HyperX Predator Quad Channel DDR4 RAM 16Gig Kit Review

Every single PC has one thing in common (well, they have a few, but this is one of them), Different capacities and speeds, RAM is something they all need, usually the more RAM you have or the faster the better your PC runs and the more work you can get done.  There are many other things as well that affect the speed of your computer, but we are focusing on RAM here, come on people FOCUS.

Today we will be taking a look at the HyperX Predator Quad Channel DDR4 16Gig Kit, Model number HX430C15PB2K4/16.  Before we get into the review, let’s check out the specifications of this memory.

• Memory Type: DDR4 SDRAM 288-Pin
• Capacity: 16Gig (4 x 4GB)
• Speed: 3000Mhz (PC4 2400)
• XMP Speed: XMP1 2400MHz | XMP2 2666MHz Optimized for Intel X99 series motherboards
• CAS Latency: 15
• Timings: 15(tCL), 16(tRCD), 16(tRP)
• Voltage: 1.35
• ECC: No
• Multi-channel Kit: Quad Channel Kit
• Buffered/Registered: Unbuffered
• ECC: No
• Predator Signature tall heatspreader
• Lifetime Warranty

Now that we have those formalities out of the way, let’s check out the unboxing, so you know what is in the box before you buy it.

Now that we have unboxed them, let’s take a closer look at what’s inside.

The memory comes packaged in a simple brown box encased in a simple plastic casing.  I like when products come packaged simple, it means you save a few bucks, but we can get to that a little later in the review.

Getting a closer we can see the memory itself.  It comes in a rather tall black aluminum heat spreader for greater heat dissipation to optimize the reliability of your overall system and also potentially improve performance.

This memory with the spreaders stands a little over 2 inches, so make sure you have the clearance for it.  Cower amongst this obelisk of grandeur.

And even closer here we can see the description of the RAM.  In case you didn’t know, let me decode the model number of the memory, it’s not just a random set of numbers and letters to impress you.

• HX = Product Line (HyperX)
• 4 = Technology, (DDR4)
• 30 = Frequency (3000Mhz)
• C = DIMM Type (UDIMM = Unregistered DIMM)
• 15 = CAS Latency (Delay between activation of row and reading a row)
• P = Series (Predator)
• B = Heatspreader Color (Black)
• 2 = Revision (2^nd Revision)
• K4 = Kit (Kit of 4 Modules)
• 16 = Total Capacity (16Gigs)

This kit comes in the 3000Mhz flavor at a CAS Latency of 15 but stick around and we will see what this can handle, at least a little of it.

The 3000MHz speed for many boards and memory combinations can be a bit tricky, so in accepting this RAM I was a bit nervous.  They look nice and all, but do they work and are they fast?  Before we can see how fast they are, we need to install them.

[nextpage title=”Installation”]

Installing RAM is one of the easiest processes when building a computer, but take that with a grain of salt.  As easy as it is to install, I personally know people that have installed RAM the wrong way, reversed and made it fit.  This means he had to force it down with all his might totally ignoring the fact that there is a piece of plastic right in the center of the DIMM slot, and a matching notch on the memory, so of course he broke the piece of plastic.  Oddly enough the PC would not boot, I wonder why?

All joking aside, don’t feel bad if you don’t know how to install RAM, we all had no clue what we were doing till we learned.  Hopefully I can show you with this video.

So in this video, you saw how you installed these sticks of DDR4 RAM.  I showed you how the RAM could have been installed incorrectly, how it seesaw’s once you slide it in the slot.  If this happens, be sure to flip the memory around (gold pins go into the slot) and try again.

On this particular board the EVGA X99 Classified, they installed in slot 1, 3, 5 and 7.  Not all boards have 8 DIMM slots, some only have 4 and some have 16 or more.  To be 100% sure on what slots you need to install your RAM to either keep it in Quad Channel or Dual channel, please be sure to check in your motherboard manual or on the motherboard manufactures website.

If you are using Quad Channel memory like the HyperX Predator HX430C15PB2K4/16, you want to make sure you get it running in Quad Channel, though it will run in Dual Channel with slight performance degradation.

