2500MHz Patriot Viper-II Sector-5 DDR3 Kit
Benchmark Reviews recently tested a
2133MHz G.Skill memory kit. The results showed that in real-world applications, there's not a lot we can receive from that kit compared to lower frequency ones. However, there are some applications that take advantage of high-end kits giving a small boost to the system increasing memory bandwidth and thus, overall speed. While 2000+ MHz kits are becoming the standard for DDR3, there's still a big gap between 2133MHz and 2500MHz. Most CPUs won't be able to run such high frequencies and motherboards could also be a limiting. Even if Benchmark Reviews proved a high-frequency kit isn't that necessary, Patriot Memory sent us their best kit for review. They really want to show their kit is a must for high-end enthusiasts and overclockers, and Benchmark Reviews is putting those words into test.
Running at 2500MHz, the Viper II 'Sector 5' Edition 4GB kit is one of the fastest kits on the market available. This kit is completely intended to be used per high-end enthusiasts and overclockers, nothing else. The reason is simple. First, to run at 2500MHz you need a very strong IMC (integrated memory controller) and you won't find something like this in a Core i3/Core i5 processor. Even with the newest Core i5 655K Unlocked processor, which allows you to enable a 2133MHz memory multiplier, you won't be able to pass 2200-2300MHz. Since you're limited to a 1600MHz memory multiplier with Core i7 CPUs, you need a 209MHz BCLK, which isn't possible with low-end and some mid-end motherboards.
Patriot ships their Viper II Sector 5 Edition kit inside a slim red box. There not much information about this kit in the package, but if you're buying this kit, you should really know what you're buying or you must have a lot of money to spend on a memory kit that you won't benefit from. Today, we will test this kit in a Core i7 860 system paired with a high-end P55 motherboard which allows some extra tweaking inside its BIOS.
Let me tell you something: when I received this kit for review, I was actually a little bit afraid to be able to reach the 2500MHz barrier. I had never tested such a high-frequency kit like this, and there aren't a lot of reviews bragging about this kit even after some weeks of being launched. To be prepared, I took a pair of high-end motherboards and 2 different CPUs to test in order to avoid any limits. I was even prepared to get a Core i7 875K CPU if needed, or just to admit I couldn't reach 2500MHz and let users know it wasn't an easy task. Finally, after fighting some days with this kit, I was able to reach the 2500 magic number, and actually, get higher frequencies than some other technology-sites with some extra tweaking and knowledge. So, let's go ahead to the features and specifications...
Manufacturer: Patriot Memory
Product Name: Viper II 'Sector 5' Edition 2500MHz 4GB
Model Number: PVV34G2500C9K
Full Disclosure: The product sample used in this article has been provided by Patriot Memory.
Patriot Viper II Sector 5 Specifications
Patriot shows close to nothing about features and specifications in their
website. Even some big retailers like Newegg are showing little info about this kit, but I've searched and joined all the information I could in the next list to help any user buying this high-end kit.
Mainboard: Intel LGA 1156
System: Desktop
System Type: DDR3
M/B Chipset: Intel P55
CAS Latency: 9-11-9-27
Capacity: 4GB (2GBx2)
Speed: DDR3-2500 (PC3 20000)
Test Voltage: 1.65-1.75 Volts
Registered/Unbuffered: Unbuffered
Error Checking: Non-ECC
Type: 240-pin DIMM
Part Number: PVV34G2500C9K
Patriot Viper II 'Sector 5' PVV34G2500C9K Features
According to Patriot's website, Viper II Sector 5 series are described in the following paragraph:
The Patriot Viper II Series, 'Sector 5' Edition, part of the Extreme Performance (EP) Enhanced Latency line, is the ultimate performance memory solution for the Intel® P55 system platform. The Viper II Series, 'Sector 5' Edition is built using only the highest quality pre-sorted IC's available and are subjected to rigorous testing and validation on the P55 platform to achieve optimized performance and maximum quality. These modules are engineered to reach PC3-20000 (2500MHz) at 9-11-9-27 timings and are available in 4GB kit capacity. Each module is 100% hand-tested for quality assurance and is backed by a full lifetime warranty and industry leading customer service.
