Sign in with
Sign up | Sign in
The Core i7-4770K Review: Haswell Is Faster; Desktop Enthusiasts Yawn
By ,
1. Haswell Turns Into Intel's Fourth-Gen Core Architecture

Editor’s Note: Eager to show off what it's doing with Intel’s Haswell architecture, system builder CyberPower PC is offering the Tom’s Hardware audience an opportunity to win a new system based on Intel’s Core i7-4770K processor. Read through our review, and then check out the last page for more information on the system, plus a link to enter our giveaway!

Do you know what it’s like to be at the top of your game, the nearest competitor several strides behind? Well, maybe not. But Intel sure does. When it comes to desktop CPUs, the company’s top-end parts continue to stave off AMD's best efforts. That applies to raw performance and efficiency.

We love fast, and we love efficient. But we also like to see healthy competition driving innovation. And again, on the desktop, there’s not enough of that to push Intel. Ivy Bridge-based CPUs are generally a small step up from the generation prior. And although the Sandy Bridge architecture included a number of notable improvements, unprecedented integration gave away Intel’s growing focus on mobility. Even as we got our hands on great features like Quick Sync, Intel was chiseling away at its enthusiast equity by limiting overclocking to K-series SKUs.

Expect more of the same from Haswell. You're going to see notable per-clock performance improvements, faster graphics, and additional features able to accelerate specific workloads. But you’re also going to witness a clumsy handling of overclocking (again), some strange decisions on the graphics side (again), and incremental gains that’ll have some of us upgrading our desktops, but more folks looking for Haswell-powered mobile platforms.

That's entirely by design, by the way. An emphasis on power is front and center with Haswell. And as a result, this architecture is going to span the broadest range of devices Intel has ever touched with one design. But I’ll argue that enthusiasts on the desktop take a back seat to make it all possible.

Meet Haswell, Now Known As Intel’s Fourth-Gen Core Architecture

Intel is rolling out the details of its Haswell-based processors in a staggered launch. The company plans to ship multiple variations of the architecture across a number of different interfaces, from very low-power segments to very performance-sensitive ones. However, the only arrangement emerging today is the quad-core SoC. Technically, Intel is talking desktop and mobile, though we’re deliberately focusing on the Core i7-4770K desktop CPU. I published a preview of Core i7-4770K’s performance almost three months ago, and that story has some information about Intel’s plans as well.

Haswell-based quad-core processors will ship in two configurations to cover the mobile and desktop markets. Only one is ready today, though. That chip features the HD Graphics 4600 engine, also known as GT2. The second, with Iris Pro Graphics 5200 (or GT3e) is coming later. Intel's engineers claim that Iris Pro scales incredibly well given a lofty power ceiling and enough cooling. However, CPUs endowed with the higher-end graphics engine are BGA-only, meaning they’re soldered down. So, enthusiasts buying LGA 1150-equipped motherboards will only find Core i7 and Core i5 CPUs with four cores and HD Graphics 4600 (technically, there’s also a 35 W Core i5 with fewer cores, but it’s still under wraps).

This implementation of Haswell is composed of 1.6 billion transistors, up from a comparable Ivy Bridge configuration’s 1.4 billion. Optimized expressly for Intel’s 22 nm node, the die measures 177 square millimeters, just slightly larger than quad-core Ivy Bridge at 160 mm².

Quad-Core Haswell with GT2Quad-Core Haswell with GT2

Quad-Core Ivy Bridge with GT2Quad-Core Ivy Bridge with GT2

Put Ivy Bridge and Haswell right next to each other and you might have a difficult time telling them apart. After all, there’s “only” a 200 million-transistor delta separating the two. That 14% growth in transistor count largely comes from a 25% increase in graphics resources compared to last generation.

That’s not to say the processor cores go untouched. Intel says it put specific emphasis on speeding up both today’s legacy code as well as applications we’ll see in the future. To that end, larger buffers enlarge the out-of-order window, which means instructions that would have previously waited for execution can be located and processed sooner. Haswell’s window is 192 instructions. Sandy Bridge was 168. Nehalem was 128. The Haswell branch predictor is improved, too. This is something Intel manages to do every generation—and for good reason, since it simultaneously enables better performance and prevents the wasted work of a branch getting predicted incorrectly. Previously, Intel’s architecture was able to execute six operations per clock cycle. However, Haswell gets two additional ports (one integer ALU and one store), enabling up to eight operations per cycle. And workloads with large data sets should see a benefit from a larger L2 TLB.

