Kingston’s FURY RAM brings auto overclocking to gamers
When Kingston released HyperX FURY memory last month at PAX East, they targeted it as a replacement for HyperX Blu, “for entry-level gamers and enthusiasts”. We really like Kingston memory and we have tested it many times at ABT. We received a 2x4GB evaluation kit from Kingston of their FURY 1866MHz DDR3 (HX318C10FWK2/8) that we will put through its paces for you. Although FURY comes in a choice of four colors, we picked white as they look great installed in any PC.
HyperX is the high-performance product division of Kingston Technology encompassing high-speed DDR3 memory, SSDs, USB Flash drives, and headsets. Kingston has 3 memory product lines: FURY, Genesis and Predator. They offer a complete series of dual-channel kits available in frequencies ranging from 1333MHz to 2800MHz, and in various combinations from 8GB to 32GB.
This evaluation will test 2x4GB HyperX FURY memory versus Kingston’s fastest 2x4GB HyperX Predator RAM (KHX28C12T2K2/8X) at 2800MHz which we already compared with running it at the stock 1600MHz specification. We will also compare with 2x8GB of Kingston’s 2133MHz Beast RAM . Previously, testing Ivy Bridge with 2x8GB Kingston’s Beast RAM at 2133MHz brought limited gains in gaming, and only to a few games. Since we have topped out with our RAM clocks at 2400MHz with Ivy Bridge, we now only use the Haswell platform since its RAM can reach 2800MHz and beyond. We are benching with the Core i7-4770K at 4.0GHz which is approximately equal in performance to i7-3770K at 4.5GHz.
We are using the GTX 780 Ti, which is able to differentiate between CPUs at 1920×1080 resolution and above, without resorting to using multiple graphics cards.
HyperX FURY replaces the HyperX Blu memory line with the most obvious visual difference being FURY’s asymmetric heatspreader design. This new high-performance memory offers automatic overclocking. Auto overclocking means that the default speed is 1866MHz instead of the motherboard’s usual assignment of 1300MHz or 1600MHz. FURY does not need the user to optimize its settings in the BIOS as is usually necessary.
HyperX FURY memory is fully Plug and Play so it automatically overclocks to 1866MHz without assigning a profile in the BIOS manually . FURY’s new heatspreader design is available in four colors ( red,blue, black, and white) which sit on black PCB. Gamers, modders, and system builders can now choose memory color-matched systems. HyperX FURY is available in 1333MHz, 1600MHz and 1866MHz frequencies, and we are testing the 1866MHz variety at its stock speed.
HyperX FURY memory is available in 4GB and 8GB modules as well as 8GB and 16GB kits of two. It is backed by a lifetime warranty, free live technical support, and Kingston reliability where every module is pretested before shipping.
Price to Performance
8GB of Kingston FURY 1866MHz RAM is currently priced at $79.99 at Newegg with free shipping. 8GB of HyperX Blu starts at about the same price at Newegg for 1300/1600MHz RAM. In contrast, 8GB of Kingston Predator 2800MHz RAM will cost you $219 at Newegg, while 8GB of Beast 2133MHz RAM costs $110.99 ($153 for 16GB, as we tested). As you can see, there are a wide range of prices to match the increasing RAM speeds, and we will compare the performance of each of our 3 speed grades.
HyperX FURY Memory Features and Specifications:
- PnP: Plug and Play automatically overclocks the memory up to the system maximum specs
- Capacities: 4GB singles, 8GB singles, 8GB kit, 16GB kit
- Frequency*: 1333Mhz, 1600Mhz, 1866Mhz
- CAS Latency: 9, 10
- Voltage: 1.5V
- Heatspreader: Stylish and aggressive in blue/black/red/white
- PCB: Black
- Reliable: 100% tested
- Guaranteed: Lifetime warranty
|4GB DIMM 1333||HX313C9FW/4||HX313C9FB/4||HX313C9FR/4||HX313C9F/4|
|8GB DIMM 1333||HX313C9FW/8||HX313C9FB/8||HX313C9FR/8||HX313C9F/8|
|8GB Kit of 2 1333||HX313C9FWK2/8||HX313C9FBK2/8||HX313C9FRK2/8||HX313C9FK2/8|
|16GB Kit of 2 1333||HX313C9FWK2/16||HX313C9FBK2/16||HX313C9FRK2/16||HX313C9FK2/16|
*HyperX FURY Part Number Decoder: HX3xx= HyperX DDR3 + Frequency; Cxx= CAS Latency; Fx= Color Code; (optional) x or K2/x= Kit of 2/capacity
The specifications also look great and we can’t wait to test it.
