Intel Sandy Bridge Core i5-2500K & Core i7-2600K Processors Review

[MAC]

With 2011 upon us, many in the tech industry are gearing up for what looks like one of the most interesting years in terms of new product launches and emerging technologies finally coming into their own. Stereoscopic 3D content will continue to be shoved down our throats, AMD’s Fusion architecture should finally be revealed, tablets are quickly becoming the must-have devices and HDTVs could go beyond the mythical 1080P “barrier”. But before all of that happens, Intel has decided to upstage everything else by officially introducing their brand new Sandy Bridge micro-architecture just a few short days after the clock struck midnight, ushering in 2011.

Sandy Bridge represents a continuation of Intel’s legendary tick / tock release pattern whereby a refined design – in this case Nehalem and its Westmere derivatives – are gradually phased out in favor of a new architecture. This will be the first step towards replacing the entire Nehalem lineup from Bloomfield to Lynnfield to Clarkdale but for the time being Sandy Bridge processors will play second fiddle to higher-end i7 Bloomfields. All of the current Clarkdale processors will move towards EOL status beginning immediately and the new processors will also take a large chunk out of Intel’s Lynnfield lineup as well.

As you may have guessed by now, the Sandy Bridge launch is a massive undertaking which bridges the mobile and desktop markets at the same time. For the purposes of this review, we will only be taking a look at the desktop parts i7-2600K and i5-2500K CPUs but even these two products only scrape the tip of a massive iceberg. A total of eight desktop CPUs will be launching today which range in price from $117 to $320 according to Intel but we expect street prices to be straddling the $125 to $350 brackets. All of these processors will also feature an onboard GPU-much like Clarkdale did.

Between the release of the last generation Clarkdale CPUs and Sandy Bridge, the market has continued to move towards high definition content and highly demanding online content. Believe it or not, the PC gaming market is also predicted to expand exponentially within the next three years. These trends along with AMD’s upcoming APUs have forced Intel to rethink their approach when it comes to graphics processing. Sandy Bridge is the first product generation to incorporate the byproducts of these realizations in the form of a thoroughly revised graphics controller layout dubbed the 2000 and 3000-series.

Intel’s timing of this launch is seemingly perfect since for the time being, the competition has absolutely nothing that can compete with this new Sandy Bridge product family. Desktop Fusion APUs are still a long ways off and the entry level i3-2100 ($117) and i3-2120 ($140) and quite obviously priced to cut the heart out of AMD’s current lineup. However, will the hoped-for performance line up with the premium being charged for some of these new chips? Let’s find out.

 

[MAC]

Sandy Bridge: Intel Core i5-2500K & Core i7-2600K

Sandy Bridge: Intel Core i5-2500K & Core i7-2600K

Sandy Bridge/Gulftown/Clarkdale/Lynnfield/Bloomfield - Click on image to enlarge​

Over the last 2 years we have witnessed the release of 4 new processor families based on the Nehalem microarchitecture, and we would be hard pressed to hide the fact that we've enjoyed the journey. The Core i5 and Core i7 processors were substantial improvements over the venerable Core 2 processors that had been the stalwart of most enthusiast's systems during the second-half of the decade. As good as Nehalem-based processors are though, Intel's Tick-Tock strategy never ceases to push CPU technology forward, and that has brought us to today's launch of Intel's brand proressor microarchitecture, Sandy Bridge.

4 cores/8 threads - 8MB L3 Cache - 995 Million Transistors - 216mm2​

While it would be tempting to describe this new microarchitecture as a hybrid between Lynnfield (advanced Turbo features/power management) and Clarkdale (integrated GPU/32nm process), Sandy Bridge represents an equal if not greater change than we saw going from Core 2 to Nehalem. The cores have been totally redesigned, the front-end branch predictors have been optimized, all the parts of the chip are connected via a high bandwidth 256-bit ringbus, the L3 cache now runs at full speed and its latency has been lowered, the memory controller frequency has increased dramatically, and there's even a new extension in the form of AVX, Advanced Vector Extension. This all combines to create a microarchitecture that is on average 10-20% faster than what is currently on the market, and that can peak at over 40% faster in certain workloads.

