ASUS X99-PRO Haswell-E Motherboard Review

[SKYMTL]

Included Software pt.2

Included Software pt.2

ASUS CPU-Z

Click on image to enlarge​

ASUS CPU-Z is a special edition of CPU-Z especially created to match the aesthetics of ASUS' other software utilities. This edition is found on the included software DVD or the ASUS website, but it is not yet available for download from CPUID.com, though you can download the ROG CPU-Z version.

Turbo LAN

Click on image to enlarge​

Turbo LAN is a utility designed to help reduce latency courtesy of cFosSpeed traffic-shapping technology. This utility provides users with a lot of control and monitoring capabilities over every application that is accessing the network. It displays CPU usage, NPU usage, ICMP and UDP average ping, and the network utilization of every system process and program. This tool also allows you give priority to certain applications, and throttle or block others to free network resources for other applications. It is your one-stop tool for monitoring and controlling all network traffic, and it even comes with a little widget for real-time bandwidth information.

Boot Setting

Click on image to enlarge​

ASUS Boot Setting allows users to boot directly into the BIOS without having to repeatedly hit delete during the POST screen. It is a pretty hand tool when you are rebooting as often as overclockers tend to do.

WebStorage

Click on image to enlarge​

The WebStorage utility is basically the ASUS equivalent of DropBox. It is cloud computing application that gives users web storage and access to data across many devices. All ASUS motherboard owners get a free 5.5GB of storage, you can buy more or be gifted some by ASUS if you referrer your friends. The web interface is pretty standard and utilitarian. Overall, there is not much to complain about, it's a nice freebie if you choose to use it.

 

[SKYMTL]

Test Setups & Methodology

Test Setups & Methodology

For this review, we are going to be testing the performance of the ASUS X99-PRO in three configurations: default settings, automatic overclock settings, and manual overclock settings. The components and software are the same across all three, and aside from manually selecting the frequencies, timings, and voltages in the manual overclock configuration, every option in the BIOS was at its default setting.

Intel Core i7 LGA2011-v3 Haswell-E Test Setup​

For all of the benchmarks, appropriate lengths are taken to ensure an equal comparison through methodical setup, installation, and testing. The following outlines our testing methodology:

A) Windows is installed using a full format.

B) Chipset drivers and accessory hardware drivers (audio, network, GPU) are installed.

C)To ensure consistent results, a few tweaks are applied to Windows 7 and the NVIDIA control panel:

• UAC – Disabled
• Indexing – Disabled
• Superfetch – Disabled
• System Protection/Restore – Disabled
• Problem & Error Reporting – Disabled
• Remote Desktop/Assistance - Disabled
• Windows Security Center Alerts – Disabled
• Windows Defender – Disabled
• Screensaver – Disabled
• Power Plan – High Performance
• V-Sync – Off

D) All available Windows updates are then installed.

E) All programs are installed and then updated, followed by a defragment.

F) Benchmarks are each run three to eight times, and unless otherwise stated, the results are then averaged.

Here is a full list of the applications that we utilized in our benchmarking suite:

• 3DMark Vantage Professional Edition v1.1.0
• 3DMark11 Professional Edition v1.0.132.0
• 3DMark 2013 Professional Edition v1.2.362
• AIDA64 Extreme Edition v3.00.2536 Beta
• Cinebench R11.529 64-bit
• SuperPi Mod v1.9 WP
• MaxxMEM² - PreView v2.01
• Sisoft Sandra 2014.SP3 20.28
• Valve Particle Simulation Benchmark v1.0.0.0
• wPRIME version v2.10
• X3: Terran Conflict Demo v1.0

That is about all you need to know methodology wise, so let's get to the good stuff!

 

[SKYMTL]

Feature Testing: SATA Express & M.2 Results

Feature Testing: SATA Express & M.2 Results

Back in April, in our ASUS 9-Series Preview article, we mentioned that there were basically zero SATA Express devices ready to hit the market any time soon. Since then the situation hasn't really changed as much as we would have liked, but during the recent CES 2015 a few companies did show off a few functional prototypes. Thankfully for us, ASUS has ASMedia - storage controller experts - as one of its subsidiaries and they were able to whip together a pretty neat SATA Express storage device in the form of the Hyper Express enclosure. We figured that it would be interesting to show you what this new high-speed interface was capable of.

