Crucial Ballistix Elite DDR4-2666 16GB Memory Kit Review

[SKYMTL]

All the way back in June, Crucial - the consumer arm of semiconductor giant Micron - announced their Ballistix Elite DDR4 memory modules. These sticks boasted fairly standard speeds of DDR4-2666 and DDR4-3000, but what really caught our eye were the heat spreaders. Frankly, they were sleekest and most badass heat spreaders that we had ever seen, and combined with the black PCB it made for an unusually eye-catching stick of RAM. Not only that, these modules featured built-in thermal sensors that allowed for real-time temperature monitoring via a custom Ballistix utility. A fast, great-looking memory module with sensors and its own utility? Sign us up!

Following that initial announcement we heard nothing for a over 6 months, and then in mid-January we finally got a notice that the memory kits were going to start hitting the retail market, so we claimed one. Since the DDR4-3000 modules haven't yet made an appearance, we picked up the Crucial Ballistix Elite DDR4-2666 16GB memory kit, known as model BLE4K4G4D26AFEA.

This 4x4GB kit features a DDR4-2666 clock speed with 16-17-17-36 timings at 1.20V, and is backed by a lifetime warranty. These are not great timings - they are a little looser than the Corsair Vengeance LPX DDR4-2666 16GB kit that we reviewed earlier - but based on our previous testing we know that Crucial is extremely conservative when it comes to specifications and that Micron ICs are capable of some impressive feats.

Given these facts, we do have high expectations for this product, especially considering its lofty $335 USD/$410 CAD price tag. There a whole lot of DDR4-2666 16GB kits in the $225 to $250 USD range, so this Crucial model better be something special.

Click on image to enlarge​

Crucial have packaged the BLE4K4G4D26AFEA in a standard plastic clamshell with a cardboard informational insert, which is the usual packaging method for most current memory kits. Thankfully, the plastic packaging does not require a knife or scissors to open, it can be opened / pulled apart by hand, which means that it is also re-sealable. Regrettably, for some anyways, Crucial does not include a branded case sticker.

Click on image to enlarge​

For this new generation, Crucial introduced a new heat spreader design for the Ballistix Elite series. It's sleek, stealthy, and aggressive by any measure. The anodized aluminum heat spreaders have a matte finish, and the modules feature a custom-designed black PCB, so these should become the de facto choice for any 'murdered out' computer build. Thanks to that thick machined spine at the top, they also have a really nice weight to them, so they feel like a quality product.

Click on image to enlarge​

Here is just a closer look at some of the heat spreader details. As you can see, they are using actual screws to keep the two sides together, it's not just sticky tape. Also, there are cool cut-outs that slightly reveal the memory ICs the lenght of the heat spreader. Definitely a unique design touch. At 40 millimeters / 1.57 inches these are what we would consider medium height modules, they aren't low profile but they also aren't tall enough to interfere with most tower-style heatsinks.

Click on image to enlarge​

As previously discussed, this memory comes in clocked at DDR4-2666 16-17-17-36 with a default of 1.20V. Those are loose timings by any standard, but as we stated in the intro Crucial are known for selecting extremely conservative timings. They aren't trying to "wow" consumers with the specs, they are instead relying on their reputation as a well established manufacturer of top-notch memory. We will find out what this kit is truly capable of in our Overclocking Results section.

These particular modules feature three XMP 2.0 profiles, but obviously the XMP-2666 profile is the one we are most interested in. It sets this memory frequency by utilizing the 125Mhz CPU strap and selecting the DDR4-2666 memory multiplier What is particularly interesting is that this memory kit defaults to DDR4-2400 instead of the usual DDR4-2133.

On our ASUS X99-Deluxe that default was DDR4-2400 16-16-16-39. That is pretty cool because even those who are completely clueless about memory speeds, timings, or even the BIOS will still get an automatic speed bump when installing this memory kit. By the way, disregard that XMP-0 profile, it is a CPU-Z issue that we haven't spotted in other programs (like AIDA64), so it's not something that is wrong with the modules themselves.

 

[SKYMTL]

Ballistix M.O.D. Utility & Thermal Sensor

Ballistix M.O.D. Utility & Thermal Sensor

As we revealed in the introduction, this Ballistix Elite memory kit features thermal sensors built into every module that allows for real-time temperature monitoring via the custom Crucial Ballistix M.O.D. utility. Now this isn't a feature that is exclusive these Ballistix Elite DDR4 models - it was actually first introduced on their Ballistix Elite DDR3 modules all the way back in 2010 - but it still a seamless and cableless capability that is unmatched by any other memory manufacturer that we know of.

