Corsair Vengeance LPX DDR4-2666 32GB Review

[SKYMTL]

In on of our last memory reviews, we took a look at the Crucial CT4K8G4DFD8213, which features a low DDR4-2133 clock speed but is the most affordable 32GB DDR4 kit on the market. As a result, for this review we decided to take a completely different approach, basically a 180° turn. As a result, today we are reviewing a Corsair Vengeance LPX 32GB DDR4-2666 CL15 memory kit, which is one of the fastest 32GB DDR4 memory kits available, and consequently also one of the more expensive offerings out there. It is substantially more than the 16GB alternative we also reviewed.

The Corsair model that is the focus of this review is the $415USD/$600CAD CMK32GX4M4A2666C15, which is part of their Vengeance LPX series and runs at DDR4-2666, which is pretty high for a 32GB kit. Although there are now some higher clocked kits that have just hit the market, what makes this model fairly unique is the fact that it features reasonable 15-17-17-35 timings - which is relatively tight compared to some other highly clocked 32GB kits - and a low 1.20V stock voltage, whereas most other in this category start off at 1.35V. Another thing that makes this model special is that it features the exact same specifications - minus the fancier heatspreaders - as Corsair's flagship Dominator Platinum, which retails for an eye-watering $890USD/$1000CAD. So you are getting the same specs, same quality, same warranty, for a good bit less money. Also, just like with the 16GB Vengeance LPX kit that we reviewed earlier, this model has a secret but awesome party trick that makes it even more attractive to well-heeled enthusiasts.

Click on image to enlarge​

Foregoing the more common plastic clamshell with cardboard insert approach, Corsair packages this model in a little branded box. It is a pretty compact affair since the modules feature fairly low profile heatspreaders, hence the LP in the LPX model name. Inside there are two separate plastic trays that each hold two modules. They can be opened by hand - no knife required unlike with blister packs - which means that they are also resealable. There is also a little pamphlet on the inside listing warranty details.

Click on image to enlarge​

Although these memory modules feature an updated look, they definitely share some design DNA with those found in the previous Vengeance LP DDR3 memory kits. Made of anodized aluminium, these are fairly low profile heatspreaders with a height of about 34mm/1.34". We could not remove the heatspreaders without risking damaging this RAM, but we do know with certainty that our double-sided modules were manufactured with Hynix MFR ICs, which are the best overclocking DDR4 memory chips on the market right now. That may change in the future, but if you can find a kit with the same version number (5.29), then you have a great chance of also having these excellent memory chips.

Click on image to enlarge​

In the introduction we mentioned that this memory kit had a secret party trick, and well here it is. Corsair have included an additional XMP 2.0 profile that pushes the modules up to DDR4-2800 while keeping the same 15-17-17 timings as the stock DDR4-2666 setting. There is voltage bump from 1.20V to 1.35V, but that's well within reasonable limits. This is pretty awesome addition since you are basically getting a DDR4-2800 RAM for less than you would ordinarily pay for such a fast high capacity memory kit.

 

[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. However, we quickly found out that this memory kit doesn't really benefit from going above 1.40V and in fact sometimes worsened overclockability depending on timings. So we decided to simply increase the voltage from the manufacturer specified 1.35V to 1.40V, and put all our efforts on timings scalability instead.

In order to make sure that there weren't any possible CPU-related bottlenecks, the CPU Cache Voltage was set to 1.30V 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.

Large capacity memory kits have always been more hit or miss when it comes to overclocking since there are more ICs and thus more chances to be held back by just one poor clocking chip. However, since the 16GB version of this memory kit proved to be an exceptionally good overclocker, we obviously had high hopes for this model especially since it features the same Hynix MFR ICs. We had no doubt that it would be able to upstage our last 32GB kit - a low-clocked Crucial part that nevertheless proved to have a lot of headroom - but by how much was a mystery. Thankfully, as you can see above, this Vengeance LPX model turned in some very solid numbers.

Obviously, we would have loved to hit the Holy Grail of DDR4-3000 15-15-15-1T, but it wasn't quite meant to be. Having said that, 36Mhz off isn't something we are going to gripe about. As a consolation prize, this kit was particularly impressive at 14-14-14 and it achieved the highest frequency that we seen so far at DDR4-2854. Further tightening the timings to 13-13-13 caused a serious decline in headroom, but we were still above this kit's stock DDR4-2666 clock speed, so it's still pretty remarkable. Impressively, when we set the timings to 12-12-12 this memory kit's overclock was almost 70Mhz higher than what we managed on the 16GB DDR4-3000 G.Skill Ripjaws4. Clearly, this Vengeance LPX 32GB model is a very versatile solution for those who want to be able to play around with frequency and timings to find their system's particular performance sweet spot.

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 may a little Prime95 stress testing should be done as well, but for gaming and day-to-day tasks our testing is more than sufficient.

Since all of these overclocks were done with the 1T command rate, we decided to see what this Corsair kit was capable of at 2T, which is what this particular model is rated for anyways. As you will see below, it is very impressive.

​

As you can see, by loosening the command rate we were able to push this kit just past the DDR4-3000 mark while still keeping the tighter than stock timings. To put this into perspective, there is only one 32GB DDR4-3000 15-15-15 memory kit on the market right now, and it retails for over one thousand dollars. If your results end up matching ours, we think you're getting a relative bargain here. So does any of this actually translate to worthwhile day-to-day performance improvement? Well check out the following two pages to see our benchmarking results with eight different timing configurations, including the two XMP 2.0 profiles.

 

[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

For a long while, this Corsair CMK32GX4M4A2666C15 model was the undisputed performance champion among 32GB DDR4 memory kits. Today, there are higher clocked products available, some even reaching the vaunted DDR4-3000 mark, but as you might expect they are exceedingly price prohibitive. You can now buy DDR4-2800 kits for about the same price as this one, but there's not much logic in doing so since they have equal voltage requirements but looser timings than this kit does via its "secret" DDR4-2800 XMP profile.

Furthermore, if our overclocking results are in any way typical, thanks to the use of excellent Hynix MFR ICs you should have no problems matching even the fastest kits on the market with a little voltage bump, and at two-thirds the price. Indeed, we had high hopes that this Vengeance LPX would have some decent overclocking headroom, and we were not disappointed with DDR4-3004 15-15-15-2T. DDR4-2854 14-14-14-1T, DDR4-2672 13-13-13-1T, and DDR4-2470 12-12-12-1T were also excellent results, once again demonstrating the solid timings versatility of the Hynix chips.

We have also grown to be fans of the Vengeance LPX heatspreaders. With their low profile design - both in terms of height and aesthetics - these modules fit and look good on every X99 motherboard that's come across our test bench thus far. And if you do encounter any issues, Corsair has a terrific online user forum, great customer service, and of course the lifetime warranty.

Overall, this Corsair Vengeance LPX model hits all the right marks for someone looking for an enthusiast-oriented 32GB memory kit. It's not in the same price category as the similarly sized but run-of-the-mill Crucial DDR4-2133 that we reviewed a while back, but it is a hell of a lot better looking and more capable across the board. What you are getting here is a large DDR4-2666 kit that's actually a DDR4-2800 kit, and one that uses the best memory chips on the market to great effect. A furry thumbs up.