All of those changes add up to significant improvement in Haswell’s IPC compared to Ivy Bridge. That’s where we expect most of the speed-up in general-purpose apps to come from this generation, since the top-end Core i7-4770K runs at the same 3.5 GHz as -3770K.

Sure enough, when we set five different processors (employing four different architectures) to the same constant 4 GHz, we see, first, how much more work Intel gets done compared to AMD and, second, a steady progression forward in Intel’s performance.

In addition to the two execution ports Intel adds to Haswell, ports one and two now feature 256-bit Fused Multiply-Add units, doubling the number of peak theoretical floating-point operations per cycle. Integer math gets a big boost as well from AVX2 instruction support.

Of course, multiplying the architecture’s compute potential means little if you can’t get data into the core fast enough. So, Intel also made a number of changes to its caches. Haswell’s L1 and L2 caches are the same size as they were in Ivy Bridge (there’s a 32 KB L1 data, 32 KB L1 instruction, and 256 KB L2 cache per core). Bandwidth to the caches is up to doubled, though, and we’ll see in our synthetic testing that the L1D is indeed quite a bit faster. Intel claims that it can do one read every cycle from the L2 (versus one read every other cycle in Ivy Bridge), but we aren’t able to replicate those figures in our own testing.


Cores / Threads
Base Freq.
Max. Turbo
L3
HD Graphics
Graphics Max Freq.
TDP
Price
Fourth-Gen Core i7 Family
4770T
4/8
2.5 GHz
3.7 GHz
8 MB
4600
1,200 MHz
45 W
$303
4770S
4/83.1 GHz
3.9 GHz
8 MB46001,200 MHz65 W
$303
4770
4/83.4 GHz
3.9 GHz
8 MB46001,200 MHz84 W
$303
4770K
4/83.5 GHz
3.9 GHz8 MB46001,250 MHz84 W
$339
4770R
4/83.2 GHz
3.9 GHz6 MB
Iris Pro 5200
1,300 MHz65 W
N/A
4765T
4/82.0 GHz
3.0 GHz
8 MB46001,200 MHz35 W
$303
Fourth-Gen Core i5 Family
4670T
4/4
2.3 GHz
3.3 GHz
6 MB
46001,200 MHz45 W
$213
4670S
4/43.1 GHz
3.8 GHz
6 MB46001,200 MHz65 W
$213
4670K
4/43.4 GHz
3.8 GHz
6 MB46001,200 MHz84 W
$242
4670
4/43.4 GHz
3.8 GHz
6 MB46001,200 MHz84 W
$213
4570
4/43.2 GHz
3.6 GHz
6 MB46001,150 MHz
84 W
$192
4570S
4/42.9 GHz
3.6 GHz
6 MB46001,150 MHz65 W
$192


The Core i7-4770K gives us an 8 MB shared L3 cache, similar to Core i7s before it. Although the Sandy and Ivy Bridge designs employed a single clock domain that kept the cores and L3 running at the same speed, Haswell decouples them. Our cache bandwidth benchmark reveals a slight hit to L3 throughput, though improvements elsewhere in the System Agent keep the results fairly even.

Haswell offers the same 16 lanes of PCI Express 3.0 connectivity as Ivy Bridge, and validated memory data rates up to 1,600 MT/s. The desktop line-up’s thermal targets are quite a bit different as a result of Intel’s fully-integrated voltage regulator, but an upper bound of 84 W isn’t extreme by any stretch and a floor of 35 W is pretty familiar.

All of Intel’s upgradable processors now drop into an LGA 1150 interface, meaning any decision to adopt Haswell is also going to require a motherboard purchase, at least. So, before you drop several hundred dollars on a brand new platform, let’s figure out if Core i7-4770K is worth the investment.

2. HD Graphics 4600: 3D And Quick Sync

Last month, Intel made a lot of noise about its new Iris Pro and Iris Graphics branding, warranted, it said, by a tremendous leap forward in performance. Core i7-4770K doesn’t get any of that. Instead, it features HD Graphics 4600, an evolution of Sandy Bridge’s HD Graphics 3000 and Ivy Bridge’s HD Graphics 4000.