We will emphasize ten games, including our eight newest games that are most sensitive to RAM speed changes. And we also shall benchmark using synthetic and real world tests. We are benching using the fastest single-GPU videocard, the GTX 780 Ti, to give you a comparison of 8GB Kingston FURY RAM at 1866MHz versus 16GB Predator RAM at 2133MHz and versus Kingston’s fastest 8GB Beast RAM at 2800MHz in game benchmarks at 1920×1200 and 2560×1600 resolutions and at the maximum details that gamers play at. We have broken with our tradition of using a GTX 680 to test the Kingston RAM with the rest of our benching suite of real world and synthetic tests, and we now use the GTX 780 Ti for all of our benching.
Before we head over to our test configuration and benchmarking, let’s open the package.
Take a look at Kingston DDR3 – (left) 2x8GB HyperX “Beast” at 2133MHz and (right) HyperX “Predator” at 2800MHz, compared with FURY 1866MHz (far right) and 1800MHz 2GB HyperX Blu. The Predator RAM is much heavier and has a much higher profile.
In fact, we had to exchange our Noctua NH-DH14 for a Thermaltake Water2.0 CPU watercooler to accommodate the extremely high profile of Predator RAM.
Kingston offers a choice of four colors for their HyperX FURY RAM which allows the enthusiast to color-coordinate their equipment. We picked White.
Even though our PC’s interior is mostly black, silver and gold, the white FURY RAM fits right in.
It’s easy to open the package and we get a look at the Kingston insert which contains warranty information as well as the installation guide. A “Powered by Kingston HyperX’ sticker is included for your case.
FURY shows up properly configured in the ECS Golden Z87 MB BIOS although there is certainly room for performance tweaking. Here is a screen shot of the ECS BIOS with the stock “automatic” overclocked 1866MHz settings. There were no other profiles available to us.
Test Configuration – Hardware
- Intel Core i7-4770K (reference 3.5GHz, HyperThreading and Turbo boost is on to 3.7GHz; also overclocked to 4.0GHz; DX11 CPU graphics), supplied by Intel.
- ECS GANK Domination Z87H3-A2X motherboard (Intel Z87 chipset, latest BIOS, PCIe 3.0 specification, CrossFire/SLI 8x+8x) supplied by ECS
- Kingston 8GB HyperX FURY DDR3 PC1866 RAM – HX318C10FWK2/8 – 2×4 GB, dual-channel at 1866MHz, supplied by Kingston
- Kingston 16GB HyperX Beast DDR3 PC2133 RAM – KHX21C11T3K2/16X – 2×8 GB, dual-channel at 2133MHz, supplied by Kingston
- Kingston 8GB HyperX Predator DDR3 PC2800 RAM – KHX28C12T2K2/8X – 2×4 GB, dual-channel at 2800MHz, supplied by Kingston
- GeForce GTX 780 Ti, 3GB (reference clocks), supplied by Nvidia
- 2TB Toshiba 7200 rpm HDD
- Thermaltake ToughPower 775W power supply unit supplied by Thermaltake
- Thermaltake Water2.0 Pro watercooler, supplied by Thermaltake
- Onboard Realtek Audio
- Genius SP-D150 speakers, supplied by Genius
- Thermaltake Overseer RX-I full tower case, supplied by Thermaltake
- ASUS BW12B1ST 12x BluRay writer
- HP LP3065 2560×1600 thirty inch LCD
Test Configuration – Software
- GeForce beta 337.50 drivers; High Quality; Single-display Performance mode; Prefer Maximum Performance, Vsync off. No overclocking; power and temperature targets set to maximum with stock voltage and fan profiles using EVGA’s PrecisionX.
- Windows 7 64-bit; very latest updates
- Latest DirectX
- All games are patched to their latest versions.
- Vsync is forced off in the control panels.
- Varying AA enabled as noted in games; all in-game settings are specified with 16xAF always applied; 16xAF forced in control panel for Crysis.
- All game results show average frame rates
- Highest quality sound (stereo) used in all games.
- Windows 7 64, latest updates; DX10 titles were run under DX10 render paths and DX11 titles under DX11 render paths.
