Sign in with
Sign up | Sign in
Core i7-3970X Extreme Review: Can It Stomp An Eight-Core Xeon?
By ,
1. Core i7-3970X Extreme: Six Cores And Up To 4 GHz

Just one week ago, I was putting together some quick thoughts for Win A Custom-Painted Falcon Northwest Tiki! "It’s amazing," I thought, "that you can get such compelling performance from a 77 W Ivy Bridge-based processor." As a result of those manageable thermals, companies like Falcon Northwest are able to cram a Core i7-3770K, overclocked, into a little mini-ITX-based machine.

Of course, it didn't even take a week before Intel set out to show me it could also go the other direction, manufacturing an even faster desktop-oriented CPU capable of sucking down twice as much power as the -3770K. Meet the new Core i7-3970X.

As the new flagship of Intel’s Sandy Bridge-E-based Core i7-3000 family, this chip succeeds the Core i7-3960X, which was launched almost exactly one year ago. It drops into the LGA 2011 processor interface, and it sells for the same $1000.

But whereas the Core i7-3960X is a 130 W CPU, Intel rates its Core i7-3970X for 150 W.

Only one other time in history have we seen a 150 W desktop part, and that was the Core 2 Extreme QX9775. Back in 2008, Intel introduced the LGA 771-based QX9775 to populate its Skulltrail platform—a dual-socket configuration aimed at content creation professionals.

Today’s Core i7-3970X is only intended to operate in single-processor setups. Its higher thermal ceiling comes from a speed bump. The -3960X features a base clock rate of 3.3 GHz. Turbo Boost takes it as high as 3.9 GHz in single-threaded apps, when headroom allows. The -3970X starts at 3.5 GHz, but then accelerates up to 4 GHz via Turbo Boost.

Both processors feature six active cores, plus Hyper-Threading, allowing them to address 12 threads concurrently. They similarly feature 15 MB of shared L3 cache, and are manufactured using the same 32 nm process. Naturally, the two Extreme Edition chips boast unlocked multiplier ratios, too.

Does anything about the Core i7-3970X alter what we’re able to achieve with regard to overclocking? Not particularly. The Core i7-3960X we tested a year ago ran stably at 4.6 GHz with all six of its cores fully utilized. Today’s -3970X rests comfortably at 4.7 GHz using the same closed-loop liquid cooler.

Higher speeds are possible using more aggressive voltages (I was trying to stay under 1.4 V), but then you have to worry about thermals. The Core i7-3970X needs to be kept under 91 degrees Celsius in order to avoid throttling, and this becomes increasingly difficult at higher voltage settings.

You'll want an RTS2011LC (or better) thermal solution to dissipate -3970X's 150 W.You'll want an RTS2011LC (or better) thermal solution to dissipate -3970X's 150 W.

Cooling is a naturally an important consideration, then. Intel says that its RTS2011AC air cooler, which sells for just under $30, is not sufficient for the Core i7-3970X. However, the RTS2011LC closed-loop thermal solution we’ve been using for testing is able to handle 150 W CPUs. Since many vendors don’t get specific about their thermal performance, you’ll need to check with your vendor of choice about compatibility. And because Intel doesn't ship its LGA 2011-based models with coolers, plan on spending $70 to $80 more for something appropriate.

The Competitive Landscape

Wait, what competitive landscape? Core i7-3960X went unmatched for a year. AMD’s fastest chip, the FX-8350, does battle in the middle of Intel's Ivy Bridge family. A Core i7-3770K is a smart choice for high-end gaming PCs. But when it comes to running threaded apps in a professional capacity or setting up a multi-card gaming box, Sandy Bridge-E-based Core i7s remain the best of the best.

We’re including all of the usual suspects in today’s benchmarks, including the Core i7-3960X and Core i7-3930K. But we also want to see how Core i7-3970X fares against the other 150 W processor in Intel’s current portfolio: the Xeon E5-2687W.

Equipped with eight cores and 20 MB of shared L3 cache, the Xeon is what Sandy Bridge-E really could be if Intel didn't disable parts of it. But because more of the die’s resources are enabled on the Xeon, generating heat, Intel has to be more conservative about its clock rates. A base 3.1 GHz frequency is comparatively slower than the Core i7 in heavily-threaded apps. But we’re guessing that, in those same workloads, having more cores helps compensate. In single-threaded benchmarks, the E5-2687W spins up to 3.8 GHz via Turbo Boost.