Click on image to enlarge​

The unit that ASUS provided us came packed with two Kingston SSDNow mS200 120GB mSATA solid state drives. These are based on the LSI SandForce SF-2241 controller and are rated at an impressive 550MB/s read and 520MB/s write speeds. Now this is not a review of this device, since this is still a pre-production product and you will not be able to buy one pre-assembled with SSDs inside. We just want to show you a little bit of what SATA Express is capable of.

As you have probably know, most current SATA 6Gb/s devices struggle to get anywhere near that interface's theoretical 750MB/s limit. Due to overhead you are realistically looking at real-life transfer rates of up to about 550 to 575MB/s. By comparison, at the moment all 9-series chipsets seem to be limited to a 10GB/s interface that tops out at about 1GB/s of bandwidth, but we will eventually see SATA Express 16Gb/s implementations capable of supporting transfer rates of up to 2GB/s.

With all of this in mind, let's take a peak at the results.

SATA 6Gb/s vs. SATA Express - Click on image to enlarge​

As you can see above, a modern solid state drive is capable of about 535-550MB/s. Often this is not a controller limitation, but an interface one. If you were to combine one of these modern SATA controllers with a faster interface, the results could be way above SATA 3's limits. That is what SATA-Express is promising to do. Not only do you get up to 1GB/s of bandwidth but there is a built-in backward compatibility with current SATA devices.

With the Hyper Express enclosure, we were able to get very close to the 740MB/s mark. This is a limitation with this configuration of our pre-production Hyper Express, but it at least gives a small glimpse at what we can expect from future SATA Express devices. There is no doubt that sometime in the (hopefully) not too distance future there will be devices that are fully capable of utilizing this interface's 1GB/s of bandwidth.


Next up, let's take a quick look at the M.2 interface:

We reviewed the Plextor M6e 256GB PCIe SSD back in May, and today we are going to be using the M.2 module (otherwise sold as the PX-G256M6e) from that product to test out this motherboard's M.2 interface. To recap, this M.2 SSD is based on a server-grade Marvell 8SS9183 controller and features eight Toshiba 19nm Toggle Mode NAND ICs. This M.2 SSD drive only requires two PCI-E 2.0 lanes - in otherwords it is based off an M.2 x2 interface - so it can't take full advantage of this motherboard M.2 x4 interface. Nevertheless, we can still give you a glimpse of what this high-speed interface is capable of.

What that out of the way, here are some quick results:

SATA 6Gb/s vs. M.2 - Click on image to enlarge​

This Plextor M.2 SSD boasts maximum transfer rates of 770MB/s for sequential read and 580MB/s for sequential writes. As you can see from these results, we are reaching a little over 780MB/s read and almost up 590MB/s write, so clearly both the drive and interface are working perfectly. Obviously this doesn't come close to saturating this motherboard's M.2 x4 interface which is capable of 4GB/s transfer rates, but we will see seeing some truly fast M.2 offerings soon.

 

[SKYMTL]

Feature Testing: Onboard Audio

Feature Testing: Onboard Audio

Since fewer and fewer consumers seem to be buying discrete sound cards, the quality of a motherboard's onboard audio is now more important than ever. We figured that it was worthwhile to take a closer look at just how good the analog signal quality is on this new X99-PRO.

Since isolated results don't really mean much, but we have also included some numbers from the ASUS Rampage V Extreme, GIGABYTE X99-Gaming G1 WIFI, MSI X99S Gaming 7, EVGA X99 Classified, and ASUS X99 Deluxe motherboards that we have recently reviewed. This X99-PRO model, the X99-Deluxe, RoG Rampage, and the MSI all feature onboard audio solutions that are built around the familiar Realtek ALC1150 CODEC, but feature different op-amps, headphone amplifiers, filtering capacitors, secondary components and layouts. The GIGABYTE and EVGA are both based on the same Creative Core3D CA0132 quad-core audio processor, but feature vastly different hardware implementations.

