above, and it provides three more four-pin fan headers and three more connectors for standard temperature probes. Thermistors attached to the fan module supply reference temperatures to the fan control intelligence managed by the board’s firmware and utility software.
A high-quality cushioned I/O shield comes with the X99-A II. This not only makes installing the motherboard in a case easier, it also may save your fingers from getting sliced in the process.
In the center of the board, all the way at the bottom is a two-digit diagnostic display that shows debug codes when the system boots. This readout can be handy if you’re trying to solve problems that occur very early in the boot process. To the left of this display, we find onboard buttons for power and reset. These can be very handy when operating the board outside of a chassis.
Sandwiched between the serial port header and the onboard buttons is a 4-pin RGB header. LED lighting strips can be connected here to tie them into Asus’ Aura lighting system. By default the LEDs glow a bright white in a slow pulsing fashion, but both the color and the nature of the pulsing can be tweaked in both the firmware and Asus’ bundled software.
Despite mobo makers getting big into the bling, the X99-A II has a handful of LEDs that serve a higher purpose than impressing your friends. Tucked away between the 24-pin ATX power connector and the DIMM slots are a series of individual LEDs that show the status of key components during the boot process: CPU, memory, GPU, and boot device. If an issue is encountered with one of these devices during boot, its LED will stay lit. While this is somewhat superfluous given the board’s two-digit display, these LEDs can provide an immediate diagnosis without having to open up the user manual to look up boot codes.
One last handy feature to touch on is Asus’ MemOK! button. Should the board be populated with unsupported RAM, pressing the MemOK! button next to the DIMM slots will make the board automatically cycle through memory profiles until it finds one that works.
Having reached the end of the hardware portion of the X99-A II tour, now is a good time to visit the restroom, stretch your legs, and grab a snack before we dive into the the board’s softer side.
Tweaking via the firmware
The X99-A II’s firmware is pretty much the same as what’s available on Asus’ Z170 boards, itself a slightly updated version of what’s found on the company’s previous X99 boards. Since our Z170-A review already provides a detailed tour, we won’t spend too much time rehashing it here. There are, however, a few points that are worth mentioning.
Unlike some of Asus’s previous boards, the X99-A II’s firmware silently takes some liberties with the CPU’s Turbo multipliers when an XMP profile is engaged. In our case, when we enabled the DDR4-2800 XMP profile the firmware took that as an open invitation to overclock our Core i7-6950X. And, not by a small amount either—the firmware wanted to run our processor at 4.0GHz with all cores engaged—600MHz higher than the stock Turbo speed for all-core loads. With 10 Broadwell cores humming along at that speed, the user would see a tremendous increase in power draw. Thankfully, the UEFI’s exit screen shows the changes that have been made during that tweaking session. Without this the user would have no idea the “free” performance they’d just been given.
Apart from this one annoyance, the X99-A II’s firmware is exactly what we’ve come to expect from Asus: an intuitive interface with an advanced mode that provides a wealth of configuration options, and excellent fan-speed controls.
With Broadwell-E support, X99 boards need a way to show the user which CPU core Intel’s characterization process has found to have the highest potential frequency overhead. In the X99-A II’s firmware, this is the core that is marked by an asterisk. For our sample Core i7-6950X, the lucky winner is core 8, as shown above. This will be the core that the Turbo Boost Max 3.0 Windows utility will push single-threaded applications to run on.
Tweaking with software
Firmware interfaces have come so far since the early days of overclocking—dramatically so since UEFI-based implementations became the standard. That said, some users may prefer the convenience of tweaking from within Windows. Asus’ AI Suite steps in to take on this challenge.
Similar to the firmware, AI Suite on the X99-A II shares a lot with its Z170-based brethren. The automatic overclocking mechanisms are the same, and tweaking options, including fan speed controls are too. You’ve simply got more cores to play with. As such, rather than rehashing all the details here, I’ll point you to our coverage from the Z170-A review.
One thing worth pointing out, however, is that core numbering in AI Suite starts at one, while core numbering in the firmware and in Intel’s utility starts at zero. Clearly AI Suite’s software engineers hail from a Fortran past.
