G.Skill RIPJAWS DDR3-1600 CL7 1.6V Dual-Channel Kit

If you're in the market for some DDR3 memory these days, you will see a lot of sets being advertised as "Optimized for use in Intel i5 and i7 systems". There's a lot of confusion in the marketplace about whether these products are suitable for use in other applications. The truth is that most all of them can be used in any system that uses DDR3 memory, as long as the BIOS on the motherboard can supply the correct voltage, usually around 1.6V for most sets. In this article Benchmark Reviews will examine how a dual-channel kit of 1600MHz CL7 RIPJAWS modules from premier maker G.Skill performs on an AMD 790FX platform.

This pair of DDR3 DIMMs is rated at 1600MHz with tight timings of 7-8-7-24 at a measly 1.60 volts. In keeping with their target application in the marketplace, they feature an XMP profile on the F3-12800CL7D-4GBRM model, which can set these SPD timings in the BIOS automatically on Intel-based P55 and X58 series motherboards.

We're going to look at several memory clock and timing configurations, to see how flexible this new low-voltage kit is on an AMD AM3 platform, with a Black Edition CPU that has an unlocked multiplier. We'll also try to overclock them a bit and see if we can wring any additional performance from them.

G.Skill RIPJAWS DDR3 Features

The Ripjaws product line from G.Skill is relatively new for them. It is aimed toward the gaming market, where many products combine a healthy dose of performance with a bit of attitude. G.Skill is no stranger to performance memory; their Pi series is well regarded in enthusiast circles for both stability out of the box, and being very willing performers when overclocked.

• G.Skill Ripjaws Gaming Series Memory is designed to optimized the high performance DIMMs for reliability in order to get gamers consistently fantastic FPS, while maintaining a solid overclock to prevent a mid-game PC failure from memory errors.
• G.Skill Ripjaws Gaming Series Memory features the latest heat spreader design for the ultimate in extreme aesthetics and sleek aluminum cooling.
• G.Skill Ripjaws Gaming Series Memory features the usual industry leading limited lifetime warranty and ever ready technical backup that all G.Skill performance memory is sold with.
• Ripjaws series memory was designed specifically to complement Core i7 processors and the X58 Express Chipset.
• Optimized for speed, low latency and high stability, Ripjaws memory is the perfect solution to faster programs and quicker load times.
• Heat management is always important for any PC builder and enthusiast. G.SKILL understands this concern, which is why each Ripjaws Series DIMM comes with a stylish comb-like design heat-spreader, which dissipates heat by exposing it to cool air over a greater surface area and will look great in any case. These DIMMs also operate at a cool 1.6V for better internal temperatures and overclocking versatility.

F3-12800CL7D-4GBRM Specifications

• Brand: G.SKILL
• Series: Ripjaws Series
• Model: F3-12800CL7D-4GBRM
• Type: 240-Pin DDR3 SDRAM
• Capacity: 4GB (2 x 2GB)
• Speed: DDR3 1600 (PC3 12800)
• Cas Latency: 7
• Timing: 7-8-7-24
• Voltage: 1.6V
• ECC: No
• Buffered/Registered: Unbuffered
• Multi-channel Kit: Dual Channel Kit
• Heat Spreader: Yes
• Features: Specifically Designed to compatible with Intel Core i5 and Core i7 for Intel P55 motherboard
• Manufacturer Warranty: Parts and Labor, Lifetime limited

About G.SKILL

Established in 1989 by enthusiast, is a leading memory manufacturer based in Taipei, Taiwan. The company's top priority is Quality. All of our products go through a series of the most rigorous tests and strict quality control processes. In addition to commissioning qualified IC testing houses to test our products, they are hand test 100% twice in factory and office, to ensure the highest product yield and quality.

G. SKILL strives to achieve the highest and most advanced quality from the initial design, through manufacturing solder-paste printing, through surface mounting, to on-line visual inspection, system compatibility testing, packaging and finally to safely and reliably shipping our products to customers.