The comparison isn’t anywhere close to fair, of course. Because the Xeon is designed to drop into dual-socket motherboards, Intel charges more for it. A lot more. Newegg has the processor listed for $1900, or almost twice as much as the new Core i7 Extreme—which brings up another point. The Xeon doesn’t have an unlocked multiplier, so you can’t overclock it.

But hey, two Xeon E5-2687Ws together are simply unbeatable. Don’t believe me? Check out Intel Xeon E5-2600: Doing Damage With Two Eight-Core CPUs.

2. Test Setup And Software
Test Hardware
ProcessorsIntel Core i7-3970X (Sandy Bridge-E) 3.5 GHz (35 * 100 MHz), LGA 2011, 15 MB Shared L3, Hyper-Threading enabled, Turbo Boost enabled, Power-savings enabled

Intel Core i7-3960X (Sandy Bridge-E) 3.3 GHz (33 * 100 MHz), LGA 2011, 15 MB Shared L3, Hyper-Threading enabled, Turbo Boost enabled, Power-savings enabled

Intel Core i7-3930K (Sandy Bridge-E) 3.2 GHz (32 * 100 MHz), LGA 2011, 12 MB Shared L3, Hyper-Threading enabled, Turbo Boost enabled, Power-savings enabled

Intel Xeon E5-2687W (Sandy Bridge-EP) 3.1 GHz (31 * 100 MHz), LGA 2011, 20 MB Shared L3, Hyper-Threading enabled, Turbo Boost enabled, Power-savings enabled

AMD FX-8350 (Vishera) 4.0 GHz (20 * 200 MHz), Socket AM3+, 8 MB Shared L3, Turbo Core enabled, Power-savings enabled

AMD FX-8150 (Zambezi) 3.6 GHz (18 * 200 MHz), Socket AM3+, 8 MB Shared L3, Turbo Core enabled, Power-savings enabled

AMD Phenom II X4 980 BE (Deneb) 3.7 GHz (18.5 * 200 MHz), Socket AM3, 6 MB Shared L3, Power-savings enabled

AMD Phenom II X6 1100T (Thuban) 3.3 GHz (16.5 * 200 MHz), Socket AM3, 6 MB Shared L3, Turbo Core enabled, Power-savings enabled

Intel Core i7-3770K (Ivy Bridge) 3.5 GHz (35 * 100 MHz), LGA 1155, 8 MB Shared L3, Hyper-Threading enabled, Turbo Boost enabled, Power-savings enabled

Intel Core i5-3570K (Ivy Bridge) 3.4 GHz (34 * 100 MHz), LGA 1155, 6 MB Shared L3, Turbo Boost enabled, Power-savings enabled

Intel Core i5-3470 (Ivy Bridge) 3.2 GHz (32 * 100 MHz), LGA 1155, 6 MB Shared L3, Turbo Boost enabled, Power-savings enabled

Intel Core i5-2550K (Sandy Bridge) 3.4 GHz (34 * 100 MHz), LGA 1155, 6 MB Shared L3, Turbo Boost enabled, Power-savings enabled
MotherboardGigabyte X79S-UP5-WiFi (LGA 2011) Intel C606, BIOS F4m

ASRock Fatal1ty 990FX Professional (Socket AM3+) AMD 990FX/SB950 Chipset, BIOS 1.9

Gigabyte Z77X-UD3H (LGA 1155) Intel Z77 Express, BIOS F17
Memory
G.Skill 16 GB (4 x 4 GB) DDR3-1600, F3-12800CL9Q2-32GBZL @ DDR3-1600 at 1.5 V
Hard Drive
Crucial m4 256 GB, SATA 6 Gb/s
Graphics
Nvidia GeForce GTX 680 2 GB
Power Supply
Cooler Master UCP-1000 W
System Software And Drivers
Operating System
Windows 7 Ultimate 64-bit
DirectX
DirectX 11
Graphics DriverNvidia GeForce Release 306.97


We've transitioned over to Gigabyte's X79S-UP5-WiFi motherboard for benchmarking LGA 2011-based processors. Armed with Intel's C606 chipset, this board is designed to accommodate Intel's Xeon E5 processors. Not all motherboards include this support. So, if you plan to do a little Xeon testing of your own, make sure your platform features the requisite firmware to recognize Intel's more business-oriented processor family.