We are going to do this using both quantitative and qualitative analysis, since sound quality isn't really something that can be adequately explained with only numbers. To do this we have turned to the RightMark Audio Analyzer, basically the standard application for this type of testing.

Since all the three motherboards support very high quality 24-bit, 192kHz audio playback we selected that as the sample mode option. Basically, what this test does is pipe the audio signal from the front-channel output to the line-in input via a 3.5mm male to 3.5mm male mini-plug cable, and then RightMark Audio Analyzer (RMAA) does the audio analysis. Obviously we disabled all software enhancements since they interfere with the pure technical performance that we are trying to benchmark.

Click on image to enlarge and reveal additional motherboards​

As you can see, from a technical standpoint this model puts up essentially identical numbers to the higher-end X99-Deluxe. We had an inkling that this would be the case, since their onboard audio designs are quite similar, but we are glad to see the numbers back that up.

Based on our listening period with a mix of Grado SR225i and Koss PortaPro headphones, Westone UM1 IEMs, and Logitech Z-5500 5.1 speakers, we are again pleased to report that the sound quality was without fault to our ears. As we tend to repeat, we aren't experts in this area, but we suspect that your average user will likewise be perfectly satisfied with this motherboard's onboard audio capabilities.

 

[SKYMTL]

Auto & Manual Overclocking Results

Auto & Manual Overclocking Results

It wouldn't be an HWC review if we didn't include some overclocking results, so we thoroughly tested out this motherboard's capabilities, especially its auto-overclocking functionality. Though it features a new chipset, the X99-PRO is still fundamentally an LGA2011 motherboard, and as a result there is nothing new to report on how to overclock on this motherboard, but if you want any insights check out the overclocking section in our review of the Core i7-5960X.

Auto Overclocking

The X99-PRO features two types of automatic overclocking. Within the UEFI BIOS you can find the OC Tuner feature. This automatic overclocking method features two main options, a multiplier-only tuning mode or a multiplier and BCLK tuning mode. Although it is very fast - the time it takes to save & exit the BIOS - OC Tuner is based on presets and as such it produces much less impressive results than an intelligent software-based approach. Thankfully, one is included. Within the Ai Suite III utility there is the usual TPU automatic overclocking feature that is part of the new 5-Way Optimization scheme. This method features two options (Fast Tuning & Extreme Tuning) as well as two different methods for achieving that overclocking (Ratio Only & BCLK First). This is an intelligent auto OC feature though, it does not rely on presets. As a result, it will slowly increase your processor's clock speed and voltage, test for stability, and repeat until it has found the sweet spot. In this new version of the software, you can specify a clock speed to start from and even what maximum temperature you feel comfortable with.

Click on image to enlarge​

First up, we have the Ratio Tuning multiplier-only tuning mode. Its application was extremely quick - the time it takes to save & exit the BIOS - and it produced decent enough results. This preset increased the CPU multipliers to 39X across all the cores, resulting in a static 3.9GHz clock speed. It's nothing too special, but it is a solid increase over the i7-5960X's stock 3.0Ghz to 3.5Ghz range. Although it is a fairly moderate overclock, the core voltage increase is thankfully quite reasonable as well. We are pleased to see a slight bump in the memory speed, from the default DDR4-2133 up to DDR4-2400.

Click on image to enlarge​

The second OC Tuner mode utilizes both the CPU multiplier and the BCLK to achieve its overclock. In our case, our i7-5960X was pushed to 3.85Ghz - 50Mhz less than the multiplier-only mode - but with an almost 400Mhz Uncore/Cache overclock. Regrettably, it only overclocked our memory modules from DDR4-2133 to DDR4-2333, whereas it set a more impressive DDR4-2666 on the X99-Deluxe. Nevertheless, even with these various difference the performance difference between both modes is quite small.

Overall, given how quick and easy it is to engage OC Tuner, consider us impressed with its current implementation. However, obviously we want even better results, and that is were the software-based TPU approach comes into play. Let's see what it can do.

Click on image to enlarge​

Once again, we were left impressed by the capabilities of the TurboV Processing Unit (TPU) and the 5-Way Optimization feature. This second auto-overclocking feature is software and hardware based, and it does not utilize presets. As a result, the process does take quite a bit longer (5-10 minutes) and requires a few automatic system reboots. Within Windows, the TPU slowly increases the system frequencies and does some stress testing at each level until it finds the limit, BSODs, reboots, and voila! The overclock is set.