We obviously can't talk Sandy Bridge without mention the integrated GPU. This new IGP is an evolution of the one found in Clarkdale chips, it does feature reworked cores, but it benefits most greatly from the 32nm manufacturing process and the new ringbus interface. The process shrink has permitted Intel to increase the IGPs clock speed dramatically (up to 1.35Ghz!) while continuing to have a minimal impact on overall power consumption and heat output. If you want to know more, click here to jump to our Intel HD Graphics 2000/3000 section.

Now before we start talking about the individual models in the new Sandy Bridge family, let's reiterate the Core i3/5/7 naming scheme that Intel has chosen to continue with this new release. Simply put, Core i7 models are eight-thread processors which feature both Hyper-Threading (HT) and Turbo Boost technology. The Core i5 models are four-thread processors with Turbo Boost. The Core i5 series will continue to be particularly confusing to consumers though, since it is comprised of both 4-core/4-thread and 2-core/4-thread processors. Lastly, there is Core i3 series models, which are 2-core/4-thread processors but without Turbo Boost.

Enough talk for now, here is how the new Sandy Bridge family breaks down:

This is what we are going to call the 'mainstream' models. On one end, the Core i5-2400 effectively slots into the spot previously held by the Core i5-750, while on the other end, the Core i7-2600K can be seen as replacement for the Core i7-875K. As you can see, the Turbo Boost implementation is nowhere near as aggressive as with Lynnfield, but the default clock frequencies are quite a bit higher.

Here we have the more affordable, budget-friendly Sandy Bridge variants. While somewhat attractively priced, these chips are going to face enormous competition from AMD processors like the upcoming Phenom II X4 840.

Last, but certainly not least, are these low wattage beauties. These are the models that are likely going to prove very popular with the HTPC crowd, or just those that keep a watchful eye on their electricity bill.

A new product lineup always mean new packaging, and Intel have indeed unveiled a new look for their Sandy Bridge processors:

Core i5-2500K on the left, Core i7-2600K on the right - Click on image to enlarge​

Here is Sandy Bridge in the flesh. As you can see, unless you are truly astute, these new LGA1155 Sandy Bridge processors are almost indistinguishable from the previous LGA1156 Clarkdale and Lynnfield models.

Based on the digits on the integrated heatspreaders, we can determine that these two chips were made in the 35th and 37th week of 2010. That might seem kind of old considering that we are now in 2011, but Intel is known for stockpiling millions of chips before a launch.

LGA1155 vs. LGA1156 vs. LGA1156 - Click on image to enlarge​

Once again, even from the back, Sandy Bridge is almost indistinguishable from Lynnfield and Clarkdale, but Intel have moved the securing notches to ensure that no one can have any installation mix-ups/disasters.

Don't expect to ever see stock clocks with Turbo Boost enabled - Click on image to enlarge​

With Sandy Bridge, Intel has integrated the clock generator off the motherboard and onto the processor itself. The most visible aspect of this change is that the 'bus speed' has been reduced from 133Mhz to 100Mhz. However, the tangible effect is the BCLK overclocking issue that have been reported over the last few months. There is a very little BCLK headroom on Sandy Bridge processors, roughly 5% to 10% maximum. This is because so many of the CPU's different parts are deriving their operating frequencies from this base clock, and since some are very sensitive to frequency changes, they can get out of whack very quickly.

Although Sandy Bridge is manufactured on the smaller 32nm process, operating voltages haven't decreased much (if at all) when compared to the 45nm Lynnfield chips. This really isn't a bad thing though, since Intel is using that voltage to produce SB processors with higher clock speeds.

 

[MAC]

In-Depth Look at Intel's HD Graphics 2000 & 3000 IGPs

An In-Depth Look at Intel's HD Graphics 2000 & 3000 IGPs

Much like Clarkdale processors they are meant to replace, Sandy Bridge chips will feature dedicated GPU cores integrated onto the CPU die. This Processor Graphics Controller (or PGC as Intel calls it) improves upon the performance of the previous generation while featuring additional instruction sets in order to better integrate into current and emerging digital media trends.