The Game benchmarks
- STALKER, Call of Pripyat
- Far Cry 3
- Tomb Raider: 2013
- Crysis 3
- Metro: Last Light
- GRID 2
- Total War: Rome II
- Batman: Arkham Origins
- 3DMark Vantage
- 3DMark 11
- Heaven 4.0
- Valley 1.0
- FireStrike/FireStrike Extreme
- Super Pi
- Fritz Chess Bench
- Custom PC Benchmark
- x264 HD
- Crystal Disk Mark
- HD Tach
- HD Tune
Synthetic Benching Suites
- SiSoft Sandra 2014
- PC Mark Vantage
- PC Mark 7
We have got an interesting project going, testing 3 categories of RAM speeds, while also looking at the effects of 16GB versus 8GB of system RAM. First, we’ll look at synthetic and at real world benches before we head for gaming results
Synthetic Benching Suites
For all of our testing, including gaming, we used our 2TB Toshiba 7200 rpm hard drive. All four of our premium Kingston 240GB HyperX SSDs are currently being used in building our FCAT (Frame Capture Analysis Tool). There are no average framerate differences running games on a SSD versus from a HDD. Although many synthetic benchmarks will score lower using a HDD, our results are completely consistent across all benching, and the performance differences will be similar for HDD- and for SSD-based PCs.
We have limited our game benchmark suite to ten games; our eight very latest games and two older games that we have benchmarked for about 5-6 years. Before we get to gaming, we want to see exactly where performance varies with RAM speeds grades, and there is no better tool for this than SiSoft’s Sandra 2014.
There are several versions of Sandra 2014, including a free version of Sandra Lite that anyone can download and use. It is highly recommended! SiSoft’s Sandra 2014 was released November 25, 2013. It has quite a few improvements over Sandra 2013 and there are regular service packs released to always keep it up to date.
First, we run the 2014 Sandra benchmark suite with the ECS Golden Z87 motherboard and Core i7-4770K as in all of our benches at 4.0GHz, first with the FURY RAM, then the Beast and finally, Predator – in increasing speed grades. Our charts comparing the individual benchmarks are here as well as in the performance summary for easy cross comparison.
Now it’s the 16GB of Kingston Beast RAM’s turn at 2133MHz with the same settings. We get a solidly higher overall score, 10.87 kPT, much of which is due to having twice the amount of RAM compared with both the faster and the slower 8GB kits.
Here are the Dhrystone and Whetstone benchmarks at 2800MHz:
Now it’s the 8GB Kingston Predator RAM’s turn at 2800MHz using the same settings, but this time, we only get a very slightly higher overall score than FURY’s 8.73 with 8.76 kPT.
Here is a summary comparison chart including some of the Sandra highlights comparing the FURY, the Beast and the Kingston Predator RAM at their stock speed grades using identical settings. The higher RAM speeds made only a slight difference in the synthetic benchmarks. And you pay nearly three times more for it over FURY.
PC Mark Vantage
PC Mark Vantage was developed primarily for Vista and it has been superseded by PC Mark 7.
We are using the Advanced Edition of PCMark Vantage. As with all of our tests, we use 2x4GB Kingston HyperX FURY, 2x8GB Predator, and 2x4GB Beast RAM. As settings and the rest of the hardware remains unchanged, the only performance differences should come from the DDR3 modules.
Final Scores: 12088 at Fury’s 1833MHz to 12654 at 2133MHz for 16GB Beast to 11555 for Predator‘s 8GB of superfast 2800MHz RAM. There can be some advantages from using faster RAM that may be overshadowed by just using more RAM.
We will keep our settings and hardware the same and now run the detailed tests of the basic version of PCMark 7.
We are testing at the same default settings that we tested with PC Mark Vantage. Even the free basic version has some very detailed test results
- 8GB of Kingston HyperX FURY at 1866MHz scores 4179
- 16GB of Kingston Beast at 2133MH scores 4472
- 8GB of Kingston Predator at 2800MHz scores 4299.
Will these increases hold true for gaming? Before we look at gaming differences, we need to look at more benchmarks including real world applications that many of us use daily, including calculation, encoding video, multi-tasking, image editing and gaming.
Super Pi is a single-threaded benchmark which can show the differences in IPC between CPU platforms. We choose to calculate 1 million digits.
- 9.457 seconds for 1866MHz FURY
- 9.067 seconds for 2133MHz Beast
- 9.100 seconds for 2800MHz Predator.
Predator is 0.357 seconds faster than FURY. Please continue on to Fritz Chess Bench.