3D Game Benchmarks And Settings
BenchmarkDetails
The Elder Scrolls V: Skyrim
Game Settings: High and Ultra Quality Settings, Anti-Aliasing: FXAA, V-sync: Disabled, 1680x1050, 1920x1200, 2560x1600, Custom Demo, 25-second Fraps run
Battlefield 3
Game Settings: Ultra Quality Settings, Anti Aliasing: Disabled and 4xMSAA (Deferred)/High (Post), Anisotropic Filtering: 16x, Vertical Sync: Off, 1680x1050, 1920x1080, 2560x1600, Demo: Going Hunting, 90-second Fraps run
World of Warcraft: Mists of Pandaria
Game Settings: Ultra Quality Settings, Anti Aliasing: 1x AA and 8x AA, Anisotropic Filtering: 16x, Vertical Sync: Disabled, 1680x1050, 1920x1080, 2560x1600, Demo: Crushblow to The Krazzworks, DirectX 11
Audio Benchmarks And Settings
BenchmarkDetails
iTunesVersion: 10.4.1.10, 64-bit
Audio CD ("Terminator II" SE), 53 min., Convert to AAC audio format
Lame MP3Version 3.99
Audio CD "Terminator II SE", 53 min, convert WAV to MP3 audio format, Command: -b 160 --nores (160 Kb/s)
Video Benchmarks And Settings
BenchmarkDetails
HandBrake CLIVersion: 0.9.8
Video: Big Buck Bunny (720x480, 23.972 frames) 5 Minutes, Audio: Dolby Digital, 48 000 Hz, Six-Channel, English, to Video: AVC Audio: AC3 Audio2: AAC (High Profile)
MainConcept Reference v2.2
Version: 2.2.0.5440
MPEG-2 to H.264, MainConcept H.264/AVC Codec, 28 sec HDTV 1920x1080 (MPEG-2), Audio:
MPEG-2 (44.1 kHz, 2 Channel, 16-Bit, 224 Kb/s), Codec: H.264 Pro, Mode: PAL 50i (25 FPS), Profile: H.264 BD HDMV
Application Benchmarks And Settings
BenchmarkDetails
WinRARVersion 4.20
RAR, Syntax "winrar a -r -m3", Benchmark: 2012-THG-Workload, 1.35 GB
WinZip
Version 17
WinZip Commandline Version 3, ZIPX, Syntax "-a -ez -p -r", Benchmark: 2012-THG-Workload, 1.35 GB
7-Zip
Version 9.20 (x64)
LZMA2, Syntax "a -t7z -r -m0=LZMA2 -mx=5", Benchmark: 2012-THG-Workload, 1.35 GB
Adobe Premiere Pro CS 6
Hollywood Sequence to H.264 Blu-ray, Output 1920x1080, Maximum Quality
Adobe After Effects CS 6
Create Video which includes three Streams
Frames: 210, Render Multiple Frames Simultaneously: on
Cinebench
Version 11.5 Build CB25720DEMO
CPU Test single- and multi-threaded
BlenderVersion: 2.63, Cycles Engine
Syntax blender -b thg.blend -f 1, Resolution: 1920x1080, Anti-Aliasing: 8x, Render: THG.blend frame 1
Adobe Photoshop CS 6 (64-Bit)Version: 11
Filtering a 16 MB TIF (15 000x7266), Filters:, Radial Blur (Amount: 10, Method: zoom, Quality: good) Shape Blur (Radius: 46 px; custom shape: Trademark sysmbol) Median (Radius: 1px) Polar Coordinates (Rectangular to Polar)
ABBYY FineReaderVersion: 10 Professional Build (10.0.102.82)
Read PDF save to Doc, Source: Political Economy (J. Broadhurst 1842) 111 Pages
3ds Max 2012
Render Space Flyby, 1440x1080, from Y: RAM Drive
Adobe Acrobat X Professional
PDF Document Creation (Print) from Microsoft PowerPoint 2010
SolidWorks 2010
PhotoView 360, 01-Lighter Explode.SLDASM Benchmark File, 1920x1080 Render, 1.44 Million Polygons, 256 AA Samples
Visual Studio 2010
Google Chrome Compile, Scripted
Synthetic Benchmarks And Settings
BenchmarkDetails
PCMark 7Version: 1.0.4
3DMark 11
Version 1.0.3
SiSoftware Sandra
Version: 2013 Beta
Processor Arithmetic, Multimedia, Cryptography, Memory Bandwidth, .NET Arithmetic, .NET Multimedia
3. Benchmark Results: PCMark 7

Thanks to its higher clock rate, the Core i7-3970X stands up to Intel’s Core i7-3770K in PCMark’s overall suite score. From experience, we know that several components of this synthetic test are really only able to tax four cores effectively. So, the Sandy Bridge-E-based chip’s extra resources are underutilized, while the more mainstream CPU’s Ivy Bridge architecture gives it an advantage.