Click on image to enlarge​

The SPD Data tab is where you can find some basic information about your Crucial modules, like capacity, model number, a look at the XMP Profile settings, and a breakdown of all of the currently approved JEDEC DDR4 specifications.

Click on image to enlarge​

The Temperature tab is obviously where you will find the real-time temperature readouts. Since every module has its own thermal sensor, you can monitor each module individually. As you can see, there is a noticeable different between idle and a full sixteen-thread HyperPI load. There is also a noticeable difference between some of the modules depending on airflow and proximity to heat-producing components.

Click on image to enlarge​

The Lights tab is not applicable to this Ballistix Elite memory kit since it doesn't have any built-in LEDs. If and when Crucial releases Ballistix Tracer DDR4 memory then this section will certainly come in handy in order to choose colours, select between different LED patterns, adjust brightness, and even turn off the LEDs.

Click on image to enlarge​

The Settings tab is where you can customize the utility to your preferences, such as selecting your preferred unit of measurement, the real-time refresh rate, and it even allows you to set two temperature alarms, and tweak the taskbar temp widget.

Click on image to enlarge​

Last and certainly least is the About tab, which basically just tells you what version of the utility you are using, and provides a helpful link to check for any updates.

Overall though, I think this piece of software and the real-time monitoring feature in general is pretty cool. It is a natural progression since enthusiasts are information obsessed when it comes to monitoring the health and functioning of system components. Since DDR4 memory kits can cost about as much as a motherboard, graphics card or processor, it makes perfect sense to have the same monitoring capabilities as we have had on those parts for years. The one improvement that I would love to see is real-time voltage monitoring of memory modules, since there can be quite a discrepancy between each memory channel, especially on quad-channel memory platforms.

 

[SKYMTL]

Test Setups & Methodology

Test Setups & Methodology

For this review, we are going to be testing a variety of different frequency and timing configurations while also attempting to keep the CPU and Uncore clocks are identical as possible across all of those configurations, unless otherwise noted. Aside from manually selecting frequencies, timings, and voltages 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]

Overclocking Results

Overclocking Results

For our memory overclocking tests we are usually interested in two main elements: how well the memory scales with additional voltage and how versatile it is at overclocking with different timings. During our early testing, we absolutely did notice that this memory kit kept scaling when we increased vDimm all the way up to 1.50V, but we don't really have any interest in stress testing with that much voltage, both for the sake of the memory ICs and our processor's memory controller. As a result, we capped memory voltage at 1.40V during in our DDR4 overclocking endeavours, which is still almost 20% over the stock 1.20V, and put all of our efforts on testing timings scalability instead.

In order to make sure that there weren't any possible CPU-related bottlenecks, the CPU Cache Voltage was set to 1.35V and the CPU System Agent Voltage set to 1.25V. We focused on four basic timing configurations (12-12-12 / 13-13-13 / 14-14-14 / 15-15-15) all with a 1T command rate for optimal performance.

With the pleasantries out of the way, let's get to the fun stuff.

Wow, this stuff is tight. We don't mean that in the youthful slang way either. The Micron memory chips that Crucial are using on this Ballistix Elite model are capable of tight timings at much higher frequencies than we have seen from any kit equipped with Hynix MFR ICs. The Hynix is still unrivalled when it comes absolute maximum frequency - DDR4-3100 15-15-15-1T is trivially easy with them - but with this Crucial memory kit what you lose at the top-end you more than make up at the low-end. DDR4-2750 12-12-12-26-1T? Yes please! For comparison, click here for a pop-up showing what our previously untouchable Corsair Vengeance LPX DDR4-2666 kit managed at those same timings. It's not even close until timings start loosening up at 15-15-15. Our G.Skill DDR4-3000 put up a stronger showing at 14-14-14, but it's still blown out of the water by this incredibly versatile Ballistix Elite model.

At 15-15-15-1T, we either hit a hard or artificial limit. Even when we increased the voltage to 1.50V and loosened the timings to 16-17-17-2T we didn't gain a single additional Mhz. As a result, either this where the frequency ends for these ICs or there is some secondary or tertiary timings, or even something more complex holding us back. We have heard of certain kits reaching up to DDR4-3200 16-17-17, and have no reason to be skeptical of that, but it's going to take some additional BIOS work on our part (and maybe the motherboard manufacturers too) to potentially make that a reality.

By the way, it is hard to explain in layman's terms, but this memory kit's Micron ICs placed more of a strain on our processor's Uncore than all three of our previously reviewed Hynix MFR-based kits did, even at similar clocks and timings. As you will see below, we combined the memory overclocks with very high 4250Mhz Uncore speeds, and usually we only need a CPU Cache Voltage of 1.35V to achieve that with full stability. However, with this kit we did actually end up having to boost that up to 1.40V. Just keep that in mind during your overclocking efforts.

