Techgage logo

Intel Core i7-4960X Extreme Edition Review

Date: September 25, 2013
Author(s): Rob Williams

Intel’s latest processor series has arrived, and we’re looking to find out if it becomes the company’s greatest. Compared to Intel’s latest mainstream part, Haswell, IV-E avails a quad-channel memory controller, a far more robust PCIe configuration, and the only place to get six-core parts. Are there other perks to be found? Let’s find out.



Introduction

This coming March, Techgage will turn 9-years-old. In all the time the site has been around, we’ve taken a look at a countless number of processors from both AMD and Intel, and generally speaking, the process for reviewing a new one – especially one with a new architecture – has largely been the same: A) Benchmark; B) Stare in awe at the improved performance; C) Convert that excitement into words; D) Profit.

I’d love nothing more than to be able to say that the trend will continue with Intel’s just-released Ivy Bridge-E based processors, but the fact of the matter is, even Intel itself knew that it was going to be a hard sell. It had to have.

When Intel released its Sandy Bridge-E processors two years ago, I was left impressed overall with what it offered over the previous generation, but I mentioned in the conclusion that many “were hoping that Sandy Bridge-E would launch with eight cores” – and of course, it didn’t. It seemed to be a sure thing, then, that two years later, its successor would ship with those coveted eight cores. But not so.

Intel 4th Gen Core Processor Box

In its press slides, Intel states that compared to Sandy Bridge-E, Ivy Bridge-E should be about 4% faster for “everyday computing”, 7% faster for 3D gaming, 5% faster for financial analysis and 10% faster for 3D modeling. This type of performance increase is what we’d expect to see from a speed-boost product – not a new architecture being released two years after the last.

The problem is further highlighted by a comparison of the Core i7-4960X that can be seen below, involving the top-end Haswell part released this past summer, the Core i7-4770K. For the first time that I can recall ever seeing from a launch of an enthusiast-class CPU, it’s being noted that “everyday computing” is 18% lower. This is thanks to Haswell’s increased instructions-per-clock and single-threaded performance, but is that what you’d expect to see from a $1,000 processor versus a $339 one?

Intel Ivy Bridge-E Performance Expectations

Despite that bizarre “lower” result listed in the slide above, the benefits a CPU like the 4960X do offer is that it can improve 3D gaming (though the result here is unrealistic outside of a benchmark), crunch numbers faster and of course, prove far more efficient at 3D work (and of course, video work, though it’s not mentioned).

The X79 platform offers a couple of important benefits over something like the Z87 as well, such as an improved number of PCI Express lanes, which is important for those running higher-end multi-GPU setups. We’re not dealing with a mere bandwidth limitation here, but rather an electrical one. On Z87, dual GPUs will give you an x8 / x8 PCIe configuration; on X79, it’d be x16 / x16. With additional chips on the motherboard, it could be x16 / x16 / x16 / x16 for a quad-GPU configuration, as is possible with our ASUS P9X79-E WS motherboard.

A rather significant perk for those who need it is increased memory bandwidth. At the same RAM speeds, going from a 4x DIMM configuration on Z87 to a 4x one on X79 could almost double the throughput. For most enthusiasts, this isn’t a benefit they can enjoy and experience, but for those opting for X79 for multimedia or workstation purposes, it certainly could.

Intel Core i7-4960X CPU-Z

Compared to Intel’s latest Z87 platform, X79 is looking long in the tooth, so it’s almost amazing that the company didn’t update things and push out an X89 chipset. Sure – the backwards compatibility is very-much appreciated (a rarity from Intel nowadays), but while Z87 natively supports 6x SATA 6Gbit/s and 6x USB 3.0, X79 supports 2x SATA 6Gbit/s and 0x USB 3.0 – yes, (Zero)x.

To get around this lacking functionality, motherboard vendors implement chipsets from places like ASMedia and Marvell to bump up the number of supported ports. With the P9X79-E WS we’re using, ASUS bumped up the number of ports to 6x SATA 6Gbit/s and 4x USB 3.0 – a great thing, but again, there are caveats. As far as performance goes on either these SATA or USB connectors, Intel has the performance crown locked-down. So by using ASMedia and Marvell, you’re actually using weaker-performing ports than those folks that are using a Z87 platform (a platform, that I must remind you, costs way less).

As a quick summary, the clear benefits Ivy Bridge-E avail vs. Haswell include a quad-channel memory controller, and a more robust PCIe configuration. The things that lack vs. Haswell include lower instruction-per-clock performance, fewer native SATA and USB ports, and of course, the fantastic power-related improvements that came with that architecture – but admittedly, those looking to build a six-core powerhouse might not be too concerned about that.