G.Skill operates a very active support forum for all their products, where their product specialists are quick to respond to issues, real or imagined. They also monitor major e-tail sites for customers having trouble with their products, and try to help them get on the right path towards resolution of their problem.

Closer Look: G.Skill RIPJAWS DDR3-1600

There is no mistaking the design of the G.Skill Ripjaws series for anything else. It has a unique look that evokes the product name in a very graphic manner. I'm very happy with my OCZ Reapers, but do they look like Reapers? Not in the least. These modules look like Ripjaws...no doubt about it. They are a bright blue, which you might not associate with jaws, but they're just about the same color as high-chromium steel after it's been heated and cooled. So, jaws and blue steel, yeah it gets my attention.

The heat spreaders on the Ripjaws are able to effectively clamp down on the memory chips without the typical spring clips that many DIMMS sport, including many G.Skill products. Unless there are some hidden screws underneath the Ripjaws label, they've come up with an innovative and stylish way to fasten the heat spreaders in place. None of the modules in this series uses any color other than green for the PCBs; I think black PCBs would look nice for the red or black versions, but that's just me.

After we get past the visuals, let's take a look at what else we have here. This is a dual-channel kit, so it's geared towards the LGA1156 (P55) platform instead of the LGA1336 (X58) motherboards that feature a triple-channel RAM interface. What makes it specifically suited for that application is the Intel Extreme Memory Profile (XMP). This a set of SPD (Serial Presence Detect) settings that work the same way as the NVIDIA EPP scheme, for plug and play memory settings above and beyond the normal JEDEC standards. For an explanation of the benefits, see one of our recent RAM reviews that used an LGA1156 platform for testing.

As I mentioned in the intro, we are going to take a slightly different approach in this review, and see how well these kits perform on the AMD AM3 platform. The standard JEDEC values reported by the memory modules are shown here, along with the single set of XMP values in the right hand column. At the standard voltage of 1.5V, CAS Latency keeps rising as the memory clock goes up. It takes a wee bit of extra voltage to get the timings down to where they need to be for a set of gaming sticks. The i5 and i7 memory controllers can't stand voltages above 1.65V, so memory makers are doing everything they can to get high frequencies and tight timings on a restricted voltage budget. The AMD AM3 platform doesn't have that restriction, and there are plenty of memory modules that will run quite happily at 1.9V and above in this environment. We want to see how the low voltage chips run, though, so let's see what we can get out of these modules at stock voltage, or at most a reasonable overvolt.

The CPU-Z screenshot below shows the best I could do for timings at 1600MHz with a slight overvolt of 1.64V. I actually took the voltage quite a bit higher, trying to achieve the XMP settings, but could not get these modules below CL8 at 1600 MHz or higher clocks. The extra volts didn't seem to help, so I throttled back to 1.64V for the remainder of the testing, and both the 1600MHz and 1744MHz overclock were stable at this voltage. I tried for 1800 MHz, but was not able to reach that, no doubt a direct result of the binning process all the memory manufacturers use to identify higher performing chips, that can be sold at a premium.

So far, I'd say my experiment has been a qualified success. I was not able to hit the XMP profile settings, but I was able to overclock the RAM beyond the stock maximum frequency at timings just one notch above the XMP profile, at 8-8-8-24. I was also able to run tighter timings than stock at lower frequencies, even at the nominal voltage spec of 1.5V. So, now that we have defined some stable configurations. let's move on to the testing portion of our review where we see what sort of gains we have or have not achieved with these various memory profiles.

RAM Testing Methodology

All benchmarks are conducted using the same system components and the same memory modules in the same DIMM slots. The memory slots sit directly between the main power connector and the CPU socket on the motherboard. Some systems work better when the memory modules are closer to the power source, and some perform better when the RAM is closer to the CPU. The manufacture of the motherboard used for testing in this review states their preference in the manual: "Install the RAM in the sockets closest to the power connector for better over-clocking capability."