As we've come to expect from ASUS, they have improved this piece of software since our last time using it with the X99-Deluxe, and as a result it was able to achieve an even more impressive automatic overclock. We allowed the software to attempt a 'Per Core' overclock which configured the processor to run at 4.5Ghz with workloads up to eight threads, and at 4.4Ghz when more than eight threads are being utilized. This overclock is exceedingly impressive and basically matches our manual overclock. Having said, as you will see below, our in-house overclock is quite a bit more strenuous due to the heavy Uncore and memory speed increases. The fact that this 'intelligent feature' still didn't touch the Uncore and only slightly increased the memory speed is still a bit disappointing, but nevertheless this is still the best automatic overclocking implementation on the market right now.

Manual Overclocking

Click on image to enlarge​

We don't have a particularly strong Core i7-5960X, and we know that our chip doesn't have a lot of headroom above 4.4Ghz @ 1.30V - at least not when all cores are loaded and without a serious vCore increase - so we focused on increasing performance by significantly boosting the Uncore and memory frequency. Just like we experienced on the X99-Deluxe and Rampage V Extreme, our manual overclocking effors proved very succesful on this motherboard. Perhaps as importantly, it was really easy too. First and foremost, we enabled the XMP profile that was programmed into our 16GB G.Skill RipJaws 4 DDR4-3000 15-15-15-35 kit. That set a 3.5Ghz CPU core clock (28x125Mhz), 3000Mhz Uncore, and DDR4-3000 memory frequency. From there, it was only a matter slowly upping the BCLK and adjusting the system voltages as needed. We achieved a 4.29Ghz Uncore by increasing the cache voltage to 1.35V and the system agent voltage to 1.20V. We kept the memory modules at their manufacturer specified 1.35V, but we were able to push them to DDR3-3118 at their stock timings. To get all of this stable, we did need to increase the core voltage by an additional 0.05V above our preferred 1.30V but this was not unexpected since we were pushing so many different parts of the processor way out of spec.

Overall, as we have come to expect from ASUS - especially with this enthusiast-oriented LGA2011-v3 platform - overclocking on this motherbard was a problem-free experience and we never once felt that the motherboard was the component holding us back. They have simply done an unparalleled job at optimizing the BIOS. The automatic overclocking results speak for themselves, they are the best we've ever seen on any X99 motherboard.

 

[SKYMTL]

System Benchmarks

System Benchmarks

In the System and Gaming Benchmarks sections, we reveal the results from a number of benchmarks run with the Core i7-5960X and ASUS X99 Deluxe at default clocks, with the TPU Extreme Tuning preset applied, and using own our manual overclock. This will illustrate how much performance can be achieved with this motherboard in stock and overclocked form. For a thorough comparison of the Core i7-5960X versus a number of different CPUs have a look at our Intel Haswell-E Core i7-5960X Review.

SuperPi Mod v1.9 WP

When running the SuperPI 32MB benchmark, we are calculating Pi to 32 million digits and timing the process. Obviously more CPU power helps in this intense calculation, but the memory sub-system also plays an important role, as does the operating system. We are running one instance of SuperPi Mod v1.9 WP. This is therefore a single-thread workload.

wPRIME 2.10

wPrime is a leading multithreaded benchmark for x86 processors that tests your processor performance by calculating square roots with a recursive call of Newton's method for estimating functions, with f(x)=x2-k, where k is the number we're sqrting, until Sgn(f(x)/f'(x)) does not equal that of the previous iteration, starting with an estimation of k/2. It then uses an iterative calling of the estimation method a set amount of times to increase the accuracy of the results. It then confirms that n(k)2=k to ensure the calculation was correct. It repeats this for all numbers from 1 to the requested maximum. This is a highly multi-threaded workload.