Being integrated onto the CPU package nets the graphics controller certain resources which were lacking from past iterations of onboard graphics. Along with the memory interface, the 3-6MB (depending on the processor) of on-die L3 cache is shared between the controller and the CPU which should facilitate draw call communication between the two and allow better texture performance than previous generations.

A simplified block diagram shows us a layout that is similar to modern discrete GPUs but also one which is more focused upon HD media content rather than pushing a massive amount of texture horsepower. At the heat of this architecture lies six to twelve Unified Execution Units which are broadly comparable to the Shader Processors in AMD and NVIDIA cards and are used primarily for processing 3D graphics.

All of these “cores” are separately programmable and perform 128-bit wide executions through every clock cycle with a 4KB register file per thread. The rest of the rendering pipeline from geometry shading, vertex processing, rasterization and so on remains identical to today’s stand-alone graphics cores. There is however a single dedicated texture unit as well as a new mid-level instruction cache layout which is supposed to help with overall performance across a variety of applications.

Intel’s HD Graphics on Sandy Bridge now feature dynamic frequency adjustments in order to automatically increase the clock speeds of the graphics controller when higher loads are detected. Much like the Turbo Boost technology on the CPU itself, this acts as a way to conserve power when high speeds aren’t needed and yet allows for on-the-call performance in demanding situations.

The media processing capabilities of this new Gen 6.0 architecture have received a thorough work-over with a massive amount of effort going into true hardware decoding and encoding. There is now a dedicated multi-format hardware assisted encode / decode pipeline which also handles a fair amount of preprocessing which can be accelerated by the onboard controller rather than relying on poorly performing software processing routines.

Additional improvements have been made through the use of a dedicated media engine which now includes support for stereo 3D through multiple video channel outputs and has the ability to simultaneously decode two HD video streams. We will be going into additional media features and the overall performance on this new architecture in a dedicated article.

Sandy Bridge’s graphics controllers are now broken down into two separate categories: the 2000 and 3000 series. The main differentiating factor between these two HD graphics engines is the number of Execution Units each offers: the 2000-series uses six while the 3000-series uses 12. Both offer a number of improvements over the previous generation but in many ways they are still a far cry from what is available in the DX11 discrete market. Nonetheless, the addition of dedicated media processing, higher clock speeds, DX10.1 / Shader Model 4.1 support, OpenGL 3.0 compatibility and overall higher efficiency Execution Units should give a noticeable improvement over Clarkdale’s somewhat lackluster performance. Unfortunately, DX11 support is still not included here.

For the time being, the higher end K-series Sandy Bridge chips (2600K and 2500K) will feature the Intel 3000 graphics accelerator while all others will make do with the 2000 controller. Considering the K-series won’t have all their overclocking functions enabled on motherboards which allow the integrated graphics processor to be used, it’s odd to see the 3000 being used exclusively on these chips. The i7 2600 does make use of a maximum graphics core speed increase to 1.35Ghz which should somewhat alleviate the performance hit incurred by its “missing” six EUs.

The lower-end i5 and i3 processors show us much of the same with only the 2000 graphics processor and its six EUs being used for all of the products.

While a maximum of twelve execution units may not sound like much when compared to today’s entry level cards, Intel claims the performance from this graphics controller approaches and in some cases surpasses most low end discrete products. Even if this claim doesn’t quite pan out, these new controllers are able to leverage a revised architecture to thoroughly outpace the i5 661; the only processor to be available with an IGP boasting a 900Mhz clock speed. It seems like Intel is making moves in the right direction here but we also have to remember that entry level cards from AMD and NVIDIA are getting better and AMD’s own competing Fusion products are right around the corner as well.

 

[MAC]

Intel H67 Express & P67 Express Chipsets

Intel H67 Express & P67 Express Chipsets

Back when Lynnfield was launched, we were introduced to an all new chipset layout from Intel where many of the usual Northbridge functions were consolidated onto the CPU die. This P55 Express “Ibex Peak” chipset allowed for a more integrated layout and also moved away from the traditional two-chip layout of a Northbridge and Southbridge towards a single chipset design.