We’re all now well accustomed to the fact that a given processor’s top stable frequency is mostly determined by the limitations of that particular chip—also known as the silicon lottery—and the CPU cooler one straps on top. This latter point is most certainly true for overclocking Broadwell-E parts, especially the top-end 10-core part, where the cooler will have to dissipate an enormous amount of heat when we shoot for high frequencies. Where the motherboard comes in to play is mostly around how easy it makes the journey from stock frequencies to the blistering heights of overclocking success.
We tested the X99-A II’s overclocking prowess using a Core i7-6950X CPU with Cooler Master Nepton’s 240M strapped to it. The Nepton has a 240-mm radiator, and before it was superseded by the company’s MasterLiquid Pro 240, it had a $110 asking price. This puts it right in the ballpark for coolers one might expect to see in a system built around the X99-A II. As I kicked off the overclocking testing, I wondered how our Nepton would fare in the face of such a beastly processor.
AI Suite got the first crack at boosting our clock speeds. We configured AI Tuner for encoding stability, told it to use AVX instructions during the stress test, and enabled the memory stress test option. After the obligatory warning message came up the system rebooted. We were off to the races.
The software started us out by testing our fastest core, labeled core 9 in AI Suite, at 4.2GHz and all other cores at 3.9GHz. After a single core test and then an all-core test, it proclaimed victory and increased clock speeds by 100MHz. This continued until we were testing core 9 at 4.6GHz and all others at 4.3GHz. Stress testing this configuration ultimately led to a blue screen.
After the reboot auto tuning completed, and we were sitting at a 45x Turbo multiplier for core 9 and 42x for all others, with the base clock left at its default 100MHz. Firing up our Prime95 stress test didn’t fare so well, however—we saw thermal throttling rear its ugly head almost instantly. Clearly with all cores humming along at 4.2GHz under a core voltage of 1.369V, our Nepton just couldn’t keep up.
Next up to the plate was the firmware’s EZ Tuning Wizard. We told it that we primarily used the system for playing games and encoding media, and that we had a liquid cooler. A quick reboot later, and we were sitting at just under 3.9GHz with a 38x Turbo multiplier on all cores (save for core 9, which was set to 42x), for a net result of just under 4.3GHz. The base clock had been bumped to 102MHz, and the core voltage was set to 1.325V. This configuration proved stable in our Prime95 stress test, with no errors nor warnings during the run. CPU temperatures peaked at 91° C, though. Not high enough to induce any thermal throttling, but not exactly comfortable for the processor. A liquid cooler with a 280-mm radiator might be a better choice for this baddest Broadwell-E chip.
With all of the automatic overclocking options out of the way, it was time to see what some manual tweaking in the firmware could do. Using multiplier tweaking alone, with the voltages on their “auto” defaults, we made it all the way to a stable 3.9GHz. At this speed, the firmware supplied 1.235V to the CPU, and temperatures maxed out at 93° C under our Prime95 load.
Moving to 40x all-core multipliers wasn’t successful. At this speed, the firmware supplied our processor with 1.27V, and within a few minutes of running Prime95 core temperatures were hitting 100°C and we were seeing thermal throttling. It was time to set voltages manually. Using an offset in the firmware, we were able to set a core voltage of 1.24V. This proved to be perfectly stable during our Prime95 stress test. Temperatures peaked at 96°C, and we saw no signs of thermal throttling.
Having conquered the 4GHz barrier—with the vector units on all 10 cores cranking away on 256-bit fused multiply-add instructions—it was time to get greedy. 4.1GHz was in our sights! A quick trip back into the firmware, and we were looking at 20 threads of Prime95 chugging along at 4.1GHz, all on the same 1.24V core voltage. Temperatures rose a few degrees, to 98° C, but we avoided thermal throttling by the barest of margins.
The limits of our slice of Broadwell-E silicon squashed our dreams when we attempted to push to higher speeds. Prime95 instantly produced errors on a number of threads when we tried for 4.2GHz. And sadly, our Nepton couldn’t keep temperatures in check when the core voltage was increased.
Our first Broadwell-E overclocking adventure, therefore, ended at 4.1GHz. To get another data point, I overclocked this same CPU/cooler combo on Gigabyte’s X99-Gaming 5P. It also managed 4.1GHz for this same stress test. Two samples doesn’t exactly make a pattern, but so far it looks like the X99-A II is not leaving anything on the table for Intel’s Core i7-6950X.