Cinebench R11.5

Cinebench R11.5 64-bit
Test1: CPU Image Render
Comparison: Generated Score

The latest benchmark from MAXON, Cinebench R11.5 makes use of all your system's processing power to render a photorealistic 3D scene using various different algorithms to stress all available processor cores. The test scene contains approximately 2,000 objects containing more than 300,000 total polygons and uses sharp and blurred reflections, area lights and shadows, procedural shaders, antialiasing, and much more. This particular benchmarking can measure systems with up to 64 processor threads. The result is given in points (pts). The higher the number, the faster your processor.

Sandra Processor Arithmetic and Processor Multi-Media Benchmarks

SiSoftware Sandra (the System ANalyser, Diagnostic and Reporting Assistant) is an information & diagnostic utility. The software suite provides most of the information (including undocumented) users like to know about hardware, software, and other devices whether hardware or software. The name “Sandra” is a (girl) name of Greek origin that means "defender", "helper of mankind".

The software version used for these tests is SiSoftware Sandra 2013 SP4. In the 2012 version of Sandra, SiSoft has updated the .Net benchmarks and the GPGPU benchmarks have been upgraded to General Processing (GP) benchmarks, able to fully test the new APU (CPU+GPU) processors. The two benchmarks that we used are the Processor Multi-Media and Processor Arithmetic benchmarks. These three benchmarks were chosen as they provide a good indication of three varying types of system performance. The multi-media test shows how the processor handles multi-media instructions and data and the arithmetic test shows how the processor handles arithmetic and floating point instructions. These two tests illustrate two important areas of a computer’s speed and provide a wide scope of results.

MaxxMem Benchmark

Created by MaxxPI², the MaxxMem benchmark tests your computer’s raw memory performance, combining copy, read, write and latency tests into one global score. This memory benchmark is a classic way to measure bandwidth of a memory subsystem.

MaxxMem uses continuous memoryblocks, sized in power of 2 from 16MB up to 512MB, starting either writing to or reading from it. To enable high-precision memory performance measurement, they both internally work with multiple passes and averages calculations per run.

Further, the main goal was to minimize (CPU) cache pollution on memory reads and to eliminate it (almost completely) on memory writes. Additionally, MaxxMem operates with an aggressive data prefetching algorithm. This all will deliver an excellent judge of bandwidth while reading and writing.

 

[SKYMTL]

Gaming Benchmarks

Gaming Benchmarks

Futuremark 3DMark (2013)

3DMark v1.1.0
Graphic Settings: Fire Strike Preset
Rendered Resolution: 1920x1080
Test: Specific Physics Score and Full Run 3DMarks
Comparison: Generated Score


3DMark is the brand new cross-platform benchmark from the gurus over at Futuremark. Designed to test a full range of hardware from smartphones to high-end PCs, it includes three tests for DirectX 9, DirectX 10 and DirectX 11 hardware, and allows users to compare 3DMark scores with other Windows, Android and iOS devices. Most important to us is the new Fire Strike preset, a DirectX 11 showcase that tests tessellation, compute shaders and multi-threading. Like every new 3DMark version, this test is extremely GPU-bound, but it does contain a heavy physics test that can show off the potential of modern multi-core processors.

Futuremark 3DMark 11

3DMark 11 v1.0.5
Graphic Settings: Performance Preset
Resolution: 1280x720
Test: Specific Physics Score and Full Run 3DMarks
Comparison: Generated Score


3DMark 11 is Futuremark's very latest benchmark, designed to tests all of the new features in DirectX 11 including tessellation, compute shaders and multi-threading. At the moment, it is lot more GPU-bound than past versions are now, but it does contain a terrific physics test which really taxes modern multi-core processors.

Futuremark 3DMark Vantage

3DMark Vantage v1.1.2
Graphic Settings: Performance Preset
Resolution: 1280x1024

Test: Specific CPU Score and Full Run 3DMarks
Comparison: Generated Score

3DMark Vantage is the follow-up to the highly successful 3DMark06. It uses DirectX 10 exclusively so if you are running Windows XP, you can forget about this benchmark. Along with being a very capable graphics card testing application, it also has very heavily multi-threaded CPU tests, such Physics Simulation and Artificial Intelligence (AI), which makes it a good all-around gaming benchmark.