Back when Intel introduced us to Sandy Bridge, fellow Tom Piazza described the work that went into modularizing different components of the graphics engine. In fact, in my Core i7-3770K coverage, I created the following numbered image to illustrate the company’s targeted approach to augmenting its partitioned design:

Here's the version that Tom used at IDF last year to illustrate Haswell. Note that there's a sixth domain, since the architecture has a video quality engine now.

Haswell maintains the same architectural partitioning, and adds more resources. Yes, there’s DirectX 11.1, OpenCL 1.2, and OpenGL 4.0 support, but performance is mostly affected by a shift from 16 to 20 programmable execution units in Haswell’s GT2 implementation. Across the next five pages, we’ll explore the impact of a more powerful graphics subsystem using average frame rates, frame rates over time, and frame time variance between consecutive frames.

The outcome, though, sounds a lot like what we said last year and the year before. Mainly, as it pertains to HD Graphics 4600, on-die graphics is fine for mainstream titles with light 3D workloads, but is quickly overwhelmed by common desktop resolutions in more taxing games. AMD isn’t much better off in this regard, but Intel still hasn’t caught up.

I’m at least happy to see the company using HD Graphics 4600 across its product line, where it previously armed lower-end chips with stripped-down graphics engines.

Improved Quick Sync

There’s quite a bit to discuss when it comes to Intel’s Quick Sync feature, which I introduced on this page in Intel’s Second-Gen Core CPUs: The Sandy Bridge Review. Intel followed up with an improved version of Quick Sync on Ivy Bridge (discussed here) that seemed to introduce mostly performance-oriented enhancements. Haswell’s implementation mixes in more speed and configurable quality dials.

For example, previous versions of Quick Sync exposed three pre-defined blends of performance and quality that Intel calls target usages. This time around, there are seven. Really, the intricacies deserve a story of their own. But at the highest-quality TU1 setting, HD Graphics 4600 is significantly better looking than 4000. Meanwhile, the fastest TU7 should be faster, higher-quality, and more battery-friendly for mobile devices on HD Graphics 4600 than 4000.

We did have a chance to run the latest beta of HandBrake, which is now available in Quick Sync- and OpenCL-optimized trim, on Intel’s Core i7s and AMD’s A10-5800K.

By no means is this meant as a slight to AMD. After all, the same task takes 226 seconds to run on the APU’s x86 cores alone, so there’s certainly an advantage to turning on OpenCL. However, Quick Sync drops a Core i7-4770K from 113 seconds (using the x86 cores-only) to 14. I had to ask Intel’s François Piednoël if there was any way this could be correct. Apparently, this is the expected behavior.

Each generation behind Haswell takes a second longer to finish the task. Just imagine if this were a full-length, Blu-ray-quality video, though.

3. HD Graphics 4600: Impressive OpenCL

These days, we think beyond 3D when someone starts talking about graphics processing. Heterogeneous computing is gaining traction, and there’s an increasingly large library of applications we’re able to test featuring OpenCL support.

Intel shipped its first OpenCL-capable drivers for the Sandy Bridge-era CPUs, though they only supported the processor. With Ivy Bridge, the company added HD Graphics 2500/4000 compatibility, allowing developers to leverage the x86 cores or symmetric execution units for general-purpose computing tasks. Haswell gets OpenCL 1.2 compliance (as does Ivy Bridge thanks to the latest driver package), along with performance improvements for OpenCL kernels running on the CPU and HD Graphics engine. What’s the result? There’s an app for that (several, actually)!

Each platform is running on integrated graphics, without interference from a discrete GPU.

Let’s start with Sony Vegas Pro 12. Because its graphics component isn’t supported, Intel’s Core i7-2700K sets our baseline with CPU-only results. Stepping up to the -3770K yields moderate gains based on architectural tweaks. But it’s not until we harness HD Graphics 4000 that the workload gets more than 50 seconds hacked out of it. Core i7-4770K furthers those gains with HD Graphics 4600.

This isn’t a story about AMD, but A10-5800K steals the show a bit by taking a relatively anemic dual-module processor and supercharging it with Radeon graphics, cutting the task by more than half. The Trinity-based APU is about as fast as Haswell without OpenCL enabled, though it’d be fairer to simply say it beats Core i7-2700K, since Sandy Bridge can’t benefit from OpenCL support.