The big, expensive Xeon can’t reach quite as high, its Turbo Boost implementation limited to 3.8 GHz. This eight-core processor goes largely unused, forcing it into fourth place.

The same combination of variables is exacerbated in PCMark’s Productivity suite. Three Ivy Bridge-based chips huddle at the top, followed by five variations of Sandy Bridge-based hardware. Four AMD processors bring up the rear, unable to achieve the performance per clock cycle needed to compete.

Parallelism is rewarded in PCMark’s Creativity sub-routine, and Intel’s Core i7-3970X grabs a first-place finish. Why doesn’t the eight-core Xeon get that honor? Presumably, a higher clock rate under full load gives the six-core CPU an advantage over the eight-core model running slower.

The same goes for the Entertainment test, where both the Core i7-3970X and -3960X outmaneuver Intel’s Xeon E5-2687W.

4. Benchmark Results: 3DMark 11

Our GeForce GTX 680 benefits most from Intel’s Xeon E5, though our three other Sandy Bridge-E-based configurations aren’t far behind. For the money, though, the Ivy Bridge-based options in fifth and sixth place are almost certainly better values, particularly in machines with just one graphics card.

In a purely graphics-bound workload, processor performance turns out to mean very little. The Core i7-3970X takes second-to-last place, but all of these systems are within a hair of each other.

This benchmark’s Physics test specifically isolates CPU performance, fully utilizing each core. The Xeon takes a commanding lead, followed by the three Sandy Bridge-E-based Core i7s.

5. Benchmark Results: Sandra 2013

Memory bandwidth and hardware-accelerated AES encryption/decryption performance are closely related, since the processor is able to crank through instructions as fast as the memory subsystem can feed them in. As a result, the LGA 2011-based platforms, with their quad-channel architectures, dominate.

.NET performance is particularly important now that Windows 8 is available, along with a growing library of Metro UI-based applications.

6. Benchmark Results: Content Creation

3ds Max is fully threaded, so Intel’s Xeon enjoys a quantifiable advantage. The Core i7-3970X does really well, though, besting its predecessor by a few seconds.

We don’t see the same scaling from Blender’s Cycles engine as we did in 3ds Max. Nevertheless, the Core i7-3970X, -3960X, and Xeon E5-2687W all fall within seconds of each other.

Cinebench gives us the ability to isolate single-core performance and then fully utilize each processor. Not surprisingly, Ivy Bridge-based processors generally demonstrate the best single-core ratings. Cores and clock rate build on those figures though, and the Sandy Bridge-E-based chips surge ahead, led by Intel’s Xeon E5-2687W.

Similar to 3ds Max, SolidWorks’ PhotoView 360 completely utilizes the Xeon, rewarding its eight cores with a first-place finish. The new Core i7-3970X runs at higher clock rates, but because it employs two fewer cores, it slides back into second place.

7. Benchmark Results: Adobe CS 6

The threaded filters in our Photoshop workload take full advantage of the Xeon E5-2687W’s eight cores, allowing it to wrap up this benchmark in half the time as Intel’s Core i5-3570K. Intel’s new Core i7-3970X lags behind by 10 seconds, outperforming the Core i7-3960X by a scant one second.

Our second Photoshop test exploits the software’s OpenCL support, tapping our system’s GeForce GTX 680 to help accelerate the benchmark.

CPU performance is still a vital component, though, and we see similar scaling as some of our other well-threaded metrics. The Xeon places first, followed by the Core i7-3970X, -3960X, and -3930K. A number of Intel’s quad-core and AMD’s quad-module processors fill in the rest of the chart.

The same story applies to Premiere Pro, which favors the eight-core Xeon, but also runs very fast on the six-core LGA 2011-based chips. Although they’re great for the price, Intel’s Ivy Bridge- and AMD’s Vishera-based processors are more mainstream; they cannot keep up.

We’ve seen After Effects respond favorably to memory subsystem improvements, but it tends to scale less aggressively based on processor performance. A mere four seconds separate the first six finishers in our benchmark. There are only eight seconds between the octet of Intel products we’re testing.

8. Benchmark Results: Productivity

Today marks the first time I’ve ever seen a single-processor system break under the one-minute threshold in our ABBYY FineReader 10 test. This application utilizes as many cores as you throw at it, allowing the Xeon E5-2687W to wrap up while AMD’s “eight-core” FX-8150 is still only halfway done.