Valve Particle Simulation Benchmark

Valve Particle Simulation Benchmark
Resolution: 1680x1050
Anti-Aliasing: 4X
Anisotropic Filtering: 8X
Graphic Settings: High
Comparison: Particle Performance Metric

Originally intended to demonstrate new processing effects added to Half Life 2: Episode 2 and future projects, the particle benchmark condenses what can be found throughout HL2:EP2 and combines it all into one small but deadly package. This test does not symbolize the performance scale for just Episode Two exclusively, but also for many other games and applications that utilize multi-core processing and particle effects. As you will see the benchmark does not score in FPS but rather in its own "Particle Performance Metric", which is useful for direct CPU comparisons.

X3: Terran Conflict

X3: Terran Conflict 1.2.0.0
Resolution: 1680x1050
Texture & Shader Quality: High
Antialiasing 4X
Anisotropic Mode: 8X
Glow Enabled

Game Benchmark
Comparison: FPS (Frames per Second)

X3: Terran Conflict (X3TC) is the culmination of the X-series of space trading and combat simulator computer games from German developer Egosoft. With its vast space worlds, intricately detailed ships, and excellent multi-threaded game engine, it remains a great test of modern CPU performance.

 

[SKYMTL]

Voltage Regulation / Power Consumption

Voltage Regulation

Usually we have a pretty extensive voltage regulation section, but this time we were a little wary of poking and prodding the live VRM section of a motherboard powering a new $1000 processor that we need for a bunch of other reviews. We usually throw caution to the wind, but this time there's just too much on the table to risk a worse case scenario type event. Therefore, in this abbreviated outlook, we can tell you that - based on the Ai Suite's voltage readings - this motherboard appears to have excellent regulation output. What you set in the bios seems to be exactly what the board put outs, without fault. This is likely due to the fact that it appears that there is some type of Load-Line Calibration (LLC) enabled by default once you set the Ai Overclock Tuner to Manual.

At this point, we would usually see how the vCore behaves with and without Load-Line Calibration (LLC) enabled. However, OCCT did not recognize this motherboard's vCore line, and thus could not actually monitor anything. We had to look for an alternative, and settled on the AIDA64 System Stability Test. This testing was done with a one-hour stability run and with our Core i7-5960X overclocked to 4.2Ghz at 1.30V (in the BIOS).

This only illustrates a portion of the 60 minute run, but we watched attentively throughout and there were never any spikes. The vCore line was straight as an arrow for the full hour and it never deviated from 1.300V. You can't really ask for better than that.


Power Consumption

For this section, every energy saving feature was enabled in the BIOS and the Windows power plan was changed from High Performance to Balanced. For our idle test, we let the system idle for 15 minutes and measured the peak wattage through our UPM EM100 power meter. For our CPU load test, we ran Prime 95 In-place large FFTs on all available threads, measuring the peak wattage via the UPM EM100 power meter. For our overall system load test, we ran Prime 95 on all available threads while simultaneously loading the GPU with 3DMark Vantage - Test 6 Perlin Noise.

Nothing unexpected here, the stock numbers are very much inline we what would expect from a fully featured motherboard such as this one. We aren't using any of the numerous power saving software options that ASUS offers, so there's definitely room for improvement if that's of interest to you. Once you start pumping extra voltage into that octo-core processor the power consumption starts shooting up pretty quickly, but this motherboard proved more than capable of handling the additional load.

 

[SKYMTL]

Conclusion

Conclusion

The ASUS X99-PRO is the seventh LGA2011-v3 motherboard that we have reviewed, so we have a pretty good idea of what we like and what we don't when it comes to this enthusiast platform. From the futuristic aesthetics, to the surprising accessories bundle, to the all-encompassing connectivity options, the PRO definitely appears to live up to the lofty expectations that were established by the higher-end X99-Deluxe model. At first glance the X99-PRO appears to just be a Deluxe with one less SATA Express port, better M.2 slot placement, a few less USB 3.0 ports, a smaller number of audio filtering capacitors, and a simpler wireless antenna. These minor differences would likely mean very little to most people on a day-to-day basis. However, it is the different PCI-E slot configuration that is the biggest differentiator.