Well, this sure has been quite the review so far, hasn’t it? Let’s get a move on with a look at Intel’s current mainstream and higher current-gen series:

 CoresThreadsClockTurboCacheIGPTDP$/1,000
Core i7-4960X6123.6GHz4.0GHz15MBN/A130W$990
Core i7-4930X6123.4GHz3.9GHz12MBN/A130W$583
Core i7-4820K483.7GHz3.9GHz10MBN/A130W$323
Core i7-4770K483.5GHz3.9GHz8MBHD 460084W$339
Core i7-4770S483.1GHz3.9GHz8MBHD 460065W$303
Core i7-4770483.4GHz3.9GHz8MBHD 460084W$303
Core i5-4670K443.4GHz3.8GHz6MBHD 460084W$242
Core i5-4670S443.1GHz3.8GHz6MBHD 460065W$213
Core i5-4670443.4GHz3.8GHz6MBHD 460084W$213
Core i5-4570S442.9GHz3.6GHz6MBHD 460065W$192
Core i5-4570443.2GHz3.6GHz6MBHD 460084W$192
Core i5-4430S442.7GHz3.2GHz6MBHD 460065W$182
Core i5-4430443.0GHz3.2GHz6MBHD 460084W$182
All models are built on a 22nm process, and utilize 3D tri-gate transistors. None provide cookies.

Intel is offering three different Ivy Bridge-E models to choose from, with all three comfortably being considered “high-end”. The Core i7-4820K is best comparable to the Haswell-based Core i7-4770K, and surprisingly, despite costing $16 less, it offers a modest clock boost and a 25% increase to cache.

On the six-core front, the Core i7-4960X we’re looking at here is clocked at 100MHz higher than last gen’s Core i7-3970X, and commands a $1,000 price tag. For those wanting to save their wallet a bit of stress, the i7-4930X is clocked at 200MHz less, and with 80% of the cache, and can be had for $583.

It’s worth noting that the quad-core Sandy Bridge-E model was not an unlocked part. The quad-core model with Ivy Bridge-E is. Kudos to Intel for that change. This means that every single one of the Sandy Bridge-E models available are ready to be poked and prodded for overclocking with no limitations whatsoever (well, except for the top-end limit of the chip, of course).

As should be clear by this point, the Core i7-4960X is a bit of an odd chip, but now it’s time to see how it actually performs, both compared to the last-gen Core i7-3970X – which it replaces – and also the Haswell i7-4770K.

Test Systems & Benchmarking Methodologies

At Techgage, we strive to make sure our results are as accurate as possible. Our testing is rigorous and time-consuming, but we feel the effort is worth it. In an attempt to leave no question unanswered, this page contains not only our testbed specifications, but also a detailed look at how we conduct our testing.

If there is a bit of information that we’ve omitted, or you wish to offer thoughts or suggest changes, please feel free to shoot us an e-mail or post in our forums.

Test Systems

The tables below list all of the hardware we use in our current CPU-testing machines.

 Intel X79 Test Machine
ProcessorIntel Core i7-4960X (Six-core, 3.6GHz, 4.0GHz Turbo)
Intel Core i7-3970X (Six-core, 3.5GHz, 4.0GHz Turbo)
MotherboardASUS P9X79-E WS
MemoryKingston Beast 4x8GB DDR3-1333 9-9-9
GraphicsNVIDIA GeForce GTX 660
AudioOnboard
StorageKingston HyperX 240GB SSD
Power SupplyCooler Master Silent Pro Hybrid 1300W
ChassisCooler Master Storm Trooper Full-Tower
CoolingThermaltake WATER3.0 EXTREME Liquid Cooler
Et ceteraWindows 7 Ultimate 64-bit
 Intel Z87 Test Machine
ProcessorIntel Core i7-4770K (Four-core, 3.5GHz, 3.9GHz Turbo)
MotherboardASUS Z87-EXPERT
MemoryKingston Beast 2x8GB DDR3-2133 11-12-11
GraphicsNVIDIA GeForce GTX 660
AudioOnboard
StorageKingston HyperX 240GB SSD
Power SupplyCorsair HX850
ChassisCorsair Obsidian 700D Full-Tower
CoolingNoctua NH-U14S Air Cooler
Et ceteraWindows 7 Ultimate 64-bit
 Intel Z77 Test Machine
ProcessorIntel Core i7-3770K (Four-core, 3.5GHz, 3.9GHz Turbo)
MotherboardGIGABYTE Z77X-UP4 TH
MemoryKingston Beast 2x8GB DDR3-2133 11-12-11
GraphicsNVIDIA GeForce GTX 660
AudioOnboard
StorageKingston HyperX 240GB SSD
Power SupplyCorsair HX850
ChassisCorsair Obsidian 700D Full-Tower
CoolingNoctua NH-U14S Air Cooler
Et ceteraWindows 7 Ultimate 64-bit

When preparing our testbeds for any type of performance testing, we follow these guidelines:

General Guidelines

To aid in the goal of achieving accurate and repeatable results, we stop certain services in Windows 7 from starting up at boot. This is due to the fact that these services have the tendency to start up in the background without notice, potentially causing inaccurate test results. For example, disabling “Windows Search” turns off the OS’ indexing which can at random times utilize the hard drive and memory.