Intel’s Core i7-3970X finishes in second, but is only three seconds faster than Core i7-3930K, which sells for less than $600.

Printing a PowerPoint document to PDF is another single-threaded workload. So, the higher IPC throughput of Intel’s Ivy Bridge architecture prevails. It’s only by the implementation of Turbo Boost that the Core i7-3970X accelerates to 4 GHz, beating Intel’s Core i7-3470, which is limited to 3.6 GHz.

In sharp contrast, our Visual Studio 2010 benchmark is very well threaded. It actually tends to take longer than any other test in our suite. On a Phenom II X4 980, for instance, compiling Google Chrome is an almost 40-minute process.

If you’re lucky enough to own a Xeon E5-2687W, however, the whole job finishes in just over 14 minutes. A Core i7-3970X makes you wait a couple of minutes longer, but once you step into the world of LGA 1155-based quad-core CPUs, performance really starts to drop off.

The latest version of the German chess program Fritz puts Intel’s Xeon E5 to good use, more than doubling the number of kilonodes/second AMD’s FX-8150 achieves.

9. Benchmark Results: Compression Apps

Corel’s recent introduction of WinZip 17 fixes much of what plagued the application’s performance in the past. Finally, it’s able to fully utilize all available processing cores. Moreover, OpenCL support accelerates the compression of files larger than 8 MB. Unfortunately, as it stands, our 1.35 GB test folder includes very few files that large, so turning OpenCL on doesn’t have much of an impact.

The Xeon finishes in first place, confirming that WinZip is now able to completely utilize the resources available to it. Intel’s new Core i7-3970X places second, followed by its predecessor, the -3960X.

Perhaps we’ve been unfairly critical of WinRAR in the past, downplaying the extent to which it’s able to utilize available cores. The fact that we again see Intel’s Xeon E5 finish first, followed by three Sandy Bridge-E-based Core i7s, illustrates that this application does scale beyond four cores.

7-Zip does as well. Even the Xeon E5’s two extra cores clearly make a really big difference. The two-core advantage that a 3.5 GHz Core i7-3970X holds over the Core i7-3770K is equally significant, despite the Ivy Bridge architecture’s superior efficiency.

We remain big fans of the Core i7-3570K for its value. But, at least in this metric, you’ll wait almost a minute more for 1.35 GB to compress on a system with that CPU compared to Intel’s newest Core i7.

10. Benchmark Results: Media Encoding

Our MainConcept benchmark isn’t as demanding as it once was; a majority of our test subjects finish the task in less than one minute. But we’re still able to see the application taxing the Xeon’s eight cores, putting it in first place. The fast six-core Core i7-3970X takes second, while the -3960X and -3930K follow, separated by one second each.

The same analysis applies to the latest version of HandBrake, which is similarly optimized for multi-core CPUs.

So much of our suite is heavily threaded. And while you can run multiple conversions in parallel to tackle multiple files simultaneously, our Lame benchmark taxes one core at a time.

Suddenly, Intel’s Ivy Bridge-based Core i7-3770K find itself in first place, accelerated by Turbo Boost technology. But because the Core i7-3970X also enjoys a frequency bump to 4 GHz in single-threaded tasks, it outperforms the Core i5-3570K for a second-place finish.

Most glaring is the divide between Intel’s slowest chip, the Xeon, and AMD’s fastest, based on its Bulldozer architecture.

On the other hand, converting a WAV file to AAC in iTunes is a single-threaded operation. Ivy Bridge’s per-clock advantage, coupled with aggressive Turbo Boost clock rates, land the Core i7-3770K and Core i5-3570K in first and second place.

But a maximum Turbo Boost setting of 4 GHz is good enough to get the Core i7-3970X into third, ahead of Intel’s Core i5-3470 processor.

11. Benchmark Results: Battlefield 3

Battlefield 3’s single-player component is decidedly graphics-bound. Consequently, we don’t see enough separation between the CPUs in today’s story to recommend one over the others, even at resolutions as low as 1680x1050.

12. Benchmark Results: The Elder Scrolls V: Skyrim

On the other hand, The Elder Scrolls V: Skyrim does exhibit greater sensitivity to platform performance.

Four Sandy Bridge-E/EP-based CPUs enjoy the lead at 1680x1050, suggesting that some combination of high clock rates and large shared L3 caches help drive performance.