- Equipped with advance aluminum heat-dispersing shields
- RoHS Compliant
- Tested on Intel P55 chipsets
- Intel Core i7 CPUs Tested and Qualified on ASUS P7P55D motherboard
- Extreme Performance PC3-20000 (2500MHz)
- Enhanced Latency (9-11-9-27)
- Voltage: 1.65v-1.75v
- Lifetime Warranty
Closer Look: Patriot Viper II 'Sector 5' PVV34G2500C9K
Leaving aside the simple package included, the Viper II Sector 5 memory DIMMs are pretty cool. If you love the red-black schemes, you'll quickly fall in love with this kit. Patriot uses 100% black PCBs in their Sector 5 Modules with a Red stripe and white letters. This kit makes a good combination for those users with ATI cards and black PC cases.
Here's a closer look to the heat-spreader. As you can see, there are to rows of black fins to help heat dissipation, especially because you're meant to run high voltages on this kit. Some of the fins came with little scratches and the paint fell of the fins, but there's nothing to worry about since this is almost impossible to see without a magnifying glass.
Here's a take from another angle to the heat-spreaders. I must say I really like the color, but even that, I still prefer low-profile heatsinks to avoid CPU heatsink installation problems with RAM clearance, especially with AMD and LGA 1156 systems, as the space between RAM slots and CPU socket is way shorter compared to LGA1366 motherboards.
Bringing out my magnifying lens, we can see a detailed shoot to the PCB and heatsink. As you can note form the next image, the heat-spreader is actually divided in two parts, and they're both glued to the ICs. I normally hate this kind of systems as you can't remove heat-spreaders without voiding warranty. Also, when I was installing the second module on my motherboard, one half of the heat-spreader came off the RAM DIMM without excessive force used. That's high quality manufacturing!
Here's on last shot with both modules. To install the fallen heat-spreader, I just had to install the RAM and stress it 20 minutes to heat them and past it again. That was possible just because the spreader didn't fall off completely, but in other case, I would have needed to take some glue/thermal paste to install it again.
Patriot Viper II 'Sector 5' PVV34G2500C9K SPD
As it is usual, at Benchmark Reviews we love to show the whole SPD to our readers. These images help users to watch how their memory will work with and without an Xtreme Memory Profile when installing their RAM for the first time. Normally, brands write very light-loaded SPDs in order to maximize compatibility in every motherboard used. Have a look at CPU-Z SPD tab:
In this case, Patriot added 900MHz 6-6-6 and 1066MHz 7-7-7 JEDECs working at 1.5 volts. I actually find those values a little bit low, especially for this kit. Many motherboards support up to 1333MHz without issues, but I won't cry a lot about this since they already included a 2500MHz XMP working at 9-11-9-27 2T timings with 1.65 volts for users who just want to install their kit and set the XMP instead of taking some extra time for tweaking and manually set every parameter in the BIOS.
Just to corroborate, I always open CPU Tweaker and check timings, but this time I was rather confused when I saw the XMP shown with 10-12-10-28 Timings. It must be a bug and nothing else though...
While some users think testing stability at default settings can be taken as an insult to the brand, I normally test each DIMM to verify if there are any errors produced by shipping damages or any other stuff. Today, this is not just to prove that, but also to make sure I'll be able to reach 2500MHz 100% stable with the motherboard and CPU used as it's not an easy task, especially if you're not used to RAM overclocking.
Patriot Viper II 'Sector 5' Voltages
There have been lots of discussions regarding RAM Voltage with Core i7 processors. Since Nehalem's launch, Intel limited maximum "safe" voltage to 1.65 volts. They said you could permanently damage your CPU IMC by using higher voltages, thus creating fear between users and forcing memory brands to start shipping low voltage kits. Nowadays, it's very common to see 1.5v kits, and even low-voltage "eco" kits working at 1.3v without a hassle. Today, this is going to change, as Patriot specs their Viper II Sector 5 2500MHz kit to be used with 1.65v and up to 1.75v if necessary. Many users have tried using higher volts to overclock their kits, and it seems nothing has damaged their CPUs, so I'm quite sure you won't have problems by adding 0.1 volts to the maximum Intel's specifications.