The most important services we disable are:

To ease the tedium of setting up an OS for a round of benchmarking, we rely on Acronis True Image to restore an install that we previously setup. These images include most of our benchmarks, a minimal number of drivers (LAN, graphics), an up-to-date OS and all of our above-mentioned tweaks. We create a total of two OS images; one for AMD, and one for Intel.

Real-World & Synthetic Benchmarks

To help us deliver a well-rounded set of test results for each processor we evaluate, we use a variety of real-world applications and synthetic benchmarks.

Techgage Processor Test Suite Desktop

Our current test suite consists of:

Real-world

Synthetic

Most tests are run twice over with the results averaged. If there is an unnatural variance between the first two runs, then we continue to run the test until we receive a result we believe to be accurate.

Workstation: Autodesk 3ds Max & Maya

If there’s design work that needs to be done, then Autodesk is sure to have the right tool. From 3D modeling to architectural design, Autodesk’s selection of highly-regarded tools is almost mind-numbing, and because both its 3ds Max and Maya applications have long been considered to be some of the best in their respective class, we opt to use them for our benchmarking here.

For the sake of all-around testing, we perform most of our benchmarking on this page with the help of SPEC’s SPECapc 3ds Max 2011 and SPECapc Maya 2012, although we also render an in-depth model/scene in the former. We’ll explain each benchmark as we go along.

SPECapc 3ds Max 2011

We kick off our testing with one of the most comprehensive benchmarks in our test suite: SPECapc 3ds Max 2011. The overarching goal of those responsible for producing SPEC’s benchmarks is to deliver as well-rounded a test suite as possible for a respective field, such as 3D rendering and modeling, to produce accurate results that those responsible for purchasing hardware can take advantage of.

Designed to utilize both the CPU and GPU, SPECapc 3ds Max 2011 comes in both free and professional flavors, with the latter being the version we use. It’s composed of 58 individual tests and takes about 6 hours to complete on a machine equipped with 12GB of RAM and a six-core Intel Core i7-990X.

SPECapc 3ds Max 2011

Intel Core i7-4770K - Autodesk 3ds Max 2011 - SPECapc

On the CPU front, the i7-4960X outperformed the i7-3970X by 3% – the other results are just about similar. While the Haswell-based i7-4770K fell behind in the CPU test – as expected – it performed much better in the GPU and large model composite test, versus the 4960X.

Autodesk 3ds Max 2011 – “Naomi” Render

For our second 3ds Max 2011 test, we render a scene commissioned from Bulgarian artist Nikola Bechev, entitled “Naomi: The Black Pearl”. The woman is composed of over 7,000 polys with the entire scene totaling just over 106,000 vertices. Three light sources are used, with the entire scene being enhanced with HDR and ray tracing techniques, and subsurface scattering applied to certain objects. The scene is rendered at 1800×3600 as a production release, with HQ detail levels being used all-around.

Autodesk 3ds Max 2011 - Naomi Render

Intel Core i7-4770K - Autodesk 3ds Max 2011

3D rendering tests like these were just made for CPUs with lots of cores, and that’s proven here. The quad-core i7-4770K fell behind the i7-4960X, but it must be said that a 12% gap between the two makes the Haswell architecture look all the more impressive.

SPECapc Maya 2012

Like its 3ds Max 2011 variant, SPECapc Maya 2012 is designed to stress various aspects of the tool, such as rendering with standard and HQ methods, working in wireframe mode and so forth across numerous models and one overarching scene titled “Toy Store”.

Autodesk Maya 2010 with SPECapc Maya 2009

Intel Core i7-4770K - Autodesk Maya 2012 - SPECapc

Haswell’s IPC performance was highlighted in our last test, but it really shines here, outperforming a CPU with 50% additional cores. Despite Intel’s claim of improved GPU performance on IV-E, it seems that it doesn’t affect a test like this one – the i7-4770K simply dominates.

Workstation: Cinebench & POV-Ray

Like Autodesk’s 3ds Max and Maya 3D tools, Maxon’s Cinema 4D is a popular cross-platform 3D design tool that’s used by new users and experts alike. Maxon is well-aware that its users are in need of some rather beefy PC hardware to help speed up rendering times, which is one of the reasons the company itself releases its own benchmark, Cinebench.

There are a couple of reasons we like to use Cinebench in our testing. For one, it’s freely available for anyone to download, unlike our Autodesk-based tests. Second, it has the capability to scale up to 64 threads, which means we’ll easily be able to rely on it until the next version hits. As a faster CPU can also help improve the GPU computational pipeline, we also like that it includes an OpenGL benchmark as well. The fact that the benchmark completes in mere minutes is another perk.