After all that excitement, it was time to try our hand at base clock overclocking. Unlike the flexibility afforded by the newer Z170 platform, Intel’s X99 chipset supports fixed base clock straps that can be used when overclocking the base clock. These straps act as a reduction gear for the DMI and PCI interfaces to ensure that they remain in spec. This allows overclockers to run the base clock at 125, 166, or 250MHz without messing with the chipset and peripheral links.
Setting the CPU strap to 1.25 in the firmware and leaving all other config options on “auto” gave us a 32x Turbo multiplier for a final CPU frequency of 4GHz.
As expected, given our previous overclocking endeavors, at 4.0GHz the firmware was supplying our CPU with 1.27V. This was too much for the Nepton 240M riding atop our Core i7-6950X. Lowering the core voltage to 1.24V, as we did before, gave us a stable Prime95 stress test run at that 4.0GHz clock speed. Attempting to use the 166MHz or the 250MHz base clock straps left us with a board that failed to boot. Thankfully the firmware caught the error and we weren’t forced to clear the CMOS.
It’s also worth pointing out that X99 boards supporting Broadwell-E, like the X99-A II, have a firmware option to set a negative offset for the Turbo multiplier that is applied when AVX workloads are run. This gives us some added flexibility when overclocking by letting us run the process twice: once for workloads that execute AVX instructions, and once for those that don’t. Once all that’s done, you can simply set your desired negative offset to account for the fact that AVX workloads can’t hit the same peak clock speeds.
Overclocking our Broadwell-E CPU was a very smooth process on the X99-A II. Manually tweaking clock speeds and voltages was a breeze in Asus’ firmware. The board’s auto-overclocking functionality gave us mixed results, however. The firmware-based method provided a stable, but relatively conservative, result that could prove to be a good starting point for manual tuning. AI Suite, on the other hand, gave an overly optimistic result that induced thermal throttling.
Now that we’ve had our fun testing the limits of our Core i7-6950X and Nepton cooler, it’s time to turn our attention towards the X99-A II’s performance.
Since so much former chipset functionality now resides on the CPU die, and since there are only a handful of third-party peripheral controllers out there, we rarely see meaningful performance differences between motherboards these days. That said, we still test system performance with different motherboards, if for no other reason than to ensure everything is functioning correctly.
When it comes to testing motherboard performance, we’ve usually gathered benchmark results using the CPU’s peak stock memory multipliers. Since DDR4 is still relatively new, however, and Broadwell-E’s 2400MHz maximum stock DDR4 speed is somewhat conservative, we’ve continued what we started with our previous X99 reviews for Haswell-E and tested this crop of Broadwell-E-supporting X99 boards with the memory clocked at the highest speed we can attain while keeping the CPU at its stock clocks.
We decided to test the X99-A II against Gigabyte’s X99-Gaming 5P that I reviewed this time last year. A simple firmware update was all that it took to get the Gaming 5P ready for Broadwell-E. Both boards were able to clock our DDR4 DIMMs up to 2800MHz while maintaining stock CPU clocks, so the results below were gathered with these settings.
The X99-A II manages to pull out a small, but measurable, win in the 99th percentile frame time testing for DiRT Showdown. For all the other tests, the two boards put in remarkably similar performance results, with the differences falling squarely within the run-to-run variance for the tests.
Boot time testing is another matter, however. Here we see the X99-A II taking over ten seconds longer to go from power-on to the Windows 8.1 desktop than the Gigabyte contender. That said, modern operating systems have perfectly functional sleep and hibernate modes, too, so boot times are much less of an issue now.
Unlike performance results, one’s choice of motherboard can have a notable impact on power consumption. We measured total system power draw (sans monitor and speakers) at the wall socket with our test system idling for a period of five minutes in the Windows desktop, and then under a full load combining Cinebench rendering with the Unigine Valley demo.
Like it’s predecessor, the X99-A II is more power-hungry at idle than the competition. Enabling the Asus EPU drops power consumption by 1W, but that still leaves it consuming 7W more than the Gigabyte board. This difference shrinks to a mere 3W under load, however. Asus’ X99 boards have a history of higher power draw, so the results aren’t terribly surprising. At least the deltas are small enough that cooling requirements shouldn’t be affected.