Click on image to enlarge​

These screenshots are just to prove that we did indeed achieve the overclocks listed, and that they were stable enough to pass a series of mainstream benchmarking and stress testing applications. If you are doing super critical work, then maybe a little Prime95 stress testing should be done as well, but for gaming and day-to-day tasks our testing is more than sufficient.

Usually, we would now include a 15-15-15-2T overclocking result to show off the maximum stable frequency that we achieved while overclocking, but as mentioned above, we did an artificial hard limit at DDR4-3000 15-15-15-1T.

 

[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 G.Skill memory kit's XMP profile applied, and using own our manual overclocks. This will illustrate how much performance can be achieved at various DDR4 memory speeds and timings. 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.

AIDA64 Extreme Edition

AIDA64 uses a suite of benchmarks to determine general performance and has quickly become one of the de facto standards among end users for component comparisons. While it may include a great many tests, we used the AIDA64 Cache & Memory Benchmark in order to tests the 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.

 

[SKYMTL]

Gaming Benchmarks

Gaming Benchmarks

Futuremark 3DMark (2013)

3DMark v1.1.0
Graphic Settings: Fire Strike Preset
Rendered Resolution: 1920x1680
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]

Conclusion

Conclusion

As we stated in the introduction, Crucial is extremely conservative when it comes to specifications since they are basically more concerned with maintaining their reputation as a well established manufacturer of highly compatible and reliable memory rather than engaging in a specs war with the smaller manufacturers. Despite this, thanks to their symbiotic relationship with Micron, they still do end up with some top-notch memory ICs and they bin the very best of them for the Ballistix Elite series.

The end result of that chip binning was revealed in our overclocking results. From DDR4-2750 12-12-12-1T, to DDR4-2838 13-13-13-1T, to DDR4-2929 14-14-14-1T this memory kit surprised us at every level. At these timings not one of the Hynix MFR-based memory kits that we've reviewed comes close. However, we did hit a wall at DDR4-3000 15-15-15-1T. Loosening the timings and increasing the voltage did nothing to surmount it, which tells us that it's probably a hard limit or maybe an artificial one caused my some secrets timings we haven't yet figured out. Therefore, in the future, we have reason to be optimistic that there's some untapped potential for those who place more value on high clock frequency above looser timings.

If you look at the benchmarks results, all the configurations achieved very comparable and impressive results. 76GB/s of read bandwidth anyone? Even though we go to significant lengths to keep the processor and Uncore clocks as identical as possible across the configurations - while at the same time not holding back the memory overclock - the 14-14-14-1T configuration had a 40Mhz CPU and 30Mhz Uncore core advantage which clearly skewed a few results in its favour.

Having said all of that, if all things were perfectly equal, with this memory kit any loosening of the timings doesn't really result in a loss of performance because it is matched by a compensating increase in memory frequency. Basically, the timing/frequency scaling is pretty much perfect, so you really don't have to make any difficult choices. This is a level of versatility that we have not seen with any other memory kit. By the way, if you're willing to pay a premium for a quality DDR4 memory kit and you don't also overclock the Uncore, you are basically throwing your money away. The performance gains simply aren't going to materialize after a certain point, usually at about DDR4-2750 with a stock Uncore.

Now this $335 USD Ballistix Elite memory kit doesn't exist in a vaccuum, so it is going head-to-head against several competitive products, many of which are less expensive. You have to consider that you can now find the previously flagship G.Skill Ripjaws4 DDR4-3000 16GB kit that we reviewed a few months ago for $299 USD, and the sleeper Corsair Vengeance LPX DDR4-2666 16GB kit can be found for as low as $235 USD. Another kit worth discussing is Crucial's own Ballistix Sport BLS4K4G4D240FSA, which runs at middling DDR4-2400 16-16-16 but retails for only $200 USD. That certainly looks promising, but are giving up the high grade binned ICs, awesome murdered-out heat spreaders, and real-time temperature monitoring. Not an easy choice for some.

When we started this review, we said that given its price tag, this memory kit better pretty special. And as it turns out, it is! Not only does it have awesome looking heat spreaders and real-time temperature monitoring capabilities, but it has a ton of untapped frequency and timings headroom. It might not be able to achieve stratospheric memory frequencies as easily as Hynix MFR-based kits, but if you are more interested in day-to-day performance than just showing off, you are going to be hard pressed to find anything better and versatile than this Crucial Ballistix Elite DDR4-2666 16GB memory kit.