Cinebench R11.5

Intel Core i7-4770K - Cinebench

Here’s the kind of result Intel would love to see – it’s just a bit unfortunate it has to come from a synthetic benchmark. Still, in the best of circumstances, it appears that the 4960X can outperform the 4770K by at least 40%. Again, the 4770K proves to be the superior in the graphics department, and by a wide margin.

POV-Ray 3.70

The “Persistence of Vision Ray Tracer” is a multi-platform ray tracing tool that allows you to take your previously-created environments and models and apply a ray tracing algorithm based on a script you either created yourself or borrowed from others. The tool is free and has become a standard in the ray tracing community, with some of the ‘Hall of Fame’ results able to be found here.

For our testing, we run the built-in benchmark in both single-threaded and multi-threaded mode. The results are presented in “pixels-per-second” – a simple metric, but one that’s easy to understand.

POV-Ray 3.70

Intel Core i7-4770K - POV-Ray

Once again, Haswell’s superb single-threaded performance shines here, but Intel’s beefier 4960X takes the delicious cake in the multi-threaded test.

Multi-Media: Adobe Premiere Pro, HandBrake & CyberLink MediaExpresso

With our 3D modeling and rendering tests out-of-the-way, let’s dive right into another popular use for high-end machines: video editing and encoding. Scenarios here could include encoding a large movie into a mobile format, ripping a Blu-ray to your PC and encoding it for HTPC use, or encoding a family video you painstakingly edited.

Adobe’s Premiere Pro likely needs no introduction. It’s a tool used by the amateur and professional video content creator alike due to the extreme control it provides along with all of the important codecs, presets, filters and tweaking options. Premiere Pro can be used for any sort of video, be it real-life, animated, 3D or even game footage.

For our benchmarking, we encode a project that consists of 35GB worth of game footage from Payday: The Heist, which we encode to MPEG2 Blu-ray 1080p/30. The resulting video can be seen here.

To ensure an encode delivers the best possible video quality, we enable the “Maximum Quality Render”, which results in nearly 100% CPU utilization on up to 12 threads (we have not tested on CPUs that have more than 12 threads).

Adobe Premiere Pro CS5.5

Intel Core i7-4770K - Adobe Premiere Pro

Video encoding is a great scenario for multi-core CPUs, so it’s no surprise to see the 4960X perform so well here. It’s worth pointing out, though, that compared to the 3970X (released two full years ago), the gains are almost nonexistent.

HandBrake 0.9.9

Premiere Pro is meant to be used as a professional tool for editing and encoding, while HandBrake acts strictly as an encoder, able to take one video format and encode it to another according to your specifications. While there are many presets available from the get-go, you’re able to customize whatever’s available, or create your own. It’s a simple tool with complex capabilities.

Here, we have a project that makes use of a Blu-ray rip of Pixies: Live at the Paradise in Boston. With it, we encode the first 10 minutes of the concert to an archival-quality 720p MKV. The archival-quality encode is time-consuming, but it can take full advantage of a 12 threaded processor. For those interested, our H.264 options are:

ref=5:bframes=5:b-adapt=2:subq=9:deblock=-2,-2:trellis=2:merange=24

HandBrake 0.9.9

Intel Core i7-4770K - HandBrake

We saw almost no difference between the performance of the 3970X and 4960X in the last test, but here, the gap does widen up a little bit. Here, the i7-4770K took about 30% longer to perform the encode than the 4960X.

CyberLink MediaExpresso

CyberLink is a company that’s quick to jump on new technologies, and it’s for that reason that CPU vendors – namely Intel – like to promote its products for use in benchmarking. In MediaExpresso’s case, this converter app can take advantage not only of basic CPU accelerators, but QuickSync and also AMD Radeon and NVIDIA GeForce.

We test a total of five configurations here:

CyberLink MediaExpresso

Because it’s a little hard to follow in a graph alone, we also include the same results in a table. This allows us to show you the fastest run overall, and then look at how each CPU fared in individual tests without having to squint through the results.

Intel Core i7-4770K - CyberLink MediaExpresso

 CPU (BQ)CPU (FC)QS (BQ)QS (FC)GTX 660
Core i7-4770K2874198019886211432
Core i7-3770K298120438957711434
Core i7-4960X21491733N/AN/A1077
Core i7-3970X21981577N/AN/A1418
CPU = CPU only; QS = QuickSync; BQ = Better Quality; FC = Faster Conversion

Where straight CPU encoding is concerned, the i7-4960X wins hands-down… no contest. It even manages to blow every other CPU out of the water when it comes to encoding with the help of the GPU. The reason? That’s a good question. It could be that GPGPU throughput is part of that +30% GPU performance Intel was talking about.