You’ll see more of this benchmark later in the story when we drop a GeForce GTX Titan into each platform. Driven only by integrated graphics, though, Haswell smokes its predecessor through a combination of faster x86 cores and a larger graphics component. AMD’s A10-5800K finishes second, ahead of the much pricier Core i7-3770K.

Another popular general-purpose title, WinZip is mildly accelerated by OpenCL. Only certain files (those larger than 8 MB) benefit from heterogeneous computing, so the impact of OpenCL on a compression test is wholly workload-dependent. Nevertheless, we see another example of Haswell faring well against prior generations and AMD’s APU.

These thumbnails represent three mathematical models for estimating the future worth of options. I ran them all at FP32 precision. They universally show Core i7-4770K at the top of our four-contender stack, with A10-5800K and Core i7-3770K trading blows underneath.

Because SiSoftware’s Sandra 2013 lets me isolate CPU, GPU, and combined acceleration, we see that a CPU working alone in these calculations is really pretty slow. The HD Graphics engine is where it’s at. And, in some cases, adding capable x86 cores on top of that does help improve the outcome.

CPU performance makes the most significant difference in LuxMark, though combining the effects of both subsystems is pretty powerful, too. For the sake of comparison, if you have a single GeForce GTX Titan installed, you get about 1,300 K samples/sec. That’s only 60% or so faster than a Core i7-4770K and its on-die graphics.

It may not be leading the field in 3D performance, but Intel certainly deserves credit for its work with OpenCL.

4. HD Graphics 4600: Battlefield 3

At least for now, the highest-end graphics implementation available on an LGA 1150-based desktop processor is HD Graphics 4600. In Battlefield 3, we were able to benchmark it against AMD’s A10-5800K and Intel’s Core i7-2700K at 1280x720. The Core i7-3770K and its latest drivers crashed upon launching the game, and the -2700K output a garbled picture at 1920x1080.

At least based on the averages, you can actually play this game at fairly mainstream resolutions using the Low quality preset. AMD maintains a small advantage over Intel’s latest-generation effort, though we’re curious to see how consecutive frame latency plays out.

Although it tracks pretty closely to AMD’s Radeon HD 7660D, Intel’s HD Graphics 4600 engine dips under 30 FPS several times during our run at the one resolution we consider playable.

It doesn’t matter as much that Core i7-4770K falls to lower minimums at 1080p—both processors are too slow at this resolution anyway.

You’re not used to seeing AMD in this position, but at 720p, it averages a scant 1.3 ms of variance from one frame to the next—its pacing isn’t all that bad, actually. Meanwhile, Intel has some serious work to do with its driver. On average, variance sits around 8.5 ms. Using our 95th percentile results, though, it's as bad as 33 ms.

This only gets worse at 1080p, though the terrible frame rates keep either processor from playable performance.

5. HD Graphics 4600: BioShock Infinite

AMD gets the advantage again in BioShock Infinite. Only 1280x720 is playable at this title’s Low quality preset, and we see a nice progression down from Trinity to Haswell to Ivy Bridge to Sandy Bridge. Core i7-4770K does a respectable job narrowing the gap, but Intel just isn’t there yet.

A10-5800K dips under 40 FPS a few times, but manages a nice smooth run through BioShock’s built-in benchmark utility. In comparison, the Core i7-4770K flirts with 30 FPS on and off.

The fastest contender, AMD’s A10-5800K, dips under 20 FPS at 1080p. Everything else is slower than that, rendering this setting mostly unplayable.

Trinity and Haswell both deliver fairly even pacing through this test, though A10-5800K turns in notably better numbers.

The A10 doesn’t achieve great frame rates, but AMD manages to maintain consistent frame delivery using its Catalyst 13.6 Beta driver.

6. HD Graphics 4600: Hitman: Absolution

Another game, another win for AMD. This victory is far less decisive, though. We’re going to need to see the variance numbers to better-determine the solution with the smoother experience.

AMD’s A10-5800K keeps its nose just above 30 FPS, through most of the benchmark (even if it dips more than the Core i7-4770K at other points). Meanwhile, HD Graphics 4600 tanks at the very end of this test, hovering just over 20 FPS. That’s not conducive to a smooth experience.

Everything starts and stays under 30 FPS. Even at this game’s most entry-level detail settings, 1920x1080 isn’t in the cards for integrated graphics.