Now, before we start testing this kit, I really wanted to verify how much voltage they needed to work at their rated speeds and timings, so I took my digital multimeter and measured CPU QPI/VTT/IMC voltage and vDIMM voltage while running Memtest to see real voltage values needed to reach a stable point.
For QPI/VTT/IMC voltage, the Xtreme Memory Profile automatically increases this value from 1.15-1.2v (stock) to 1.3 volts. My motherboard suffered from little overvoltage, so the actual value was 1.38v, but it seems 1.34v were enough for 2500MHz.
The XMP automatically sets the vDIMM to 1.75v, and my motherboard showed some under-voltage running 1.74v real volts. Trying to lower this voltage to 1.7v made this kit a little bit unstable for Memtest and other benchmarks, and lowering this voltage to 1.65v resulted into a completely unstable kit, so make sure your motherboard is able to supply 1.75v at least to run this kit 100% stable, otherwise, you won't reach its maximum potential.
Also, remember you need /at least) a 209 BLCK capable motherboard to reach 2500MHz with a 1600MHz memory multiplier. Otherwise, your maximum memory frequency will be capped. You'll probably need to lower your CPU multiplier a little unless you want to run 4GHz+ frequencies, and you might need to increase CPU vCore. For more information, Benchmark Reviews recently published an
Intel LGA 1156 Core i3/i5/i7 Overclocking Guide.
Testing Methodology: DDR3 RAM
At Benchmark Reviews we need to test memory without adding any other factors to the results. While testing stock speeds is usually easy, it only gets more difficult when we want to add overclocked results. We need to make some changes to the BIOS values and test each configuration to ensure maximum stability. First, I set the values on the BIOS and run Memtest to check if the kit is stable or not. After that, I test with OCCT Perestroika to check if the CPU is stable along with the whole system.
Finally, Benchmark Reviews decides which frequencies are going to be shown in the graphics in order to leave the CPU factor outside of the equation. This way you'll probably see small differences between memory frequencies, latencies and voltage reflected on our charts without the CPU interfering in the results. Benchmark Reviews will use the following tests suites and applications:
Passmark Memory Test v7.0 x64
SiSoft Sandra Lite 2010
Lost Planet 2 DX11 Benchmark Test B
WinRAR 3.9 x64 Benchmark
KitGuru V1(4) Adobe Photoshop CS5 Script
Test System
Motherboard: ASUS Maximus III Formula P55 (2001 BIOS)
System Memory: 2x2GB Patriot Viper II 'Sector 5' Edition 2500MHz
Processor: Intel Core i7 860 with Noctua NH-D14 heatsink
Video: Nvidia Geforce 460GTX SLI @ 800MHz GPU core
Disk Drive 1: Intel X-25M G2 80GB
- Disk Drive 2: Seagate Barracuda 500GB
Enclosure: Corsair Obsidian 700D
PSU: Antec Signature 850 watts
Before even starting this journey, I found I couldn't reach 2500MHz. Actually I couldn't pass 2250MHz before ending with a completely unstable setup. I tried everything I could: raising voltage, lowering CPU multiplier, adding QPI/IMC voltage, relaxing timings etc., but everything was worthless. Finally, before taking out my second motherboard ASUS decided to launch a new BIOS version (2001), which said to increase memory compatibility and avoid high-frequency errors, so I gave it a try. It resulted the old BIOS (1801) just hated this kit, because when I flashed the Maximus III Formula with the newest BIOS, I set the XMP and the Viper II Sector 5 Kit just booted at 2500MHz without problems. I just had to do some tweaking here and there to make it 100% stable, but it wasn't difficult. According to my experience, I really recommend to check your motherboard's manufacturer and get the latest BIOS available before installing this kit. Actually, since Patriot only qualifies the ASUS P7P55D motherboard for this memory, I wouldn't recommend you to try it with non-expensive brands because they don't guarantee reaching 2500MHz. If you're looking for a high-end motherboard to pair with the Sector 5 Edition kit, I'd recommend looking at ASUS or GIGABYTE's products, as they offer some of the best overclocker products at the moment.