Multi-Media: Adobe Lightroom & dBpoweramp

Photo manipulation benchmarks are more relevant than ever given the proliferation of high-end digital photography hardware. For this benchmark, we test the system’s handling of RAW photo data using Adobe Lightroom 4.4, an excellent RAW photo editor and organizer that’s easy to use and looks fantastic. You can check out our full review of the tool here.

For our testing, we take a total of 500 RAW files spread across 250 .NEFs captured with a Nikon D80 and 250 .CR2 captured across a Canon 40D and 5D Mark II. We export all of these files to a matte-sharpened quality 90 JPEG resized to a resolution of 1000×660 – similar to a lot of photos we use here on the website. The test is timed indirectly using a stopwatch as the program doesn’t record the duration itself.

Adobe Lightroom 4.4

Intel Core i7-4770K - Adobe Lightroom

Lightroom doesn’t utilize 4+ cores as much as we’d like, but it’s clear that six-cores are definitely better than four when running batch exports.

dBpoweramp R14

You own hundreds, thousands, or even tens of thousands of songs, all encoded to a pristine lossless format such as FLAC. Your mobile device on the other hand, supports either MP3 or AAC. What’s the solution? There are several, but the one I’ve relied on for almost ten years has been dBpoweramp. It’s both flexible and powerful, which happen to be two important factors for those who take their music seriously.

Recent versions of dBpoweramp have opened up the ability to encode more than one track at once, up to a limit of one-per-thread. With twelve-thread CPUs on the market, that ability can greatly improve overall times. For our testing here, we take 500 unique FLAC files that average about ~30MBs and encode them using the “high-quality” setting to 320Kbit/s MP3.

dBpoweramp R14

Intel Core i7-4770K - dBpoweramp

This is a perfect example of a scenario where six-core CPUs can be a boon. 50% cores in this case do in fact add up to about a 50% gain in performance. If all scenarios performed like this on six-cores, there’d never be a hesitation for a lot of people to pick them up.

Mathematics: Sandra Arithmetic, Multimedia, Cryptography & Financial

SiSoftware’s Sandra is a piece of software that needs no introduction. It’s been around as long as the Internet, and has long provided both diagnostic and benchmarking features to its users. The folks who develop Sandra take things very seriously, and are often the first ones to add support to the program long before consumers can even get their hands on the product.

As a synthetic tool, Sandra can give us the best possible look at the top-end performance from the hardware it can benchmark, which is the reason we use it to test much of our PC’s hardware. The fact that a free version exists so that you can also benchmark against our results is something we greatly appreciate.

The more threads a CPU has coupled with its frequency and architecture refinements, the faster it should be able to calculate complex math. We’re not talking about simple math that can be done on a calculator, but rather advanced calculation that is often used behind the scenes. Sandra’s Arithmetic test stresses the popular Dhrystone integer and Whetstone floating-point algorithms that have acted as a base for a countless number of benchmarks dating back as far back as the 70s.

SiSoftware Sandra 2011 SP5

Intel Core i7-4770K - Sandra Arithmetic

The six-cores are clear winners here, though it’s no surprise since Sandra tends to give us the “best case” performance from any CPU. Still, it’s good to see what these things are capable of when put to proper use.

Sandra Multi-Media

One of the best reasons for upgrading or building a new PC is to increase the performance for multi-media work, whether it be editing or encoding. As we saw earlier in our results, faster CPUs can save minutes or even hours of time. To test such capabilities here, Sandra renders the famous Mandelbrot set in a total of 255 iterations and in 32 colors.

This is a test that’s been around for close to forever, but it still scales extremely well with thread counts and can benefit from new media-centric instruction sets, including AVX.

Intel Core i7-4770K - Sandra Multimedia

The arithmetic test put the six-cores in a dominant lead, but multimedia changes things up a bit – neither scales quite as their +50% cores would suggest.

Sandra Cryptography

You might not be aware of it, but cryptography plays a major role in computing. With some algorithms proving more complex than the others, having a faster processor can dramatically improve performance – especially important on the server front. In Sandra’s benchmark, the mega-popular AES and SHA algorithms are computed, both with 256-bit key sizes.

Intel Core i7-4770K - Sandra Cryptography

With tests like these, you’d expect that adding in a couple of cores would put the six-core models far ahead of the quad-cores, but not so. But despite that, the six-cores still come out ahead overall. It might not be +50% performance, but it’s substantial gains nonetheless.

Financial Analysis

There’s little that can stress a CPU’s worth quite as much as number-crunching, and for that reason, we take full advantage of Sandra’s financial analysis benchmark. In the past, we ran similar tests using an Excel spreadsheet that allowed us to run a macro based on the Monte Carlo pricing, but here, Sandra allows us to also test Black-Scholes and Binomial.