Quad Channel memory allows the communication between the memory controllers on the CPU, in this case the Intel Core i7 5930K processor and the RAM.  4 Lanes or communication, allows more information to pass faster than 2 lanes, or Dual Channel.  Not all programs/drivers will take advantage of Quad Channel versus Dual Channel so you may not see a performance improvement but as more programs are written to utilize this advancement the performance improvement will become more evident.

So how can we see how to configure it and benchmark it, that’s going to my next page.

[nextpage title=”Configuring”]

So now that the memory is installed, we need to configure it and then benchmark it

To configure it, when we turn on the computer and enter the BIOS.  To enter the BIOS, you may need to find the key needed to enter the BIOS in your motherboard manual, though most board need for the delete key to be press, F2 or Escape when you see the BIOS Splash screen.

If you have the splash screen disabled, the message may appear something like this

With the splash screen enabled, you will typically have your motherboard manufactures logo flash quickly, mine looks like this.

Very classy looking if I do say so myself, and I do.

One thing to remember when setting up this memory is that it does have Intel’s XMP 2.0 Profile; in this case it has 2 Profiles.

In this picture, I have XMP Profile 1 set.  This sets my memory frequency to 2400,… but wait its 3000Mhz memory.  You will notice right below it though it reads “Target Memory Frequency” 3000Mhz, but how do you figure?  Well before we find that, notice below that that under “Basic Timing Configuration” the tCL (CAS Latency) to 15, tRCD (Row Address to column Address Delay) to 16, tRP (Row Precharge Time) to 16 and tRAS (Row Active Time) to 39, this is all saved in the XMP Profile, along with the memory voltages.

Some of you may ask, what is an XMP Profile and if I get it, can I cure it?   Well, it’s not something you need to cure, but it is something you want.

XMP stands for Extreme Memory Profile and it was developed by Intel and it was developed to enable tested predefined performance tuning of RAM beyond JEDEC SPD standard specifications.  These profiles are stored in the SPD of the XMP DIMM and are extracted by the BIOS to tune the memory controller for optimal memory performance.  This was originally designed for DDR3, now in DDR4 it is XMP 2.0.

I know, I through a few more acronyms at you, but let me explain.

• tCL (CAS Latency): The top and most important of memory timings. CAS stands for Column Address Strobe and the amount of cycles, or time in cycles between sending and receives commands.  The lower the tCL or CAS the better performance.
• tRCD (Row Address to Column Address Delay): The amount of cycles or time in cycles the RAM is issuing active command and read and write commands.
• tRP (Row Precharge Time): Minimum time between active commands and reads and writes of the next bank  in the memory module.
• tRAS (Row Active Time): Time between a row being activated by the precharge and then deactived.  Once the tRAS has completed, the row can be deactivated.  The lower the tRAS the better the performance.
• JEDEC (Joint Electron Device Engineering Council): JEDEC standards are designed to keep manufacturers and specs in sync.  Finding RAM for your system is a little hard today, without the JEDEC standards, it would be next to impossible and would lead to memory being more proprietary thus making prices skyrocket.
• SPD (Serial Presence Detect): Information stored in an electrically programmable and erasable memory that is read only.
• DDR (Double Data Rate): Double data rate synchronous random access memory.
• DIMM (Dual in line memory module): A module containing one or several random access memory chips on a circuit board with pins to connect it to the motherboard.
• BCLK (Base Clock): The base frequency of the CPU is running. It is derived by the multiplier of a CPU, in this case 35 times the base clock of 100Mhz for 3500Mhz.

Ok, back to the memory and its configuration.  So I left you with the memory frequency, and why would the almighty XMP profile set the frequency so low?  That my friends has to done with the BCLK, more commonly referred to as the B Clock.  On this Haswell series of processors as well as some previous processors, the base clocks 100Mhz frequency is tied to memory, CPU Core and PCI-E slots, so when you overclock the b clock, you overclock these as well affecting your video card, RAM, CPU, Sound Card and more, be careful with this setting.

In the picture below at the bottom, you can see the BCLK is set to 125.  If you start overclocking the BCLK too much you will start having to adjust vcore voltages, cpu voltage, memory voltages and more and passed 125 can be a bit dangerous.  Just above that you can see that the CPU cores have been downclocked to 30, so that would be 3000Ghz but that’s the CPU, we don’t want to underclock that.  The trick here is that by raising the BCLK frequency to 125Mhz, we have to multiply 30 by 125 which equals 3750Mhz, so actually the CPU has been overclocked just by setting the XMP Profile to profile 1.  This also brings the memory frequency to 3000Mhz.