When we reviewed the GIGABYTE GA-X99-Gaming G1, we took issue with fact that if you installed an M.2 SSD it would automatically disable the SATA Express port as well as a couple SATA ports. Our main argument was what good is a brand new platform with next generation storage interfaces if you're forced to choose between them. With the X99-PRO, the tough decisions that you will need to make aren't quite as drastic, but regrettably they are there, they are numerous, and they could be quite confusing for novice users.

The main issue we have is that when combined with a 28 PCI-E lane processor like the Core i7-5820K, this motherboard cannot handle a proper x16/x16 two-way CrossFire or SLI configuration. With two graphics cards installed you're looking at x16/x8 and frankly that's not sufficiently better than the x8/x8 that you get on the Z97 LGA1150 platform. Perhaps even worse, if you combine this motherboard, an i7-5820K, and three graphics cards you're looking at x16/x8/x4. Since x4 is not supported by SLI, that feature goes out the window and you're left with three-way CrossFire only.

A couple of PCI-E switches would have allowed the PRO to have a more logical x8/x8/x8 three-way breakdown, so their omission is a mystery to us. No matter the CPU type, if you use the fourth PCI-E x16 slot it will disable the M.2 x4 slot. And vice versa, obviously. This issue is somewhat alleviated by the inclusion of the Hyper M.2 x4 expansion card...but make sure that you don't install that card in the second PCI-E x16 slot or it will disable the onboard Wi-Fi/Bluetooth module. See what we mean by confusing?

Having said all of that, as we stated in the GIGABYTE review, every manufacturer has their own unique way of dividing the plentiful but finite amount of PCI-E lanes that you get on the X99 platform. Some focus on pure performance, others on maximizing connectivity and expansion, and some try to thread the middle with varying success. You really need to pay close attention, do your research, and find the design implementation that best suits your current and future needs. This X99-PRO has its quirks, but they might be a complete non-issue for some configurations (i7-5930K or i7-5960X/5820K + single GPU) and complete non-starter for others (5820K + 2-3 GPUs + M.2 SSD). Our message to i7-5820K owners with multi-GPU aspirations - and only secondary interest in overclocking - is that they should strongly consider the MSI X99S Gaming 7, since it divvies up those 28 PCI-E lanes in such a way that you can retain both three-way CrossFire/SLI and M.2 capabilities.

With all that out of the way, we still found the X99-PRO as fun, easy and rewarding to use as we did the X99-Deluxe and even Rampage V Extreme. As we revealed in the Overclocking Results section, this motherboard has some fantastic automatic overclocking capabilities. The two auto-overclocking options, one fast and preset-based while the other is slower but intelligent, differ greatly in their approaches but we can appreciate their distinct purposes. Most importantly though, our results and experiences indicate that they both work exceedingly well. The more advanced smart implementation pushed our Core i7-5960X up to 4.5GHz at 1.29V, which is actually higher than our manual overclock but not quite as holistic since it doesn't touch the Uncore and barely increases the memory frequency.

Speaking of our manual overclock, although we have long ago established our chip's frequency/voltage/heat wall at about 4.4Ghz and 1.30V, this motherboard proved quite capable of pushing the Uncore and memory frequency to very lofty levels. We managed to overclock the Uncore from its stock 3000Mhz up to 4288Mhz, an impressive 43% increase. Likewise, we were able to extract an additional 59Mhz from our excellent G.Skill DDR4-3000 memory kit, reaching DDR4-3118 15-15-15-35 without loosening any timings or increasing the voltage above the specified 1.35V. Basically, as we've come to expect from ASUS, they have mastered overclocking on this LGA2011-v3 platform, and the end result of their optimizations is that this motherboard was rock solid, never held us back, and recovered perfectly when we did go over the limit.

When reviewed the X99-Deluxe, we said that it was a motherboard that had "absolutely no compromises, no flaws at all really". The same cannot be said for this X99-PRO model. There are definitely some compromises or at least choices that prospective owners will need to make, but it is still a flawless experience. The polish that ASUS have put into the BIOS and the software is a cut above the competition, and if you are an overclocker and a tweaker - in other words anyone who's not satisfied with "default settings" - that high quality engineering work really starts to overshadow everything else.