Intel Core i7-4770K - Sandra Financial Analysis

The six-cores again reign supreme.

Mathematics: 7-Zip & Euler3D

When hard drives densities measured in the megabytes or single-digit gigabytes, data compression became something that even the layman computer user took advantage of. In fact, even entire hard drives could be used in compressed mode to help increase the overall storage. Today, such methods aren’t required thanks to hard drives ranging in the thousands of gigabytes, but compression is still used on a regular basis by many people, either for storing a folder for backup, encoding music, converting a photo and et cetera. On servers, compression is often used to shrink mega-large log files.

For our compression testing, we enlist the help of 7-zip 9.20. We take a 772MB folder that consists of 39,236 highly-compressible files and archive it using an ‘Ultra’ level of compression using the LZMA2 algorithm. This results in an archive weighing in at about 137MB.

7-Zip

Intel Core i7-4770K - 7-Zip

Given the fact that LZMA2 can make use of all 12 threads on the biggest CPUs we test, it seems that actually having to rely on all of them is going to be a scarce thing.

Euler3D

In terms of complexity, Euler3D is one of our most advanced benchmarks, and also one of the quickest to run. It calculates the fluid dynamics properties of the AGARD 445.6 aeroelastic test wing as it was tested in-house at NASA’s Langley Research center. It’s calculated using Euler equations, with results printed out as Hz and time-to-complete (seconds). A benchmark such as this is useful to those who work designing products where physics has to be considered, whether it be a wing, a car, a ship and so on.

Euler3D

Intel Core i7-4770K - Euler3D

As with previous synthetic benchmarks, the performance boosts with the six-cores are nicely exhibited here.

Compile & Execution: SPEC CPU2006

SPEC’s CPU2006 is the most comprehensive benchmark in our test suite. Its goal is to test both the general execution performance of a machine and also the chosen compiler, and as such, it makes great use of all available threads across one or more CPUs along with the memory sub-system.

You might not have heard of SPEC before, and if so, it’s likely because the non-profit group creates benchmarks targeted at the enterprise rather than the desktop. The folks responsible for each one of its benchmarks take things extremely seriously, and nothing gets released without extensive review. Many companies belong to SPEC as members, offering input and other insight. Some of these include AMD, Intel, Apple, ASUS, HP, Fujitsu, IBM, Lenovo, Microsoft, NEC, NVIDIA, Novell, Red Hat, Super Micro, VMware, Dell and EMC.

The CPU2006 suite is a about as complicated to explain as it is to run. We’ve prepared what we feel to be the best possible configuration for use with the tool, and as the result of much testing, we use Intel Compiler version 12 coupled with Microsoft Visual Studio 2008 for our testing. This is one of the few current configurations that can deliver submittable results, as Intel Compiler supports the most recent C standard, C99, whereas most compilers do not (in Linux, gcc would be a good replacement).

SPEC CPU2006

Due to its inherent design to run each test three times over, we do not run the entire CPU2006 more than once, as it would be redundant. At the same time, a full run on an Intel Core i7-2600K takes just over 13 hours to complete, so it’s not feasible to run the entire suite multiple times over. Because all current CPUs include AVX acceleration, we enable that in here in our testing.

More information on the suite and how we use it can be read about in this forum post.

Intel Core i7-4770K - SPEC CPU2006

Before diving into testing the i7-4960X, I had guessed that CPU2006 might be a benchmark where rather substantial gains would be seen – but again, no. The 4770K outperforms it overall, which really says more about the excellent architecture of Haswell than it does the downsides of IV-E. SPEC is putting the final touches on the next iteration of its CPU benchmark, so it’ll be interesting to re-test these CPUs when that gets released.

CINT2006i7-4960Xi7-3970Xi7-4770Ki7-3770K
400.perlbench33.533.039.832.8
401.bzip226.725.828.926.8
403.gcc30.629.835.633.9
429.mcf78.675.283.982.0
445.gobmk25.025.126.526.0
456.hmmer65.560.468.565.5
458.sjeng31.230.836.331.4
462.libquantum1380135011001100
464.h264ref42.040.946.644.7
471.omnetpp28.727.630.830.5
473.astar33.532.635.732.8
483.xalancbmk48.146.556.146.8
CFP2006i7-4960Xi7-3970Xi7-4770Ki7-3770K
410.bwaves181178143132
416.gamess28.827.732.830.2
433.milc69.364.978.077.1
434.zeusmp122115107111
435.gromacs32.231.036.835.5
436.cactusADM242240232178
437.leslie3d12011989.486.7
444.namd27.427.328.927.2
447.dealII51.649.855.853.7
450.soplex50.144.147.746.2
453.povray50.649.360.950.2
454.calculix42.442.548.245.1
459.GemsFDTD89.588.071.367.7
465.tonto31.930.234.933.4
470.lbm228219162156
481.wrf88.586.188.286.3
482.sphinx362.660.964.864.4

The performance improvements we saw the 4770K deliver over the 3770K in our look at that CPU last month didn’t seem overly impressive, but it’s too bad we didn’t see at least that sort of gain with the 4960X over the 3970X.