With the ASUS Maximus II Formula, I'm capable of reaching very high CPU/Memory frequencies and tweak voltages and different memory settings to make them completely stable. This time, I decided to include 5 different configurations, all of them overclocked, as this kit screams to be used with a high-end system and overclocked settings. All of these configurations were tested with the Core i7 860 running at 3.8GHz with a 10MHz tolerance between different setups. I included the most common 1333MHz setup, and started to overclock going to 1600MHz, 2000MHz and up to 2500MHz with different timings. I must add that all of these settings were tested with relatively tight latencies (against memory frequency), except the 1333MHz setup which I tested with 9-9-9 latencies.
1333MHz 9-9-9-27 1T 1.75v
1600MHz 7-7-7-21 1T 1.75v
2000MHz 7-9-7-24 1T 1.75v
2400MHz 8-11-8-27 1T 1.75v
- 2500MHz 9-11-9-27 1t 1.75V
Note that all the setups used in this review use a Command Rate of 1T. This kit proved to be great as it didn't needed to use a 2T CR to reach super-high frequencies, even if the SPD was programmed with a 2T command rate. In the next section, I test and analyze synthetic benchmarks at different frequencies.
Benchmark Tests Results
Benchmark Reviews runs each tests at least 3 times to avoid special situations where scores are over/under exposed. Passmark Performance Memory Test does some bench to the memory and reports a test score. I´ve chose Memory Mark along with uncached read and write speed since they´re the most important benchmarks for memory testing on this suite.
As it becomes clear, there is no special advantage between 1333MHz and 2500MHz in this test. While there is a 2.5% increase in all the tests, it just doesn't worth the money. Even at write and Read speeds the difference remains the same. Let's have a look at our next benchmark and see if this changes a little.
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.
Everest showed a bigger gap between 1333MHz and 2500MHz. Every MHz added results into higher read bandwidth. However, write performance remains the same at different frequencies. Finally, copy bandwidth increases with higher frequency, but also takes advantage of tighter latencies, thus giving advantage to the 2000MHz 7-9-7-21 configuration.
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.
Sandra showed up to 83% increase between 1333MHz and 2500MHz. Even at 2400MHz, there's a lot of increase against 1333 and 1600MHz. This test cares a lot of memory frequency. We just wished all applications showed similar results to take full advantage of a high-end memory kit like this one. In the next section, we run some real-life applications and compare performance between different frequencies.
Application Performance Results
While we normally choose Far Cry 2 or Crysis benchmark but this time I opted for Lost Planet 2. This game is being developed by Capcom and takes a pretty decent GPU load in DX11 mode, especially with test B. I run the test 3 times at 1920x1080 with 4xA/16xAF and everything maxed out. Let's have a look at the results obtained:
The difference is so minimal that it can be discarded for being inside a 1% tolerance. There's nothing new in these results. We all know memory MHz really don't help for gaming, even worse, not even CPU overclocking helps that much at GPU dependent games. That's why it's so ridiculous watching companies bragging about their high-end memory kits targeted at hardcore gamers. If there's something special between this and a 1600MHz kit, that should be the ability to reach higher frequencies and overclock like hell obtaining higher scores and world records, but for gaming, that's just not going to happen.
Adobe Photoshop CS5 is our new contender for the memory suite. This program can benefit from both CPU and memory, and additionally, it supports Nvidia CUDA and ATI Stream. While we normally used Retouch Artists Test, it was very difficult to get precise results since it's not a real benchmark application. However, KitGuru has developed a new script which maximizes the use of CPU and memory while working on a very big and complex image applying filters, layers and other processes making it really interesting at the end. This script also allows you to take note of the time every process takes to be done. Just add them all and you have a precise result. The test takes almost 9 minutes on a Core i7 overclocked setup, so expect it to take longer if you're using a weaker processor.