Sandy Bridge motherboards will use the “Cougar Point” 6-series chipsets of which the P67 boards will target the upper end of the spectrum while others like H67 and Q67 will be destined for slightly lower-end products. Cougar Point won’t change the ballgame like Ibex Peak did last year. Rather, this new chipset design uses the same building blocks as past PCH-equipped chipsets yet features expanded capabilities in several key areas.

The high-end P67 Express chipset layout is nearly identical to the P55 Express but there are several key changes. Sandy Bridge processors themselves will feature an integrated PCI-E controller which supplies 16 PCI-E 2.0 lanes. When installed on a P67-equipped board, they can be used in one of two ways: either one slot operating at x16 or dual slots running at x8. This means both Crossfire and SLI are supported but not at their full theoretical bandwidth.

Sandy Bridge processors will also retain the same basic memory layout as the last generation with two memory channels (versus the three memory channels of the high-end Bloomfield series) with the capability of running DDR3-1333 modules in each of the two channels. A total of 32GB can installed when using dual sided 8GB DIMMs.

The link between the CPU and the P67 Platform Controller Hub (PCH) is still done via the DMI interface but in this case, the specifications of this interface have been upgraded from past generations. It now features four lanes in each direction which can operate at speeds of up to 2 GB/s. This results in 4 GB/s of aggregate bandwidth if both upstream and downstream lanes are used to their theoretical maximum.

The Cougar Point chipset (in this case P67 Express) acts as the control hub for all of the peripheral and storage connectors on Sandy Bridge motherboards. Not much has changed here since it still features eight PCI-E lanes, an Intel HD Audio module along with Intel’s Extreme Tuning support. Intel Extreme Tuning will only be available on the P67 boards and adds Windows-based overclocking if you have an unlocked K-series processor as well as some basic system monitoring tools.

For the most part, external storage capabilities of Sandy Bridge-based boards haven’t changed all that much from the previous generation since up to fourteen USB 2.0 ports are available but native support for USB 3.0 is missing. The real difference lies in the Serial ATA interface as Intel has decided to add native support for the new SATA 6 Gbps standard. Of the six support SATA ports, two can be converted to SATA 6. Naturally, motherboard manufacturers still have the option of adding third party controllers for both USB 3.0 support and additional SATA 6 Gbps ports.

The reference specification for H67-based motherboards will retain many of the P67’s capabilities but since they are targeted at a more price-conscious market, several features have been removed. First and foremost among these is the ability for dual graphics card support via the redistribution of the CPU’s sixteen PCI-E lanes. This means only one GPU is “officially” supported on a single x16 slot. However, companies have been known to add the option for “unofficial” Crossfire and SLI setups on these lower-end chipsets as we saw with ASUS’ P7H57D-V EVO which still split the CPU’s lanes between two slots. In addition, the H67 boards will not support Intel’s Extreme Tuner.

Unlike P67 boards, H67 products will add the option to use the CPU’s on-board Intel HD graphics module along with display outputs located on the motherboard itself. Since the H67 PCH also controls display outputs, Intel has provided a Flexible Display Interface (FDI) in order to transport display information between the GPU and the chipset. This allows for support for HDMI 1.4, Displayport and other display outputs as well.

One other advantage P67 Express-based boards will have is their ability to support the full range of overclocking and unlocking capabilities for Sandy Bridge processors. At this point, lower end products simply lack the support for processor tweaking and tuning but they will allow some extension to on-board processor graphics performance.

While board partners are once again sure to add in some basic overclocking options into the BIOSes of certain lower-end products, Intel seems to be pushing users to choose a K-series chip along with a P67 board in order to push their processors to the limit.

 

[MAC]

Test Setups & Methodology

Test Setups & Methodology

For this review, we have prepared four different test setups, representing all the popular platforms at the moment, as well as most of the best-selling processors. As much as possible, the four test setups feature identical components, memory timings, drivers, etc. Aside from manually selecting memory frequencies and timings, every option in the BIOS was at its default setting.

Intel Core i5/i7 LGA1155 Test Setup​

​

Although Windows Vista SP1 was our principal OS for the majority of benchmarks, we did use Windows 7 (with all the latest updates) when benchmarking AIDA64 and HDxPRT 2011.