Fritz Chess Bench
Fritz Chess Benchmark is found within the game’s program files and basically it crunches numbers to test your processor’s speed. Deep Fritz takes advantage of massive calculations and multi-threaded performance to work any CPU fully. It loads all threads 100% and will drive your CPU temperatures way up. We are using version 12 which is the same version we used for testing Core i7-920.
The results are given as, relative speed/Kilo nodes per second:
- 8GB FURY, 1866MHz – 30.03/14415
- 16GB Beast, 2133MHz – 32.44/15572
- 8GB Predator, 2800MHz – 32.61/15654
On to Custom PC benchmark.
CustomPC benchmark use widely available open-source applications to carry out the tasks that most of us perform on a regular basis. There are three tests, each of which measure different aspects of a PC’s performance. These tests themselves are not synthetic benchmarks but instead they use real world image, video and multi-tasking tasks to test the performance of your computer.
The tests are:
- GIMP Image Editing
- H.264 Video Encoding
- 8GB FURY, 1866MHz – 2340.
- 16GB Beast, 2133MHz – 2447
- 8GB Predator, 2800MHz – 2625
Using 2800MHz Predator RAM makes a difference over using slower-clocked RAM on the Socket 1150 platform in this real world benchmark. Beast’s 16GB of RAM takes a small lead over the 8GB kits in multi-tasking, and FURY falls a bit behind in video encoding.
CINEBENCH is based on MAXON’s professional 3D content creation suite, CINEMA 4D. This latest R15.0 version of CINEBENCH can test up to 64 processor threads accurately and automatically.
We are testing with the newest version. v11.5
Here are our results.
If you use Cinebench and video encoding a lot, or for your work, you want the fastest RAM. Otherwise, FURY at 1866MHz turns in respectable results.
Basically this test encodes a HD video clip into a x264 video file. The first pass is very quick and the second one is much slower and much more demanding of a task as it does the actual encoding.
This benchmark is heavily multi-threaded.
In all cases, both passes are quickest at the fastest RAM clocks. As with Cinebench, if you do a lot of video encoding, choose the fastest RAM you can afford. For the rest of us, FURY turns in a very respectable performance.
The next series of tests are going to focus on the hard drive and storage to see if there is a difference using faster premium RAM over using RAM at larger capacity or versus using slower RAM speeds.
- Measure sequential reads/writes speed
- Measure random 512KB, 4KB, 4KB (Queue Depth=32) reads/writes speed
We run the default tests by clicking “all”.
There is not really a lot of difference.
- Benchmark measures the performance of:
- Transfer Rate
- Access Time
- CPU Usage
- Burst Rate
- Random Access test
- Write benchmark
- Hard Disk information which includes partition information, supported features, firmware version, serial number, disk capacity, buffer size, transfer mode
- Hard Disk Health
- S.M.A.R.T. Information (Self-Monitoring Analysis and Reporting Technology)
- Power On Time
- Error scan
- Temperature display
The tests are inconclusive regarding RAM speeds; FURY is fine.
ATTO is a disk benchmark. The ATTO Disk Benchmark measures a storage system;s performance with various transfer sizes and test lengths for reads and writes. Options are available to customize the tests, including queue depth, overlapped I/O and even a comparison mode.
HD Tach is a low level hardware benchmark for random access read/write storage devices including SSDs and HDDs. HD Tach uses custom device drivers and low level Windows interfaces to determine the physical performance of the device. It is no longer supported and needs to be run in compatibility mode for Windows 7.
The results are
- 8GB FURY, 1866MHz – 432.3 MB/second
- 16GB Beast, 2133MHz – 434.1 MB/second
- 8GB Predator, 2800MHz –427.9 MB/second
Next, we move on to game-related benchmarks and games. Next up are the synthetics.
3DMark 11 and Firestrike are useful tests to track changes within the same system although they are often used to give a rough comparison between platforms.
First, we look at Futuremark’s 2011 DX11 graphics-heavy test.
The results are:
- 8GB FURY, 1866MHz – 11616
- 16GB Beast, 2133MHz – 12749
- 8GB Predator, 2800MHz –12710
Now we look at Futuremark’s 2013 DX11 graphics-heavy test.
The results for the Basic Test:
- 8GB FURY, 1866MHz – 9568
- 16GB Beast, 2133MHz – 9681
- 8GB Predator, 2800MHz –9707
The results for the Extreme Test:
- 8GB FURY, 1866MHz – 4828
- 16GB Beast, 2133MHz – 4881
- 8GB Predator, 2800MHz –4862
Practically, there is little difference. We can also see the scores summarized in the Game Benchmarks chart below.