Again, there's no real gain between different memory frequencies. Except for the 1333MHz result, which is around 1.5% slower than the rest of the results, there's no visible advantage from going to 2500MHz or similar. Let's have a look at our final benchmark; WinRAR 3.9 x64 is a very used application for files compression. I bet many users use WinRAR or similar applications to zip/RAR their files and send them via e-mail or just to organize them in their storage drives. Have a look at the results obtained:
Finally! WinRAR shows some improvement going from 1333MHz to 1600MHz, and then again some extra performance when increasing memory frequency to 2000 and 2400MHz. There is a 28% increase between 1333MHz and 2400MHz. Unluckily, gong to 2500MHz but relaxing timings to 9-11-9 gives a worse result than running at 2400MHz with 8-11-8 timings.
Summarizing the application performance results; overclocking RAM isn´t a big deal nowadays. CPUs with bigger caches definitively ignore RAM frequency and timings. That trend doesn't seem to end any near in the future. Let's head to the overclocking results, where I show how well the PVV34G2500C9K kit scaled with different timings.
Patriot Viper II 'Sector 5' Overclocking Results
Here's the interesting section of this review. As I told at the beginning of this revision, this kit is really meant to be used by overclockers. Since the Viper II Sector 5 kit runs at 2500MHz with relatively decent timings (9-11-9), it just means it still has some room to increase frequencies or lower timings. Of course, this kit can't contain other than PSC ICs, which are very common for reaching high-frequencies with some extra voltage and a good IMC. To do the overclocking tests, I limited vDIMM to 1.75v as I think that's already a high voltage for Core i7 processors. I actually tested with 1.85v, but just to discover that my Core i7 860 didn't have a strong enough memory controller. Instead of aiming for higher frequencies and relaxed timings, I went for tight timings and lower frequencies. The results are exposed in the next chart:
Before you start asking why I did decided to increase tRCD instead of using similar value timings (8-8-8 and 9-9-9 for example), I must tell you the key for overclocking PSC ICs is hidden in tRCD. As you can see, when I decrease tRCD from 11 to 10, I lose 150-250MHz, which is a lot compared to 50-100MHz lost when I lower CAS latency and tRP. In fact, I already knew this before receiving this kit as this isn't the first kit with similar behavior I've tested in my life. I must say that even if you're not planning to use the PVV34G2500C9K at 2500MHz, it is still a very good option for 2400 and 2000MHz, as you tight your latencies from 9-11-9 to 8-11-8 and run a 200MHz BCLK. The result is more impressive at 2000MHz, since you can lower your timings down to 6-9-6-20 1T and still reach that frequency, giving you a very nice performance per clock in every benchmark (especially if it takes advantage of tight timings).
Finally, I'll ad some CPU-Z screenshots with the maximum frequencies I managed to reach at different timings:
Have you ever reach 2000MHz with a CAS latency value of 6? That's really impressive, I must say.
Using 7-10-7-24 1T timings, we passed the 2150MHz number.
Increasing timings to 8-11-8-27, we managed to get 2440MHz out of this kit. This could be even a better configuration than the stock XMP included in this kit.
Finally, at 9-11-9-27 timings, we reached 2550MHz, adding 50MHz extra from the stock values. More MHz could be possible with a stronger IMC and some extra voltage.
Viper II Sector 5 Final Thoughts
RAM memory is always taking big leaps to new heights where only overclockers and hardcore enthusiasts can benefit from. This is really the case of the Patriot Viper II Sector 5 Edition 4GB kit. As an enthusiast, I really feel great reviewing and testing such a kit like this, but in real life and usual applications, there's close to no advantage from taking this kit over a common 1333/1600MHz kit. The only real application that showed a noticeable difference was WinRAR, which I'm glad to use as many other users do, but for games or even CPU intensive applications, those $150-$200 compared to a 1600MHz kit, could be spent on a better GPU (a second GTX460 for example) or to buy 8GB of RAM, or simply to save them for another occasion if necessary.