On XMP Profile 2 we can see this approaches things a little differently.  This profile bumps up the target frequency to 2667Mhz from Profile 1 but then also tightens the timings a bit, so it’s not at 3000Mhz anymore, maybe the tighter timings help compensate for the lack of frequency.

Then we can see here that the multiplier is still at 38 and the base clock has been lowered to 100Mhz.  Hmmm, didn’t I get 3000Mhz RAM here, did I get jipped?  Well, there is only 1 way to tell, it’s benchmarking time.

Then also, something very important is that you can verify if your memory is running in Quad Channel using a few different pieces of software, but I will show you inside of CPU-Z.

[nextpage title=”Benchmarking”]

OK, now I will walk you through some benchmarks and first discuss my suite of tools.  Here is what I am using.

• 3DMark
• 7Zip
• AIDA64
• CPU-Z
• Intel Extreme Tuning Utility
• PCMark
• Sisoft Sandra

Let me explain why I use these pieces of software, please chime in if you would recommend anything different.

3DMark: While not a benchmark for memory, it does provide overall system performance.  It does of course steer more in the direction for video but memory speeds do play a role in its performance.

7Zip: While 7Zip is a tool for file compression, it also provides a tool to benchmark the speed of your processor and RAM.  The Compression speed test relies heavily on your RAM latency, Data cache size and speed and TLP or translation lookaside buffer.  The TLB is a memories cache that stores recent translations of virtual memory into physical addresses for faster retrieval of that cache.

Decompression relies more on the CPU’s integer operations and very little on memory and its own cache, but it is wrapped in the test.  The performance is based off of MIPS (Million Instructions per second.  The overall score is calculated from the measured speed of its calculations.  I run this test for 100 passes, not only test get a better tested score but also to test system stability under each configuration.

AIDA64: Is a benchmarking suite that can benchmark the entire system, but I will focus this test only on Memory, since we have a few other suites benchmarking everything.  The memory test here benchmarks reading, writing with data transfer bandwidth and latency.

CPU-Z:  Well this is not really a test, but a utility to tell you what speed your CPU, Memory and the likes are running at.  This piece of software provides information to confirm what you have done in the BIOS making sure all is well, confirming Bus Speeds, Multipliers, Frequencies, Channel’s and more.  Yes, many pieces of software do this as well, but this is 1 of them.

Intel Extreme Tuning Utility: honestly this is one that I completely forgot about till I was done with all the testing.  Yet another piece of software to double check on settings, but this also provides a Stress Test utility and a Benchmarking Utility.  The benchmark of course is CPU related, but the CPU is where the memory controller is but the Memory Stress Test will stress the CPU and/or Memory as well.  While I don’t mention this during the review really, I test all settings in this as well to make sure the system is stable aside from the other suites as well.

PCMark:  Does not focus on memory specifically, memory and memory speed will affect its end result.  PCMark has various testing packages, Home Test, Creative test, Work Test, Storage Test, Application Test and Battery Life testing.  Some packages are self-explanatory; I will focus on the Creative Test.

The Creative Test includes works loads typically found on enthusiasts and professionals that work heavily with media and entertainment.  The suite performs various tests in the lines of Web Browsing, Photo Editing, Video Editing, Group video chat, Media transcoding and gaming.  These are all different types of workloads, but all are very memory centric, including overall system performance.

Sisoft Sandra:  Sandra is also a suite that can potentially benchmark the entire system and can also be used for diagnostic purposes, like most if not all of the previous software titles mention.  For memory, I will be running Sandra’s Memory Bandwidth and Latency tests.

For testing purposes I will be testing XMP 1 versus XMP 2 to see exactly what the 2 offer in performance and maybe help you select what profile you would be using.  This review also will serve as a base for future DDR4 reviews as well.

I will also list my build as a point of reference.