System: Futuremark PCMark

Futuremark’s no stranger to most enthusiasts, as its benchmarking software has been considered a de facto standard for about as long as it’s been fun to benchmark. While its 3DMark software is undoubtedly the company’s most popular offering, PCMark is a great tool for summing up the performance of a PC with gaming being a minor focus rather than a major one.

Futuremark’s latest PCMark, 8, consists of five main test suites: Home, Creative, Work, Storage, and Applications. The goal of each is to show how a system will perform overall in a given scenario, and their titles sum up each respective goal nicely. The Applications suite consists of two sub-suites; one for Adobe’s Creative Suite (or Creative Cloud), and the other for Microsoft Office. Of all these suites, we run them all except for the Storage, as it’s not that relevant.

For fun, we also include the overall test results with PCMark 7 (just can’t bear to let it go!).

Futuremark PCMark 8

Intel Core i7-4770K - PCMark 8 - Home

Intel Core i7-4770K - PCMark 8 - Work

Intel Core i7-4770K - PCMark 8 - Creative

Intel Core i7-4770K - PCMark 8 - Adobe CS6

Intel Core i7-4770K - PCMark 8 - Microsoft Office 2013

Intel’s Core i7-4770K dominates in all tests except for the one we all could have guessed a six-core would excel: Creative. Overall, it’s safe to say that for general computing purposes, all of the CPUs are going to suffice, but for when high performance is required, the 4770K gets the nod – unless you spend most of your time video encoding or 3D rendering.

System: Sandra Memory, Cache Performance & Multi-Core Efficiency

The faster the processor, the better its bandwidth and latencies are. Where memory is concerned, however, there are many more factors at play. While frequency plays a major role in overall memory performance, the memory controller can make an even greater improvement, based on its implementation and also its capabilities.

With Intel’s Sandy Bridge-E, we were given a quad-channel controller, while Intel’s (and AMD’s) other platforms stick to a dual-channel design. A quad-channel controller could in theory provide twice as much bandwidth as a dual-channel one. How the controller is integrated into its chip along with the memory’s frequency determines the latency.

While faster memory bandwidth and lower latencies can improve overall computer performance, the faster each core can work with one another along with how much bandwidth a cache can handle rounds out the most important factors of PC performance. The results of all of these are tackled on this page.

SiSoftware Sandra 2013 SP3a

Intel Core i7-4770K - Sandra Memory Bandwidth

Note: Haswell = DDR3-2133; Ivy Bridge-E = DDR3-1333. The reason this was done was because on our X79 test motherboard, the CPU automatically became overclocked when we adjusted our RAM to DDR3-2133 speeds (or any speeds, for that matter). At CPU stock speeds and DDR3-2133 RAM speeds, the results on IV-E are closer to 45GB/s.

Intel Core i7-4770K - Sandra Cache Bandwidth

Intel Core i7-4770K - Sandra Cache & Memory Latency

Intel Core i7-4770K - Sandra Multi-Core Latency

Latency-wise, Intel’s quad-cores win hands-down. With raw bandwidth, including memory, the six-cores do. It’s a total toss-up here, though for those requiring big memory bandwidth, the only choice is IV-E (or SB-E).

Gaming: Futuremark 3DMark & Total War: SHOGUN 2

Game benchmarks stand to see the least amount of gain in comparison to our other tests, but they’re necessary for the sake of completeness. Also, while we benchmark hands-on for our graphics card content, we opt for synthetic testing here, as we’re utilizing the same GPU across each setup.

First up is the ever-popular 3DMark benchmark, and for the sake of completeness, we run all three tests (Ice Storm, Cloud Gate and Fire Strike).

Futuremark 3DMark

Intel Core i7-4770K - Futuremark 3DMark

The differences might be small, but here’s one test where the i7-4960X finally comes out ahead of the rest.

Total War: SHOGUN 2

Real-time and turn-based strategy games tend to be the most stressful on both the GPU and CPU, and Total War: SHOGUN 2 does well to live up to that stereotype. The game is so stressful on a PC, in fact, that the developers included built-in benchmarks that are meant to test a PC in a worst-case scenario sort of way. For our testing here, we use we use both the 720p GPU and CPU benchmarks.

Total War: SHOGUN 2

Intel Core i7-4770K - Total War: SHOGUN 2

GPU-wise, the quads win, as has become a theme, though overall performance is what I’d consider non-noticeable in the real-world.