The Game Benchmarks with the GTX 780 Ti
Here are the benches for the three Kingston HyperX RAM kits. All games have their setting completely maxed out and levels of AA are identified on the chart. The GTX 780 Ti is run at reference clock speeds, while our i7-4770K is clocked at 4.0GHz.
If you are a gamer, you can definitely “get by” with using RAM at stock speeds although 4GB of system RAM is no longer sufficient to play some games (Call of Duty: Ghosts, for example), and 8GB appears to be the new minimum for a fast gaming platform. Using FURY compared to the fastest Predator RAM with nearly 1000MHz difference, only Crysis and Far Cry 3 showed a small deficit of 1-2 fps at 1920×1080, hardly noticeable.
Let’s head for our conclusion.
There is absolutely no doubt that running RAM at a faster clock speed produces performance gains and saves time in several common tasks, especially with multi-tasking, but also including video encoding and decoding. However, for the gamer on a budget, we see that our sweet spot for gaming is indeed 2x4GB of Kingston FURY DDR3 at 1866MHz for $80. Paying $110 for the faster Beast 2133MHz does give slight advantages, but one would have to be an overclocking and performance-oriented enthusiast to pay $220 for 8GB of 2800MHz Predator premium RAM, or else work with video encoding regularly.
Kingston Predator HyperX FURY 1800MHz memory – Pros and Cons
- The Kingston HyperX FURY RAM is very fast automatically overclocked 1866 MHz DDR3 that is built for gaming, and it is a great replacement for HyperX Blu.
- FURY comes with a lifetime warranty along with Kingston’s great customer service and tech support.
- If necessary, Kingston RMAs are quick, professionally handled, and painless.
- Choice of 4 colors
If you are a gamer with premium components who is looking for great performance without spending a fortune for memory, then the Kingston HyperX FURY DDR3 at 1866MHz is an excellent choice. We have tested 1300MHz and even 1600MHz memory, and if you are running at those speeds, overclocking or upgrading may be in order if you have an Ivy Bridge platform, or newer. For Haswell platform Core i7, 1866MHz to 2133MHz seems to be the sweet spot for bang-for-buck gaming performance, and FURY is an excellent choice for gamers and enthusiasts.
The Kingston HyperX FURY RAM is priced about the same as other manufacturer’s high-quality 1866MHz DDR3, but it is backed by Kingston who gives a lifetime warranty and excellent support. We feel it deserves ABT’s Great Value Award. We have had an opportunity to RMA a Kingston HyperX SSD recently, and we got amazing service. The turnaround was less than a week – and the Kingston representative did not know we were a tech site until after the RMA was processed!
If you are on a budget, you are more likely to spend less on faster RAM speeds. Many budget conscious enthusiasts settle for less highly-clocked RAM, or perhaps try for a manual overclock. For gaming, 8GB is sufficient, although for other tasks that involve multi-tasking, 16GB or more of RAM might be preferable, as we saw in some cases from using our 16GB Beast RAM modules. Kingston’s HyperX Predator memory at 2800MHz is amazingly fast, and it is for those who demand the very highest performance from their PC components.
FURY is available in 4 colors for those who wish to color-coordinate their PC, and we think that White looks great.
For ABT’s flagship Core i7-4770K PC, running everyday tasks and for most of our benchmarking, we have settled on using 16GB of Kingston HyperX Beast DDR3 at 2133MHz. It is slightly faster than the FURY stock speeds of 1866MHz; it brings advantages of 16GB over 8GB, and the DIMMs are much lower profile than the Predator RAM which allows us to easily test more CPU coolers. For testing extreme overclocks, or for extracting every bit of performance out of our Haswell platform, we will use the HyperX Predator RAM at 2800MHz.
For everyday tasks, and especially for gamers, we would recommend Kingston’s HyperX FURY 1866MHz DDR3 as being excellent bang-for-buck!
We are looking forward to publishing our next evaluation early next week which will feature Optical Storage as we compare OWC’s latest external 16x BluRay to our older ASUS 12X internal writer. We will also post our impressions of SHIELD and Half Life 2, as well as a Kingston 240GB mSATA evaluation. And we are still working on FCAT. Stay tuned!
In the meantime, make sure that you follow us on our ABT forum.
ABT Senior Editor