AMD Phenom II AM3 Test Setup​

​

Intel Core i3/i5/i7 LGA1156 Test Setup​

​

Intel Core i7 LGA1366 Test Setup​

​

For all of the benchmarks, appropriate lengths are taken to ensure an equal comparison through methodical setup, installation, and testing. The following outlines our testing methodology:

A) Windows is installed using a full format.

B) Chipset drivers and accessory hardware drivers (audio, network, GPU) are installed followed by a defragment and a reboot.

C)To ensure consistent results, a few tweaks were applied to Windows Vista and the NVIDIA control panel:

• Sidebar – Disabled
• UAC – Disabled
• System Protection/Restore – Disabled
• Problem & Error Reporting – Disabled
• Remote Desktop/Assistance - Disabled
• Windows Security Center Alerts – Disabled
• Windows Defender – Disabled
• Windows Search – Disabled
• Indexing – Disabled
• Screensaver – Disabled
• Power Plan - High Performance
• NVIDIA PhysX – Disabled
• V-Sync – Off

D) Programs and games are then installed & updated followed by another defragment.

E) Windows updates are then completed installing all available updates followed by a defragment.

F) Benchmarks are each run three times after a clean reboot for every iteration of the benchmark unless otherwise stated, the results are then averaged. If they were any clearly anomalous results, the 3-loop run was repeated. If they remained, we mentioned it in the individual benchmark write-up.

Here is a full list of the applications that we utilized in our benchmarking suite:

• AIDA64 Extreme Edition v1.50.1200 (Windows 7)
• ScienceMark 2.0 32-bit
• MaxxMEM2 Preview
• wPrime Benchmark v2.03
• HyperPI 0.99b
• PCMark Vantage Advanced 64-bit Edition (1.0.2.0)
• Cinebench R10 64-bit
• Cinebench R11.5.2.9 64-bit
• WinRAR 3.94 x64
• Photoshop CS4 64-bit
• Lame Front-End 1.0
• X264 Benchmark HD (2nd pass)
• 7-Zip 9.20 x64
• POV-Ray v3.7 beta 40
• Deep Fritz 12
• HDxPRT 2011 v1.0 (Windows 7)
• 3DMark06 v1.2.0
• 3DMark Vantage v1.0.2
• Crysis v1.21
• Far Cry 2 1.02
• Left 4 Dead version 1.0.2.3
• Valve Particle Simulation Benchmark
• Word in Conflict v1.0.0.0
• Resident Evil 5 1.0.0.129
• X3: Terran Conflict 1.2.0.0

That is about all you need to know methodology wise, so let's get to the good stuff!

 

[MAC]

Feature Test: Hyper-Threading (HT)

Feature Test: Hyper-Threading (HT)

As implemented on Sandy Bridge, Hyper-Threading (HT) is a feature that allows a processor with four physical cores to simultaneously process up to 8 threads. A core usually processes the pieces of the different threads one after another, however an HT-enabled core can process two threads in a simultaneous manner. While Hyper-Threading did not perform particularly well on the Pentium 4's Netburst microarchitecture, Nehalem and now Sandy Bridge were designed to remove many of the processing bottlenecks that had plagued previous implementations. Depending on the workload, and how effectively multi-threaded an application is, the performance increases could be 20% or higher.

This is a technology that was very successfully brought back with the Nehalem microarchitecture, so it should really excel on Sandy Bridge processors. To demonstrate this, we have benchmarked a small selection of multi-threaded applications with HT enabled and disabled.

As you can see, there are some very tangible gains to be had from Hyper-Threading, but the application has to be able to support the additional threads. In highly multi-threaded applications, HT can make a significant difference, speeding up specific workloads by 10% to 40%. This is roughly about the same level of improvement that we saw on Nehalem-based chips. However, most contemporary games simply aren't designed to take advantage of more than four threads, much less eight. This obviously negates any possible HT-related performance gains. In some exceptionally rare cases, Hyper-Threading can have a slightly negative impact, but only by about 1-2%.