The following page is loaded with detailed motherboard specifications, system configurations, and test procedures. For those who take the path less traveled by not skipping straight to the conclusion, thank you. I’m yet to automate the generation of those tables. You’ll also get to see pictures of the test components inside and out of the Antec P380 case we use to house our test rigs.
We’ve already gone over the X99-A II’s most important details, but for completeness, here’s the full spec breakdown.
|Platform||Intel X99, socket LGA2011-v3|
|DIMM slots||8 DDR4, 128GB max|
|Expansion slots||3 PCIe 3.0 x16 via CPU
1 PCIe 2.0 x16 via X99 (x1, or x4 when second PCIe 2.0 x1 and USB 3.1 are disabled)
2 PCIe 2.0 x1 via X99 (second one shares bandwidth with above PCIe 2.0 x16
|Storage I/O||1 M.2 type 2242-22110 via CPU (PCIe 3.0 x4, shared with U.2)
1 U.2 via CPU (PCIe 3.0 x4 NVMe, shared with M.2)
1 SATA Express via X99
6 SATA RAID 6Gbps via X99
4 SATA 6Gbps via X99
|Audio||8-channel HD via Realtek ALC1150
Real-time digital encoding via DTS Connect
Surround virtualization via DTS Studio Sound
|Ports||1 PS/2 keyboard/mouse via nuvoton NCT6791D Super I/O
1 USB 3.0 via X99 (supports USB BIOS Flashback)
3 USB 3.0 via ASMedia ASM1074 hub chip connected to X99
4 USB 2.0 via X99
2 USB 3.1 (1 Type A and 1 Type C) via ASMedia ASM1142
4 USB 3.0 via internal headers and X99
4 USB 2.0 via internal headers and X99
1 Gigabit Ethernet via Intel I218-V
1 Serial/COM via internal header and nuvoton NCT6791D Super I/O
1 analog microphone in
4 configurable analog ports
1 digital S/PDIF out
|Overclocking||All/per-core Turbo multiplier: 12-80X
AVX instruction core ratio negative offset: 1-31x
Min. CPU cache ratio: 12-80X
Max. CPU cache ratio: 12-80X
Base clock: 80-300MHz
CPU strap: 100, 125, 167, or 250MHz
Base clock frequency to memory frequency ratio: 100:100 or 100:133
Memory frequency: 800-4000MHz
CPU core voltage: 1.0-2.0V (0.003125V increments)
CPU cache voltage: 1.0-2.0V (0.003125V increments)
CPU system agent voltage: 0.8-2.0V (0.003125V increments)
CPU input voltage: 0.8-2.70V (0.01V increments)
DRAM voltage: 0.8-1.90V (0.01V increments)
PCH core voltage: 0.7-1.80V (0.00625V increments)
PCH I/O voltage: 1.20-2.20V (0.00625V increments)
VCCIO CPU reference voltage: 0.7-1.80V (0.00625V increments)
VCCIO PCH reference voltage: 0.7-1.80V (0.00625V increments)
VTTDDR termination voltage: 0.2-1.0V (0.00625V increments)
|Fan control||1 x CPU (DC and PWM), 1 CPU_OPT (DC and PWM), 1 W_PUMP,
2 x SYS (DC and PWM), 1 H_AMP (DC and PWM, up to 3A)
Predefined silent, standard, and turbo speed profiles
Manual profile with three temp/speed points per fan
The PCIe configurations are for 40-lane CPUs. With a 28-lane chip, the lane assignments are x16/x8/x0 or x8/x8/x8.
Our testing methods
As promised, here are the test components we used:
While performance testing and overclocking was carried out on an open-air testbed, we also installed our rig inside Antec’s P380 full tower case. You can read Jeff’s review of the case here.