All three of the top processors exhibit worst-case variance between consecutive frames that we’d expect gamers to notice. With that said, Ivy Bridge and Haswell fare best, followed by AMD’s Trinity design.

7. HD Graphics 4600: The Elder Scrolls V: Skyrim

Although Skyrim is notoriously platform-bound, AMD’s A10-5800K establishes a commanding lead that makes the game enjoyable using Medium quality settings. We again see evidence that 1920x1080 is not particularly playable, though there are quality options you could sacrifice to bring performance up at that resolution.

A10-5800K kisses 40 FPS, but spends all of its time between 40 and 50 frames per second. Meanwhile, Core i7-4770K oscillates above and below the 30 FPS mark. Core i7-3770K and -2700K are simply not playable.

You might want to try turning the quality settings down to get Skyrim playable at 1920x1080…

Although AMD’s APU exhibits the best frame rates, the dual-module Piledriver-based CPU component is demonstrating that platform performance still does matter in this title by hitting the A10 with high variance between consecutive frames. Game play still feels fluid, fortunately.

Again, top frame rates are no guarantee of consistent frame delivery. AMD’s 95th percentile exceeds 32 ms. The jolt from one frame to the next is noticeable when it takes 32 ms longer.

8. HD Graphics 4600: World of Warcraft: Mists Of Pandaria

In a surprising turn of events, Intel’s Core i7-4770K claims its first and only gaming victory in World of Warcraft. Or maybe we shouldn’t be surprised. This is another notoriously processor-limited title with documented issues on AMD hardware. It looks like the A10 might still be playable at 720p. But it’s symbolic that Haswell, which we know only gets faster when we look at the BGA-based, Iris Pro Graphics 5200-equipped mobile version, already wins in this test with 20 execution units.

The frame rate over time is notably choppier in WoW than the other titles we’ve tested. With that said, Core i7-4770K spends most of its time up above 40 FPS, while A10-5800K hovers in the mid-30 range. Core i7-3770K dips into the mid-20s too often for our tastes.

Haswell again fares well, though playing through WoW at this setting is clearly too much for the HD Graphics 4600 engine to handle. Dialing down to Fair quality might help.

Relative to the other processors we’re testing, Core i7-4770K compares pretty well. In absolute terms, however, almost 14 ms of variance between consecutive frames is palpable. Fortunately, HD Graphics 4600 averages closer to 5 ms at 720p. The experience on AMD’s A10 is noticeably more jittery, particularly in our flight path-based benchmark where the animations are expected to be smooth.

Average and “worst-case” variances alike quickly balloon. This is actually a good representation of how frame time variance and the impact of inconsistent pacing manifest in a game. Naturally, when you’re looking at a tenth of a second difference between the time it takes to serve up one frame and the frame next to it, the effect is exaggerated. But there’s certainly no denying it’s there.

9. Intel 8-Series Chipsets: Z87 Is Nice

One of the brighter spots of Intel’s desktop Haswell introduction is its 8-series chipsets, including Z87, H87, H81, and B85 Express. Naturally, Z87 Express is the enthusiast-oriented platform controller hub that most of the motherboards we review will employ.

From my Haswell preview:

Eight-series chipsets are going to be physically smaller than their predecessors (23x22 millimeters on the desktop, rather than 27x27) with lower pin-counts. This is largely attributable to more capabilities integrated on the CPU itself. Previously, eight Flexible Display Interface lanes connected the processor and PCH. Although the processor die hosted an embedded DisplayPort controller, the VGA, LVDS, digital display interfaces, and audio were all down on the chipset. Now, the three digital ports are up in the processor, along with the audio and embedded DisplayPort. LVDS is gone altogether, as are six of the FDI lanes.

It turns out that one of the three digital ports is eDP-only with Panel Self-Refresh support, capable of cutting power consumption by putting the on-die GPU in a sleep state between frames. The second port can either be DisplayPort or HDMI, with daisy chaining enabled by DP 1.2 support. If you use a single screen, there’s enough bandwidth to support 4K output at 24 Hz. The third port is for DisplayPort, and that does resolutions of up to 3840x2160 at 60 Hz.

Moving down from the CPU to the less-complex Z87 PCH, we still get eight PCI Express 2.0 lanes for peripherals, along with an integrated gigabit Ethernet MAC and High-Def Audio. Support for Rapid Storage Technology (software-based RAID 1, 5, and 10), Smart Connect Technology (configurable wake from sleep to receive data like email from Outlook), and Rapid Start Technology (fast resume from hibernate) carry over from Z77 also.