From an overclockers point, this kit is just great to break records and test your overclocking abilities. It's very fun to tweak and set every memory parameter by yourself if you're patient enough to do so, and every extra MHz gained represents something to be proud of. Also, this kit could be used for benchmarks where every extra MHz or tight timings at relatively high frequencies (2000MHz for example) are needed to take the crown of your neighborhood. Aside from that, there's no real use for this kit. For comparison, a 2500MHz kit could cost up to $300, while a 2000MHz kit costs $200. To make things worse, a 1333-1600MHz kit could cost $100 only, and you could still get some extra performance by overclocking, while getting the same memory amount and quality. So, it's really up to you to decide if you really want to spend $300+ on a high-end kit just to test your OC abilities and break some benchmark numbers, or you still prefer to spend that into a more "useful" component. Considering you can buy 12GB of 1333MHz memory with the cost of this 2500MHz 4GB kit, it really puts things in a consumer-perspective. Let's head to the Patriot PVV34G2500C9K conclusions in the next paragraph...
Patriot PVV34G2500C9K Conclusion
Patriot Viper II Sector 5performance is top notch. You receive a high-binned 2500MHz kit, and you're still able to squeeze some extra MHz if your CPU's IMC is able to support it. All depends on your CPU to overclock this kit. If you've got a weak IMC, you could even have problems reaching 2400MHz, so please be advised that you need a decent Core i7 processor for the task. With lower latencies, this kit is still able to reach very decent frequencies, being 2400MHz 8-11-8 and 2000MHz 6-9-6 some of the best setups we have seen between memory kits.
The appearance is quite good, but as always, it has its downsides. If you love black/red colors, you'll love this kit. Pair it with ASUS ROG motherboard and an ATI Radeon card and your PC will be the envy of your friends. However, put this kit with a blue/green PCB motherboard or an Nvidia GPU and it will lose all its attractive. Also, some users will have problems with the heat-spreaders and CPU heatsinks, so be warned, especially if you're using a 120mm heatsink with 2 fans. In my case, the Noctua NH-D14 allows you to move the fan upward or downwards depending on your needs, but many other CPU heatsink don't give you a choice.
Construction felt very solid until I installed the second DIMM and half of a heat-spreader fell off. I really think this was a little mistake, and many users won't suffer from problems like this, but still, I'd hate to buy a $300 kit just to see how parts fall off in the installation process.
Benchmark Reviews has proven RAM functionality doesn't necessarily increases with higher frequency. According to our tests, this kit shows some extra benefit against lower frequency setups, but it's not really worth the price's difference, since buying more RAM or adding better PC components gives you higher performance than when spending $300 on a RAM kit. Even if this kit overclocks like hell, some users just can't justify buying a 4GB kit instead of buying 8 or 12GB.
For $309.99 at Newegg, you get a high-binned 2500MHz kit. Sure, it's very expensive, but you get what you pay for. Unluckily, you can get 4GB kits for $100 or even 200MHz kits for $200. So, unless you're trying to overclock the hell out of your PC, or you have tons of money to buy computer hardware, you'll be just fine with a less expensive kit. Maybe the Sector 5 200MHz kit or something similar will fit better to your needs. Pros:
+ Specially binned PSC ICs
+ Patriot Limited Lifetime Warranty
+ Interesting design and colors will appeal a bunch of users
+ 2500MHz kit is one of the fastest available at the moment
+ Superior overclocking capabilities
+ Works great with Command Rate: 1
Cons:
- Might not work with many CPUs and motherboards
- Excessive price even for enthusiasts
- Requires "high" voltage exceeding Intel's maximum specifications
- The heat spreader's design could interfere with a big CPU heatsink
- Needs overclocked BCLK to achieve 2500MHz
Ratings:
Performance: 9.50
Appearance: 9.50
Construction: 9.00
Functionality: 8.00
Value: 8.00
Final Score: 8.8 out of 10.
Recommended: Benchmark Reviews Seal of Approval.