• Thermaltake Core X31 Case:https://geni.us/6NAIJBN?zWTvJH
• Intel Core i7 5930K:https://geni.us/6NAIJBN?4C8Itd
• EVGA X99 Classified:https://geni.us/6NAIJBN?E9eamo
• Arctic Liquid Freezer 240MM Liquid Cooling:https://geni.us/6NAIJBN?vEaJAf
• Kingston HyperX Predator 3000Mhz 16Gig:https://geni.us/6NAIJBN?w9kPe5
• Sapphire Nitro R9 390 Video card:https://geni.us/6NAIJBN?klUObl
• Samsung 850 EVO 500GB SSD:https://geni.us/6NAIJBN?1gf0fs
• Hitachi 1TB SATA 3G HD:https://geni.us/6NAIJBN?pU2QOo
• Patriot Ignite 480GB SSD:https://geni.us/6NAIJBN?eoPVsG
• Kingston HyperX 240GB SSD:https://geni.us/6NAIJBN?8leEDW
• Plextor 256GB PCIE SSD:https://geni.us/6NAIJBN?gVBR
• Cooler Master silent Pro Gold 1200W Gold Power Supply:https://geni.us/6NAIJBN?Umwm
• LG WH16NS40 BluRay Rewriter:https://geni.us/6NAIJBN?60fP
• MicrosoftWindows 10 Professional:https://geni.us/6NAIJBN?GYbBRY

Again for reference, here are the XMP Profiles:

XMP Profile 1.0

XMP Profile 2.0

Now back to your regularly scheduled benchmark.

3DMark

The memory frequency differences were seemingly significant from the overclocked 3000Mhz in XMP 1.0 and stock 2666 in XMP 2.0.  XMP 2.0 has a lower memory frequency, but tighter timings and a 50Mhz increase in CPU performance, while XMP 1.0 has a BCLK Frequency of 125 increasing the memory frequency of 2666 to 3000, slightly looser timings and 50Mhz

XMP 2.0 had a greater improvement it seems due to the tighter timings, I would not have though it would have made such a huge difference.   Being that 3DMark’s focus is more heavily on the GPU I would have thought it would have benefited from the overclock base clock, increasing the bus on the PCI-E slot, but that seems not to have been the case.

PCMark

In PCMark XMP 2.0 performs yet again better than XMP 1.0.  This is more over all overall system result and did not focus on graphics; we can see that the raised Memory frequency did not help what so ever or even the increase bus frequency yet again.

7Zip

On 7-Zip we can see the combination of raised memory and bus frequency have helped XMP 1.0 hold its own.  XMP 1.0 was 304MIPS higher than XMP 2.0, a .897% improvement.  Putting into a percentage makes the improvement seem so much more minuscule doesn’t it?  I was to be expected though since we already know that this test focus’ almost entirely on the memory performance.

AIDA64

XMP 1.0 this time around takes the lead completely as you can see above coming above all other XMP 2.0 results, and yes I mean all of them.  While latency is lower, you want latency to be lower as it is the delay time between the memory controller which is on the CPU and the memory itself, the higher the delay the slower the process, the lower the delay the quicker the process.

This test is a memory centric test much more than the other tests, even though it does involve the CPU, but it is inevitable since the memory controller is on the CPU.

Sisoft Sandra

Sandra is bringing something to light, for true memory performance it looks like XMP Profile 1 takes the lead yet again.  Both Sandra and AIDA64 show that memory frequency allows memory to perform much better, but in this we can see that overall system performance is improve if the CPU is clocked higher as well as the memory timings tightened up some.  This means that each memory profile is tailored towards a specific type of user.

XMP 1.0 seems to be tailored more for professionals where memory is more of a requirement that processing power.  People that use their systems for encoding and decoding as well as photo editing, CAD and the likes would benefit more with XMP 1.0.

XMP 2.0 seems to be tailored more for overall system performance.  Nothing wrong with a well-rounded system and in XMP 2.0 this will take care of just that, making everything perform very well and seems to be tailored more for gaming and basic functions.

No matter what a good CPU will improve performance, but you want to pair that CPU with some nice memory like the HyperX Predator Quad Channel DDR4 16Gig Kit HX430C15PB2K4/16.  Any one of these 2 profiles will help you in every task, though specifics are always key to get that extra umph but is there more?

[nextpage title=”Overclocking and comparative performance”]

There is no good review if there isn’t a bit of overclocking to show you what more the memory can do above standard clocks, and we got that here is well.  This is to help you decide which should be higher overclocked to get better performance, Memory or CPU?

Please take note before you get into the overall results, there is much more that can be done, but overclocking, tuning and tweaking does not take 5 minutes, it can take some time and I want to make sure I can bring you this review in a timely manner.