The biggest changes are as many as six native USB 3.0 ports and six 6 Gb/s SATA ports, eradicating the four 3 Gb/s ports previously found on 7-series platforms. This is good. Given the proliferation of fast SSDs, we really needed more than two full-speed ports on some of our lab systems. And extra 5 Gb/s USB connectivity is welcome, too.

Should I Worry About My USB Flash Drive?

Prior to Haswell’s introduction, it was rumored that 8-series chipsets had a bug that’d cause USB 3.0-based thumb drives with certain controllers to disconnect when the platform woke from a sleep state. This turned out to be true, though the steps to reproduce actually had more to do with a pulse from the device greater than 400 mV.

Stepping C1 of the chipset is affected. Stepping C2, which should already be shipping, fixes it. Single-chip BGA-based Haswell implementations won’t exhibit the issue, as Intel intervened with updated chipset components on those soldered-down packages.

So far, there are no reports of data loss due to this, so it’s being labeled a nuisance. Our sources say a small number of drives trigger the bug, and if you find one that does, using a different thumb drive should be your solution. At the very worst, you may need to reconnect your device or restart your video player if you watching a movie from the drive when it disconnected. Given the list of scenarios where this errata might surface, and in light of the actions you’d need to take, it’s not worth factoring into a buying decision.

10. Overclocking Haswell: You’ll Pay For That

If you purchase one of the two K-series parts in Intel’s Haswell-based portfolio (and you have a Z87-based motherboard), you’ll have access to the same knobs and dials that were available from the Core i7-3770K and i5-3570K, with one notable change: the BCLK ratios popularized by X79 Express are back, facilitating some manipulation of the number that gets multiplied against the clock ratio. Intel exposes 1.0x, 1.25x, 1.67x, and 2.5x straps.

There’s a colossal difference between what Haswell-based LGA 1150 processors can do in theory, and what they can do after Intel’s product guys get done disabling the bits and pieces to create a differentiated stack, though.

Overclockers maintain control over core frequency using ratios up to 80x in 100 MHz increments. The on-die graphics engine is also adjustable in 50 MHz increments via ratios as high as 60x. Further, the memory controller is technically unlocked, allowing options for 200 and 266 MHz steps with a logical ceiling at 2,933 MT/s. Finally, the platform controller hub’s clock generator is unlocked too, supporting frequencies as high as 200 MHz.

But as with generations past, there remains a relationship between the DMI clock and PCI Express clocks—and you want PCI Express to stay as close to 100 MHz as possible. So, Intel implements 5:5, 5:4, 5:3, and 5:2 ratios to maintain constant PCIe clocks with a variable BCLK. Unlike the LGA 2011 platform, which used the CPU to adjust those ratios, Haswell requests this from the PCH. The outcome is similar, though. Expect a useable range between 5-7% around the ratio you choose for fine-tuning your overclock.

That’d all be well and good if Intel was enabling this new level of flexibility for the folks who don’t spend extra on K-series parts, giving them the ability to pick higher BCLK settings without access to the ratio multiplier. However, the company instead chooses to restrict the ratios to the Core i7-4770K and i5-4670K—the same ones you can already overclock in 100 MHz increments. Anyone buying one of the 11 other SKUs in Intel’s new Core i7 and i5 line-up is out of luck.

Overclocking Core i7-4770K

Moving on, what can you expect from a Core i7-4770K, in terms of overclocking headroom? We have a couple in the lab, and are getting 4.7 GHz, at most, across all cores using Prime95 to test for stability. However, those samples come from Intel. We were much more interested in feedback from someone with many, many retail parts at their disposal.

Our first-hand information involves a high double-digit number of processors, including samples and final shipping boxed CPUs. Sort testing was limited to 1.2 V to keep heat manageable. Ring/cache ratios are pegged at 3.9 GHz, with the memory controller operating at 1,333 MT/s. Of the chips available for sorting, only one is stable at 4.6 GHz under full load. A few are capable of operating at 4.5 GHz. More run stably at 4.4 GHz. Most are solid at 4.3 GHz and down. As you stretch above a 1,600 MT/s memory data rate or a ring ratio to match your highest single-core Turbo Boost ratio (which helps maximize performance), your top stable core frequency tends to drop.