Once assembled and powered on, here’s what the test system looked like:
We used the following components in our testing:
|Processor||Intel Core i7-6950X|
|Cooler||Cooler Master Nepton 240M|
|Motherboard||Asus X99-A II||Gigabyte X99-Gaming 5P|
|Platform hub||Intel X99|
|Audio||Realtek ALC1150||Creative Sound Core3D (CA0132)|
|Memory size||16GB (4 DIMMs)|
|Memory type||Corsair Vengeance LPX DDR4 SDRAM at 2800MHz|
|Graphics||Sapphire Radeon HD 7950 Boost with Crimson Edition 16.6.2 Hotfix drivers|
|Storage||OCZ ARC 100 120GB|
|Power Supply||Cooler Master V750 Semi-Modular|
|Operating System||Microsoft Windows 8.1 Pro x64|
Thanks to Antec, Cooler Master, Corsair, Intel, and OCZ for providing the hardware used in our test systems. We also extend thanks to Asus and Gigabyte for providing our test boards.
We used the following versions of our test applications:
- 7-Zip 9.20 64-bit
- TrueCrypt 7.1a
- Chrome 40.0.2214.115
- x264 r2431
- DiRT Showdown demo
- Fraps 3.5.99
- Cinebench R15
- Unigine Valley 1.0
Some further notes on our test methods:
- All testing was conducted with motherboard power-saving options enabled. These features can sometimes lead to slightly slower performance, particularly in peripheral tests that don’t cause the CPU to kick into high gear. We’d rather get a sense of motherboard performance with real-world configurations, though; we’re not as interested in comparing contrived setups with popular features disabled.
- DiRT Showdown was tested with ultra detail settings, 4X MSAA, and a 1920×1200 display resolution. We used Fraps to log a 60-second snippet of gameplay from the demo’s first race. To offset the fact that our gameplay sequence can’t be repeated exactly, we ran this test five times on each system.
- Power consumption was measured at the wall socket for the complete system, sans monitor and speakers, using a Watts Up Pro power meter. The full-load test combined Cinebench’s multithreaded CPU rendering test with the Unigine Valley DirectX 11 demo running with extreme settings in a 1280×720 window. We reported the peak power consumption during the Cinebench run. Our idle measurement represents the low over a five-minute period sitting at the Windows desktop.
- Our system build was done using all of the hardware components listed in the configuration table above. Completing this process as our readers would allows us to easily identify any pain points that arise from assembling a system with this particular motherboard.
The tests and methods we employed are publicly available and reproducible. All tests except power consumption, were run at least three times. Unless otherwise indicated, we reported the median result for each test. If you have questions about our methods, hit our forums to talk with us about them.
Asus’ X99-A II continues the fine legacy set forth by its predecessor. The board still offers everything we want for a high-end desktop build and nothing we don’t. The main difference is that this time around, it’s targeted at those with an eye for Broadwell-E CPUs rather than Haswell-E. What’s even sweeter is how it comes in at a lower price point than the original X99-A. You can nab yourself one right now for $230 online.
Beyond my oft-repeated complaint with regards to motherboard firmwares that silently take liberties with Turbo multipliers, there’s very little to dislike about the X99-A II. Asus has taken the original X99-A and injected some modern hotness, all without tampering with the successful formula that made the first iteration of the board one of our favorites.
We get USB 3.1 Type-A and Type-C ports, and a U.2 connector for 2.5″ NVMe SSDs, all the while still retaining the M.2 slot, the high-end Realtek ALC1150 audio implementation, some thoughtful builder-friendly perks, and excellent overclocking features and fan controls. And let’s not forget the all-important RGB LED lighting. That’s a must-have addition for any board with claims to the high end in 2016.
Perhaps the only fly in the ointment is that the older X99-A/USB 3.1 is currently selling for $210 online. That older board lacks USB 3.1 Type-C ports, though, and it also doesn’t have some of the niceties the X99-A II does, like dedicated fan headers for liquid-cooling hardware. Those considerations might swing the pendulum in the newer board’s favor for most builders. That said, the X99-A/USB 3.1 is probably only an option while supplies last. The X99-A II will presumably supplant the older offering in the market with time.
All told, I think the X99-A II offers as smooth a ride as one could want when pairing a board with one of Intel’s Broadwell-E CPUs, even the range-topping Core i7-6950X. We haven’t had a ton of experience with refreshed X99 boards yet—something we hope to rectify in time—but Asus’ value-priced entry in its X99 refresh lineup has impressed us. Making the right revisions to a winning formula without goofing up the things that made a product great in the first place is a fine line to walk, and Asus’ deft handling of that challenge makes the X99-A II an easy pick for another Editor’s Choice award.