Each benchmark begins after a complete system restart and is repeated five times; the high and low results are discarded, and the average of the three remaining results is calculated and reported in the text. To ensure system stability and the reliability of our results, each new memory configuration was fully tested with Memtest86 v4.00.

One of the advantages of an AMD based-system for memory testing is the wider availability of unlocked CPUs, allowing the FSB and CPU multiplier to be raised and lowered in tandem, so that the CPU clock remains the same while the FSB and base memory clock is increased.

One of the disadvantages of an AMD-based system is a problem that cropped up with the release of the Phenom II chips and the AM3 platform. Namely, there was a "Product Errata" called: #379, DDR3-1333 Configurations with Two DIMMs per Channel May Experience Unreliable Operation. The name is unfortunate, in that it doesn't exactly convey the fact that all memory clocks greater than DDR3-1066 were impacted, not just DDR3-1333. Luckily, time heals all wounds, and there are dozens of CPUs and motherboards today that do not experience any problems with these higher speeds.

Below is a table summarizing the hardware settings used in this review. I stuck with JEDEC standard frequencies, but experimented with tighter timings than the SPD values embedded in the modules. At all the standard frequencies I left the FSB base clock the same, at 200 MHz. When I started to overclock the ram, I had to raise the FSB, but I reduced the CPU multiplier to keep the CPU frequency at 3.6 GHz.

Test System

• Motherboard: ASUS M4A79T Deluxe (2205 BIOS)
• System Memory 1: 2X 2GB OCZ Reaper HPC DDR3 1600MHz (7-7-7-24)
• System Memory 2: 2X 2GB G.Skill Ripjaws DDR3 1600MHz (7-8-7-24)
• Processor: AMD Phenom II 720 Black Edition (Overclocked to 3.8 GHz)
• CPU Cooler: CoolerMaster Hyper Z600
• Video Adapter: ATI Radeon HD5770, Engineering Sample
• Drive 1: G.Skill Titan SSD, 128GB
• Optical Drive: Sony NEC Optiarc AD-7190A-OB 20X IDE DVD Burner
• Enclosure: CM STORM Sniper Gaming Case
• PSU: Corsair CMPSU-750TX ATX12V V2.2 750Watt
• Monitor: SOYO 24" Widescreen LCD Monitor (DYLM24E6) 1920X1200
• Operating System: Windows 7 Ultimate Version 6.1 (Build 7600)

Benchmark Applications

• Passmark Performance Test v7.0 Build 1011
• EVEREST Ultimate Edition v5.30.1900
• SiSoftware SANDRA v2009.9.15.124
• Crysis v1.21 Benchmark Tool
• Memtest86 v4.00
  

Performance Test Results

Four benchmark applications for memory performance have been in rotation here at Benchmark Reviews for some time now, and there are no new contenders that offer any more or better information: Passmark Performance Test, Lavalys EVEREST, SiSoftware Sandra, and Crysis. The first three are synthetic benchmark suites specifically targeted at several aspects of memory performance. Each one has a unique approach, which provides a diverse set of measurements so that performance trends are brought to light. The last benchmark, Crysis, offers insight into how memory performance affects a gaming application that stresses the CPU and memory almost as much as it does the graphics subsystem. CPU speed is always a factor in memory tests, and we did our best to eliminate it as a variable. During overclocking, we had to adjust the Northbridge clock frequency, which has a halo effect on the overall system, but we were able to keep the CPU clock the same.

In Passmark Performance Test, there were very minimal gains, either from higher clock frequencies or tighter timings. The cached memory read test saw literally no difference between the five tested configurations; any differences are buried by experimental error. The uncached read test scored less than 1% improvement between the 1066 MHz and 1744 MHz settings. One of the nice aspects of this benchmark is the consistency of the results, I feel confident that even the small improvement measured here is real and repeatable.

The write performance was the bright spot of this test, clocking in a 3% gain as clock speed increased. Once again, the results were very consistent for this test, and while 3% may not seem like a lot, at least it is real, measureable and repeatable. I am hoping for more differentiation in the remaining tests, though.