Load-line calibration has very little, if any, effect with Haswell. There’s simply less flexibility to coax a killer overclock out of one of these CPUs using levers that were popular previously. Although the most enthusiast-friendly motherboards still include pages of options, a number of them simply don’t do anything beneficial. Flatly, it sounds like engineers and enthusiasts alike are still trying to figure out what the more obscure settings actually do.

We’re going to have to accept that Haswell-based parts get hotter, faster, it sounds like, and that they might fall a few hundred megahertz short of comparable Ivy Bridge-based parts with conventional cooling methods. It’s a good thing, then, that the architecture is inherently faster to help compensate.

11. Test Setup And Benchmarks
Test Hardware
ProcessorsIntel Core i7-4770K (Haswell) 3.5 GHz (35 * 100 MHz), LGA 1150, 8 MB Shared L3, Hyper-Threading enabled, Turbo Boost 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 i7-2700K (Sandy Bridge) 3.5 GHz (35 * 100 MHz), LGA 1155, 8 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

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

AMD A10-5800K (Trinity) 3.8 GHz (19 * 200 MHz), Socket FM2, 4 MB Total L2 Cache, Turbo Core enabled, Power-savings enabled
MotherboardMSI Z87 Mpower Max (LGA 1150) Intel Z87 Express, BIOS 1.2B1

MSI Z77 Mpower (LGA 1155) Intel Z77 Express, BIOS 17.8

MSI X79A-GD45 Plus (LGA 2011) Intel X79 Express, BIOS 17.2

MSI 990FXA-GD80 (Socket AM3+) AMD 990FX/SB950, BIOS 13.2

MSI FM2-A85XA-G65 (Socket FM2) AMD A85X, BIOS 2.0
Memory
G.Skill 16 GB (4 x 4 GB) DDR3-1600, F3-12800CL9Q2-32GBZL @ DDR3-1600 at 1.5 V
Hard Drive
Samsung 840 Pro 256 GB, SATA 6 Gb/s
Graphics
Nvidia GeForce GTX Titan 6 GB
Power Supply
Corsair AX860i, 80 PLUS Platinum, 860 W
System Software And Drivers
Operating System
Windows 8 Professional x64
DirectX
DirectX 11
Graphics DriverNvidia GeForce Release 320.18


As we were planning out our test platforms, MSI responded to a call for vendor consistency across the motherboards we wanted to use. The company sent us one board for each of the processor interfaces we planned to compare, making it easy to tune each firmware exactly the same way. In this case, we wanted all power-saving features turned on and all automatic overclocking capabilities turned off (including settings that pushed all cores to the maximum Turbo Boost setting).

The first platform we set up centered on the 990FXA-GD80 board and FX-8350 processor. Four 4 GB DDR3-1600 memory modules from G.Skill populated the motherboard's slots, while a Samsung 840 Pro attached to its first SATA 6Gb/s port.

Next, we configured MSI's FM2-A85XA-G65 using the same drive image, complementing the platform with AMD's highest-end A10-5800K APU. While a number of our earlier benchmarks exploited the chip's on-die Radeon HD 7660D, everything from here on out leverages an Nvidia GeForce GTX Titan.

With the AMD testing out of the way, we were able to start configuring Intel-based systems. We started with the Z87 Mpower Max, which glows yellow up top and white underneath. To be quite honest, I haven't been so impressed with an MSI motherboard in a long time. The company appears to have really stepped up its game for the 8-series generation.

The Z77 Mpower board serves our LGA 1155 needs, hosting Intel's Core i7-3770K and -2700K, yielding comparisons to two prior-generation architectures.

I want to point out something MSI did for us leading into this launch. Like pretty much every motherboard vendor out there, it has a setting in its BIOS called Enhanced Turbo. This sets all four cores to operate at the processor's maximum single-core Turbo Boost ratio as a sort of sneaky overclock. At our request, MSI turned this off by default with its 8-series boards so that Haswell-based CPUs operate according to Intel's specification.

We had to turn this quiet "cheat" off manually on the Z77 Mpower and Z79A-GD45 Plus, below. However, we appreciated that MSI was willing to get back to Intel's spec for this generation. We prefer control over this, and to not have it be a default setting.