 

[MAC]

Feature Test: Turbo Boost 2.0

Feature Test: Turbo Boost 2.0

3.4Ghz Core i7-2600K Turbo'ing up to 3.79Ghz​

For those of you who aren't familiar with it, let's recap what Turbo Boost is and what it does. Turbo Mode is a performance feature that automatically unlocks additional speed bins (multipliers) and allows the processor to self-overclock based on thermal conditions and workload. For example, if the Power Control Unit (PCU) senses that only one core is active and the other three are in an idle state, it will use the unused power and thermal headroom to overclock that single active core to ensure superior single-threaded performance. Conversely, if you are running a multi-threaded application, the PCU will measure the thermal headroom and if the processor is running cool enough it will overclock all six cores. On Sandy Bridge processors, Turbo can provide a 400Mhz speed boost in single-threaded workloads, 300Mhz in dual-threaded workloads, 200Mhz in triple-threaded workloads, and 100Mhz in applications that utilize four threads or more.

Let's check out the performance gains that Turbo Boost can provide on top-end Core i7-2600K model. Since single-threaded applications are diminishing in importance and prevalence, we selected some benchmarks that were admittedly skewed towards 2 threads or more.

The gains from Turbo Boost are perhaps not as great as with past processors like Lynnfield, but frankly you are getting a 10-20% clock-per-clock performance improvement anyways just due to the microarchitecture improvements. Furthermore, I think just about everyone will prefer having a higher default clock speed instead of additional Turbo Boost bins.

 

[MAC]

Synthetic Benchmarks: AIDA64 / MaxxMEM²

Synthetic Benchmarks: AIDA64 / MaxxMEM

AIDA64 Extreme Edition 1.50 - CPU & FPU Benchmarks

AIDA64 Extreme Edition 1.50 - Cache Benchmark

AIDA64 Extreme Edition 1.50 - Memory Benchmarks

MaxxMEM² - Memory Benchmarks

 

[MAC]

Synthetic Benchmarks: SuperPI 32M / wPRIME 1024M

Synthetic Benchmarks: SuperPI 32M / wPRIME 1024M

SuperPi Mod v1.5

When running the SuperPI 32MB benchmark, we are calculating Pi to 32 million digits and timing the process. Obviously more CPU power helps in this intense calculation, but the memory sub-system also plays an important role, as does the operating system. We are running one instance of SuperPi via the HyperPi 0.99b interface. This is therefore a single-thread workload.

wPRIME 2.03

wPrime is a leading multithreaded benchmark for x86 processors that tests your processor performance by calculating square roots with a recursive call of Newton's method for estimating functions, with f(x)=x2-k, where k is the number we're sqrting, until Sgn(f(x)/f'(x)) does not equal that of the previous iteration, starting with an estimation of k/2. It then uses an iterative calling of the estimation method a set amount of times to increase the accuracy of the results. It then confirms that n(k)2=k to ensure the calculation was correct. It repeats this for all numbers from 1 to the requested maximum. This is a highly multi-threaded workload.

 

[MAC]

System Benchmarks: Cinebench R10 / Cinebench R11.5

System Benchmarks: Cinebench R10 / Cinebench R11.5

Cinebench R10

Cinebench R10 64-bit
Test1: Single CPU Image Render
Test2: Multi CPU Image Render
Comparison: Generated Score

Developed by MAXON, creators of Cinema 4D, Cinebench 10 is designed using the popular Cinema software and created to compare system performance in 3D Animation and Photo applications. There are two parts to the test; the first stresses only the primary CPU or Core, the second, makes use of up to 16 CPUs/Cores. Both are done rendering a realistic photo while utilizing various CPU-intensive features such as reflection, ambient occlusion, area lights and procedural shaders

Cinebench R11.5

Cinebench R11.5 64-bit
Test1: CPU Image Render
Comparison: Generated Score

The latest benchmark from MAXON, Cinebench R11.5 makes use of all your system's processing power to render a photorealistic 3D scene using various different algorithms to stress all available processor cores. The test scene contains approximately 2,000 objects containing more than 300,000 total polygons and uses sharp and blurred reflections, area lights and shadows, procedural shaders, antialiasing, and much more. This particular benchmarking can measure systems with up to 64 processor threads. The result is given in points (pts). The higher the number, the faster your processor.