EVEREST Ultimate Edition offers three simple memory bandwidth tests that focus on the basics; Read, Write, and Copy. In order to avoid concurrent threads competing over system memory bandwidth, the Memory benchmarks utilize only one processor core and one thread.

The Everest Read benchmark measures the maximum achievable memory read bandwidth. The code behind this benchmark method is written in Assembly and it is extremely optimized for every popular AMD and Intel processor core variants by utilizing the appropriate x86, MMX, 3DNow!, SSE, SSE2 or SSE4.1 instruction set extension. The benchmark reads a 16 MB sized, 1 MB aligned data buffer from system memory into the CPU. Memory is read in forward direction, continuously without breaks.

In Lavasys Everest we see more dramatic performance differences between the speed settings, and we can also see the effect of timings. From best to worst, there is a 25% improvement in read performance. We can also see how the tighter timings that were achieved at 1333 MHz almost made up the difference in speed between 1333 and 1600 MHz. It's also interesting to note that the timing changes at 1066 MHz made very little difference.

The Everest Write benchmark measures the maximum achievable memory write bandwidth. The code behind this benchmark method is written in Assembly and it is extremely optimized for every popular AMD and Intel processor core variants by utilizing the appropriate x86, MMX, 3DNow!, SSE or SSE2 instruction set extension. The benchmark writes a 16 MB sized, 1 MB aligned data buffer from the CPU into the system memory. Memory is written in forward direction, continuously without breaks.

The write performance is relatively flat, as speed settings increase, until we get to the overclocked configuration, where we were able to bump up the memory clock by increasing the Front Side Bus (FSB) by 9%, from 200 to 218 MHz. We reduced the CPU multiplier to keep the CPU clock the same, but as most people know, increasing the FSB clock makes almost everything faster. In fact, the best performance is usually achieved by pushing the FSB even higher and using a lower FSB:DRAM strap. But that's not a fair way to test memory products...

The Everest Copy benchmark measures the maximum achievable memory copy speed. The code behind this benchmark method is written in Assembly and it is extremely optimized for every popular AMD and Intel processor core variants by utilizing the appropriate x86, MMX, 3DNow!, SSE, SSE2 or SSE4.1 instruction set extension. The benchmark copies an 8 MB sized, 1 MB aligned data buffer into another 8 MB sized, 1 MB aligned data buffer through the CPU. Memory is copied in forward direction, continuously without breaks.

Copy performance was influenced the most by cranking up the memory clocks. We achieved a 38% increase in performance on this benchmark, which seemed to depend mostly on clock speed and less on clock timings. Overall, there were some significant performance gains to be had in the Everest set of benchmark tests. Not bad for a product that is supposedly optimized for a completely different operating environment. So far I see no reason that these new, low voltage RAM sets can't be used to good effect on the "old" AMD platform.

Sandra is based on STREAM, a popular memory bandwidth benchmark that has been used on personal computers to super computers. It measures sustained memory bandwidth not burst or peak. Therefore, the results may be lower than those of other benchmarks. STREAM 2.0 uses static data (about 12M) - Sandra uses dynamic data (around 40-60% of physical system RAM). This means that on computers with fast memory Sandra may yield lower results than STREAM. It's not feasible to make Sandra use static RAM - since Sandra is much more than a benchmark, thus it would needlessly use memory.

A major difference is that Sandra's algorithm is multi-threaded on SMP/SMT systems. This works by splitting the arrays and letting each thread work on its own bit. Sandra creates a thread for each CPU in the system and assigns each thread to an individual CPU. Another difference is the aggressive use of scheduling/overlapping of instructions in order to maximize memory throughput even on "slower" processors. The loops should always be memory bound rather than CPU bound on all modern processors.

The other major difference is the use of alignment. Sandra dynamically changes the alignment of streams until it finds the best combination, then it repeatedly tests it to estimate the maximum throughput of the system. You can change the alignment in STREAM and recompile - but generally it is set to 0.