One thing to mention before we get into the overclocks, what works for me may not work for you.  Even though you may have the exact same RAM, the slightest differences in the RAM, different IC’s, PCB’s, Date and Lot codes, anything can improve or reduce performance.  This has nothing to do with Kingston or HyperX, this is with any piece of memory, CPU, motherboard, video card,… anything.

Before I get into the testing, I will provide the settings I used as I did previously.

CPU running at 4.275Ghz, Base Clock at 112.50, XMP Profile 1, Memory Frequency at 2667, overclocked to 3000Mhz.

XMP1.0 OC

CPU running at 3.80Ghz, Base Clock at 100.00, XMP Profile 2, Memory Frequency at 3200Mh.

XMP2.0 OC

This will take memory timings out of the equation completely, focusing only memory Frequency versus CPU frequency with a little overclocking of the bus, no way to avoid that.

3DMark

In 3DMark, it seems that it prefers the XMP 2.0 OC profile; this profile focuses more on the Memory Frequency, being that it now 3200Mhz.  Mind you, the difference in this score is only .15%, but it’s still higher.  Even though XMP 1.0 OC employs a 4.275Ghz overclock on the CPU resulting in the 112.50 bus and a decent 3000Mhz frequency on the RAM, it can’t stand to the memories 3200’s frequency on XMP 2.0 OC.

PCMark

XMP 2.0 OC overtakes XMP 1.0 OC yet again, this time by slightly more.  These results have a .59% difference in favor again of XMP 2.0 OC.  Be interesting to see what the rest get compared to what they go before.

7Zip

XMP 2.0 OC for the win, showing again that it relies much more on Memory frequency over CPU frequency.  It proves that even though XMP Profile 1.0 OC has a significantly higher CPU Frequency the Memory Frequency determines its fate.

AIDA64

XMP 2.0 Oc leads yet again completely as you can see above.  Since this is a memory benchmark only, at least this portion the memory frequency being higher (3200Mhz over 3000Mhz) takes precedence over CPU or bus overclocking.

Sisoft Sandra

Sandra again shows memory performance improved with the frequency set to 3200 on XMP 2.0 OC, it looks like XMP Profile 2 takes the lead.  It seems that throughout this test, comparing CPU overclock versus Memory Frequency overclock that actually the Memory Frequency overclock improves all aspects of the PC.

[nextpage title=”Final thoughts”]

Memory as we can see, and I am sure as many of you have already know can really change the performance of a PC for all users, Apple Mac’s too.  The tiniest of settings off can make it perform not so well; something that has plaque computer users for a long time until the XMP Profile was introduced, making life just a little easier.  In this case, HyperX has decided to employ 2 x XMP Profiles to make things that much more simple, to give you the option of how you want to use the memory, a nice feature.

You are giving the opportunity even with those profiles to overclock the RAM even more, Kingston has ensured that your memory lasts by provided a taller heatspreader to dissipate all that unwanted heat.  More than that, they are so sure it will last that they give you a lifetime warranty on the memory, that means a lot.

Pros

• Tall Aluminum black heat spreaders
• 2 x XMP Profiles
• Low voltage
• High Memory Frequency
• Solid and Stable
• Lifetime Warranty

Cons

• Priced high compared to similar 16Gig 3000Mhz CL15 kits.
• While it is a Pro and can also be a Con, the heat spreaders might be too high for some configs, be careful and measure.
• Only comes in black, no color options

The memory is solid, overclocks like a champ and has headroom for more, even tighter timings as well as high voltages if need.  The reason I cannot award this kit 5 stars is that the price is too high compared to other 16GB 3000Mhz CL15 kits,  then sometimes you have to factor in shipping, so it comes out a little higher. The heat spreaders being a little tall will knock it down a little more, but not by a full star, I can’t say even the price knocks it down a full star since it is not EXTREMELY more expensive, just slightly.  The color is a preference even though I do like it, others may not but it will not affect the rating.

I really want to award this a 5 star editor’s choice, but alas I cannot.  With that said though, I will give this 4 stars and it is very highly recommended.  Verify that you have enough clearance in your case and who knows, the price may be cheaper when you next see it, if it is though, I might knock it back up another .5 stars.

Thanks guys, please let me know what you think.