As with any workload that increasingly emphasizes some other component, however, scaling up to 1920x1080 and then 2560x1600 quickly levels off average frame rates. Our highest resolution tips the scales in favor of Intel’s Ivy Bridge architecture. Sandy Bridge-derived CPUs clump together in the middle, while AMD’s portfolio lags behind (albeit by less than 10 FPS, on average, under the High settings preset).

13. Benchmark Results: World of Warcraft: Mists Of Pandaria

World of Warcraft: Mists of Pandaria is also sensitive to processor performance, rewarding Intel for its multi-core designs, large caches, and aggressive Turbo Boost settings. Meanwhile, the average frame rates achieved by AMD’s CPUs trail off rapidly at 1680x1050 and 1920x1080. It’s really only when you’re at 2560x1600 using 8x MSAA and the Ultra quality preset that the Bulldozer- and Piledriver-based FX processors perform comparably.

14. Power Consumption And Efficiency

Logging power consumption during the entire benchmark run gives us a window into the behavior of each processor and platform.

We’ll start by looking at the Core i7-3970X’s green line. A 150 W TDP should have been enough for us to know that high power consumption would be measurable. Indeed, the -3970X spikes up to use the most power under load.

What about the eight-core Xeon E5, which is also a 150 W processor? Despite that rating, its consumption is far below the new Core i7s, and in fact looks to be even more conservative than FX-8350’s power use.

Although the Core i7-3960X is only just slightly slower than the -3970X, the yellow line is quite a bit lower on our chart. And of course, the Core i7-3770K is downright miserly.

Averaging those power numbers out reveals that Intel’s Core i7-3970X fares worst. And although the Xeon E5 bears the same 150 W rating, it ends up using even less power than the 130 W Core i7-3960X.

Calculating efficiency requires that we know two things, though. First, we need the consumption figures above. Then, we need some measure of performance.

We capture this data by logging our scripted benchmark suite in two-second intervals. Multiplying out the number of resulting samples by two, dividing by 60 (minutes), and dividing by 60 again (hours) tells us how long each processor takes to wrap up testing.

Eight cores reign, it appears, though the Core i7-3970X and -3960X aren’t far behind. Opting for a Core i7-3770K gets you significantly lower average power use; however, it also takes 10 minutes longer than the -3970X to complete testing.

We’ll leave the FX-8350 alone in all of this. Its high power consumption and last-place performance are countered only by a $220 price tag.

Not surprisingly, Intel’s 77 W Core i7-3770K is the most efficient choice. More interesting is the fact that the 150 W Xeon combines brutal speed with lower-than-expected consumption to finish second in our efficiency metric.

Core i7-3970X isn’t able to add enough performance to make up for dramatically higher power use, and this reflects in its loss to the -3960X.

Then there’s the FX-8350, which, again, combines modest performance and high average power use, hurting its efficiency story.

15. Core i7-3970X: Faster, But Less Efficient At The Same Price

Intel didn’t have to launch the Core i7-3970X. Its -3960X was already the fastest single-socket workstation processor you could buy. But, in the year that has passed since Sandy Bridge-E first debuted, Intel also introduced its Ivy Bridge architecture. Better per-cycle performance gave enthusiasts something to consider: do you save some money by buying a quad-core Ivy Bridge-based processor, or splurge on an older architecture for its higher core count and more generous PCI Express connectivity?

With the Core i7-3970X, Intel gives power users with money to spend a little more reason to lean in favor of Sandy Bridge-E. An extra 200 MHz under full load is reflected in apps like Adobe Photoshop, Premiere Pro, Autodesk 3ds Max, Microsoft Visual Studio, and Maxon Cinema 4D. A 100 MHz-higher Turbo Boost bump helps iTunes, PowerPoint, and Lame.

Although Intel is maintaining its $1000 pricing on this desktop flagship, you’ll need to pay closer attention to cooling. A massive 150 W TDP is more than Intel’s LGA 2011-oriented heat sink can handle. Instead, the company recommends its closed-loop thermal solution, adding nearly $80 to the already-steep CPU's cost.

Unfortunately, at least in our suite of tests, the extra infusion of performance isn’t significant enough to counteract higher power consumption. Efficiency suffers as a result. There’s a good chance you won’t care if more speed in a money-making application covers the difference.

Though, in that case, we’re obligated to mention Intel’s Xeon E5-2687W. It’s a $1900 processor, yes. It’s also rated for 150 W. But if you need your software to run as quickly as possible, this eight-core monster armed with 20 MB of L3 cache tears through threaded workloads—and it does so more efficiently than even a year-old Core i7-3960X.