The results from SiSoft Sandra look a lot like the Read performance results in Lavasys Everest. They scale more as a result from increasing clock speeds than clock timings. Interestingly, the Integer and Floating Point results are almost identical, and the individual results were also very consistent from run-to-run. The overclocked pair, running 1744 MHz at CL8 bested the 1066 MHz CL7 set by 50% in both tests. That's a pretty significant gain, and a testament to the strength of the memory controller built into the AMD Phenom II architecture.

Crysis needs no introduction on this website. It is well known as one of the most demanding benchmarks, and our move to DirectX 10 has only increased the overall difficulty of achieving reasonable frame rates at high resolutions. In this scenario, where we want to reduce the influence of the video card in the results, we are primarily interested in the low resolution tests, and minimizing the video processing that is handled by the graphics subsystem.

Starting on the right and moving to the left, we can see that at 1680x1050 and 1280x1024 resolutions, there are minimal differences in gaming performance with changes in memory. Concentrating on the lowest resolution we tested, 1024x768, there is a noticeable, 12 FPS difference in average frame rate between the lowest and highest performing memory configurations. I say noticeable, meaning that it is easily measured; I doubt that you or I could visually tell the difference between an average of 109 and 121 frames per second in Crysis.

Overall, the synthetic tests mostly showed measureable performance improvements from increased memory speeds and tighter timings. Our toughest gaming benchmark, in terms of CPU and memory usage only showed measureable changes at low resolution. But, as GPU power increases in the system, this influence will be felt at higher resolutions. Similarly, if you are still using DirectX 9, where the GPU has an easier task, the impact will be greater.

We're left with the question of value, then. How much difference does premium, high speed memory make, especially compared to investing money in other system components. Continue on to Final Thoughts for the answer to that question, and a discussion of how I really feel about XMP and other memory standards.

G.Skill RIPJAWS Final Thoughts

Who spends over $100 on DDR3 memory and then runs games at 1024x768 screen resolution? Nobody. So, why should you spend the extra cash for premium, high speed, low latency RAM when Crysis couldn't care less? The answer is hidden in the testing details above, where I explained how I went to all kinds of trouble to make sure the CPU clock didn't vary during these tests. In the real world, where we're not testing, we're optimizing; every attempt is going to be made to get both the CPU clock and the FSB clock as high as possible, within the limits of overall system stability. In that case, you need as much flexibility as you can get in memory clock speed, so you aren't prevented from dialing an extra hundred, or couple hundred megahertz into one of the clocks that DO matter. Once you get those clocks maxed out, you mess with the FSB:DRAM straps and the timings to get the maximum possible performance from your memory subsystem. Doing this will definitely gain you some substantial increases in performance on most all your gaming applications, unless the system is severely GPU limited, and we know that's not going to be the case, don't we...?

On another note.... I have a love-hate relationship with standards. On the one hand, I love industry standards, especially when there is a strong, unified, forward thinking agency that can get out ahead of the product development curve and provide some stability for the marketplace. Can't think of one? Well, most good standards organizations are a victim of their own success. The reason you don't think of them is because you don't have to; they're just quietly doing their thing behind the scenes, saving you from the death throes of incompatibility. On the other hand, I hate proprietary standards, like EPP (NVIDIA), XMP (Intel), and Black Edition Memory Profiles (AMD). They are all mutually exclusive, and force you to pick sides and commit to a certain platform for your memory purchases. I mean, I understand why AMD and Intel CPUs don't fit in the same socket, and require unique chipsets and motherboards to support them, but these memory profiles are just extensions of the existing JEDEC SPD (Serial Presence Detect) standard. Yet, they are all incompatible with each other. Wonderful, eh...?

G.Skill F3-12800CL7D Conclusion

One of the questions we had going in to this review is, how well do some of the new low voltage DDR3 kits work on the AMD AM3 platform? Do DIMMS designed for P55 and X58 suffer when taken out of their native environment? In some ways they do, in that AM3 boards do not currently support the Intel XMP (Extreme Memory Profile) spec, so there is no explicit guarantee that the modules will perform at the specific performance level they're rated for. On the other hand, these modules had no problem going well past their official speed rating at slightly looser timings. The performance is still there, it's just not a cut-and-dried, plug-and-play experience to achieve it.