Power Consumption & Final Thoughts

To help consumers understand what sort of power their CPU will draw from the wall, companies like AMD and Intel give us a “TDP” rating, which is in effect the realistic top-end wattage the CPU will hit. CPUs can sometimes go above these wattages, but that’s typically only when tools are used to specifically stress-test the processor.

To help keep our CPU at 100% load during a realistic scenario, we make use of our CyberLink MediaExpresso benchmark; specifically, the CPU Only + Best Quality configuration. But please note: the load results below are just that, (max) “load”. Most often (or never, depending on your usage), your CPU is not going to be at full load, making this a worst case scenario. Also, different motherboards can introduce a variance of up to 10 watts, so there’s no true apples-to-apples comparison unless more than one CPU is benchmarked on the same platform.

Our methodology here is simple: boot the machine up, let it sit idle for 10 minutes, capture the idle wattage, and then begin the benchmark. Because the encode stresses the CPU 100% throughout most of the encode, we only monitor the last couple of minutes’ worth.

Intel Core i7-4770K - Futuremark 3DMark

We might not have seen major performance improvements with IV-E over SB-E, but power consumption wise, some nice improvements can be seen. Alright – awesome improvements. A drop of 53W at full-load? That’s what I’m talking about.

Final Thoughts

Intel’s Core i7-4960X (and Ivy Bridge-E in general) is easy to sum-up, but not for the greatest of reasons. I’d love to be able to write that IV-E is a major improvement over SB-E, but it isn’t. I’d like to say that it dominates the Haswell-based Core i7-4770K, but it doesn’t. While there are clear improvements in certain scenarios, they’re mostly confined to video and 3D rendering. In most other regards, the architectures in both the 4770K and 3770K prove to be superior to IV-E.

As an “enthusiast” platform, IV-E does offer a couple of important perks: quad-channel memory, and more robust PCIe possibilities. For the latter perk alone, X79 + IV-E, despite falling short in the performance department a lot of the time, is going to be the best choice for those planning to build higher-end multi-GPU setups. You’re sure to be fine with mainstream cards in SLI, but with cards like the GTX 770 or higher, the extra PCIe lanes could be beneficial. This is especially true with 3-way or 4-way – X79 is an absolute no-brainer here. Running 3 GPUs at x16 each is going to be hugely preferable to running them at x8 / x8 / x4, which would be the configuration on Z87.

And again – for those with insatiable memory bandwidth needs, X79 again can’t be beat. For some, 45GB/s @ DDR3-2133 is going to be an important gain over 25GB/s on Z87 at the same speeds. Admittedly, I am not quite sure of a scenario that will make use of this extra bandwidth, but those that do are sure to be on the workstation / server side.

Intel Ivy Bridge-E Wafer Close-up
Ivy Bridge-E Wafer

Had Intel released the i7-4960X as an eight-core / sixteen-thread part as I would have predicted two-years-ago, this review would have been a lot more positive. The IPC (instructions-per-clock) issues wouldn’t have been fixed, of course, so a CPU like the i7-4770K would have still come out ahead in single-thread tests. But with a literal doubling of the cores, the CPU would have handily dominated most tests, whereas here, the 4770K matches or beats it more often than not.

IV-E’s design supports 8 cores, as SB-E did. So what’s the reason Intel didn’t go that route? The answer is likely two-fold: At 130W, adding two additional cores would have likely pushed the CPU above 150W at the frequencies Intel would have wanted to offer it at. At six-cores, these processors run hot enough, and while eight-core Xeon chips exist, none of them run at speeds like this 4960X does.

Thanks to these issues, Intel might have had wanted to release a lower-clocked eight-core part instead – but how attractive would that have looked? Not very, I’m afraid. In that case, we’d be crippling single-threaded performance to gain in multi-threaded performance, but few of Intel’s target audience for IV-E is going to be running strenuous workloads all-too-often.

Then there’s the even more important fact: Had Intel released an 8-core desktop part, companies in the enterprise might purchase those instead of the pricier Xeon parts. Intel of course doesn’t want that. So this means that the desktop folk might be able to pick up an Intel 8-core part as soon as 10+ core CPUs become common on the Xeon side. This is the same sort of reason that Intel drops certain virtualization technologies off of its unlocked desktop parts, such as the 4770K. Intel makes mad bank off of its Xeon product-line, so it’s not going to want to risk that by releasing an overly impressive desktop part, as unfortunate as that is.

It stands to reason, then, that the i7-4960X shouldn’t be priced as it is, but alas, I don’t think Intel could imagine foregoing its $1,000 part, and because it does include more cache, it’s going to be attractive to those planning to take full advantage of what the processor offers.

Summing-up, IV-E is worth going with if you are building a truly high-end or workstation rig; for everyone else, you’re probably better-off building a much more affordable Z87 rig instead.

If it were Haswell-E we were reviewing here, the outcome would have been far different.

Copyright © 2005-2019 Techgage Networks Inc. - All Rights Reserved.