The Ripjaws definitely look the part of gaming kit, and G.Skill deserves a lot of credit for going outside the box with their design. The low operating voltages involved has given module manufacturers a little more wiggle room in the thermal dissipation department, and G.Skill made the most of it. Just like with their other products, they have color coded the modules within the product line, but gone are the pastel blues and greens from their standard line, replaced with a bold, vibrant color scheme. The two slightly taller "teeth" on the ends of the assembly interfere with the low profile a bit, but they still fit under my CoolerMaster Hyper Z600 CPU cooler in the first set of DIMM slots. I think the appearance fits the target market perfectly; performance with a bit of an attitude.

Construction quality of the G.Skill Ripjaws series is tough to judge by eye. Most of the module is covered up by the heat spreader, and this is one assembly I'm not going to tear down. So you're left with the lifetime warranty and the reputation of the brand to go on. G.Skill comes up aces on both counts, with excellent product support to boot, if you just have a problem with how to use the product to its full potential.

Functionality, for a memory product, is kind of a limited bag. In my testing, they ran cool, they ran at all the speeds, timings, and voltages I expected them to. I can't fault them for not running their XMP settings on an AMD board; that was my own way of pushing them outside of their envelope. One aspect I appreciated was their semi-low profile. They are taller than a bare stick, but they still fit into the first pair of slots next to a monster of a CPU cooler. I did some off-hand testing with 8GB of memory, and it was good to have some DIMMS that would fit into that spot.

With a retail price of $104.99 at Newegg, the 4GB G.Skill F3-12800CL7D-4GBRM kit is reasonably priced; the lowest price 4GB, 1600 MHz set is $92.99, and that's for a CL9 pair that wouldn't perform nearly as well. The highest priced set was $169.99, so you can see that this pair is definitely priced well below the median. I honestly think they're a better value if you put them in a P55 motherboard, and take advantage of the XMP memory profile, but if you want to put them in an AMD-based system, there's nothing to stop you. Just a little extra work on your part to get them running at maximum performance.

G.Skill has a winning proposition with this new series. They've hit a sweet spot of value, performance, and style that puts them on the leader board where everyone can see them. They haven't been invisible in the past, at least not to enthusiasts, but they lacked widespread brand recognition in the gaming marketplace. The Ripjaws product line is going to change all that, I think.

Pros:

+ 1600MHz rating
+ Low latency
+ XMP profile is useful for many
+ Lifetime warranty
+ Looks matter, right?
+ Value
+ Product Support from G.Skill
+ Shorter profile than some RAM

Cons:

- Proprietary memory standards (...not G.Skill's fault)
- Taller profile than many DIMMs

Ratings:

• Performance: 9.00
• Appearance: 9.25
• Construction: 8.75
• Functionality: 8.75
• Value: 9.00

Final Score: 8.95 out of 10.

Excellence Achievement: Benchmark Reviews Silver Tachometer Award.

Questions? Comments? Benchmark Reviews really wants your feedback. We invite you to leave your remarks in our Discussion Forum.

Related Articles:

• Patriot Gamer-Series DDR3-1600 Memory Kit
• 6GB Kingston HyperX 2000MHz DDR3 Memory Kit
• Kingston HyperX Genesis 16GB DDR3-2133
• G.SKILL 2133MHz DDR3 Pi-Series Memory Kit
• ASUS ENGT240 1GB GDDR3 Video Card
• Mushkin PC3-10666 DDR3 1333MHz RAM Kit HP3-10666
• G.Skill ECO 4GB DDR3 Memory Kit F310666CL7D
• Gigabyte GA-MA770T-UD3P AM3 Motherboard
• 6GB Corsair Dominator-GT 2000MHz DDR3 Memory
• G.Skill Triple-Channel 1600MHz DDR3 Memory Kit