Finally, theX79A-GD45 Plus hosted our venerable Core i7-3930K processor. Old though it might be, this is still one of our favorite CPUs around. And as you'll see, the old timer has no trouble beating Haswell down in threaded apps able to exploit its six cores.

Noctua sent over its LGA 1150-compatible NH-U14S to help with testing, too. We used an NF-F12 fan to blow over our memory modules. This seemed to help put a stop to the errors we were getting in our long, taxing Visual Studio compile workload.

Corsair also sent over some hardware for us to use. We leaned on its AX860i power supply for its promised compatibility with Haswell's low-power states. We also used its CMY16GX3M2A2400C10R memory kit for our overclocking efforts.

Benchmark Configuration
3D Games
Battlefield 3Campaign Mode, "Going Hunting" 90-Second Fraps, Low-Quality Preset, 1280x720 and 1920x1080
BioShock Infinite
Built-in Benchmark Utility, Fraps, Low-Quality Preset, 1280x720 and 1920x1080
Hitman: Absolution
Built-in Benchmark Utility, Fraps, Lowest-Quality Preset, 1280x720 and 1920x1080
The Elder Scrolls V: SkyrimUpdate 1.5.26, Celedon Aethirborn Level 6, 25-Second Fraps, Medium-Quality Preset, 1280x720 and 1920x1080
World of Warcraft: Mists of Pandaria
Flight Point Benchmark, Fraps, Good Quality Preset, 1280x720 and 1920x1080
Adobe Creative Suite
Adobe After Effects CS6Version 11.0.0.378 x64: Create Video which includes three Streams, 210 Frames, Render Multiple Frames Simultaneosly
Adobe Photoshop CS6Version 13 x64: Filter 15.7 MB TIF Image: Radial Blur, Shape Blur, Median, Polar Coordinates
Adobe Premeire Pro CS6Version 6.0.0.0, 6.61 GB MXF Project to H.264 to H.264 Blu-ray, Output 1920x1080, Maximum Quality
Audio/Video Encoding
iTunesVersion 10.4.1.10 x64: Audio CD (Terminator II SE), 53 minutes, default AAC format 
Lame MP3Version 3.98.3: Audio CD "Terminator II SE", 53 min, convert WAV to MP3 audio format, Command: -b 160 --nores (160 Kb/s)
HandBrake CLIVersion: 0.98: Video from Canon Eos 7D (1920x1080, 25 FPS) 1 Minutes 22 Seconds
Audio: PCM-S16, 48,000 Hz, Two-Channel, to Video: AVC1 Audio: AAC (High Profile)
TotalCode Studio 2.5Version: 2.5.0.10677: 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
Productivity
ABBYY FineReaderVersion 10.0.102.95: Read PDF save to Doc, Source: Political Economy (J. Broadhurst 1842) 111 Pages
Adobe Acrobat XVersion 10.0.0.396: Print PDF from 115 Page PowerPoint, 128-bit RC4 Encryption
Autodesk 3ds Max 2012 and 2013
Version 14.0 x64: Space Flyby Mentalray, 248 Frames, 1440x1080
BlenderVersion: 2.64a, Cycles Engine, Syntax blender -b thg.blend -f 1, 1920x1080, 8x Anti-Aliasing, Render THG.blend frame 1
Visual Studio 2010Version 10.0, Compile Google Chrome, Scripted
File Compression
WinZipVersion 17.0 Pro: THG-Workload (1.3 GB) to ZIP, command line switches "-a -ez -p -r"
WinRARVersion 4.2: THG-Workload (1.3 GB) to RAR, command line switches "winrar a -r -m3"
7-ZipVersion 9.28: THG-Workload (1.3 GB) to .7z, command line switches "a -t7z -r -m0=LZMA2 -mx=5"
Synthetic Benchmarks and Settings
3DMark 11Version: 1.0.1.0, Benchmark Only
PCMark 7Version: 1.0.4 x64, System, Productivity, Hard Disk Drive benchmarks
SiSoftware Sandra 2013Version 2013.01.19.11, CPU Test = CPU Arithmetic / Multimedia / Cryptography / Memory Bandwidth / Cache Bandwidth
12. Results: Synthetics

This one’s mainly interesting for the Physics benchmark, a pure measure of CPU performance. Core i7-3930K asserts itself early on, with quad-core Haswell, Ivy Bridge, and Sandy Bridge chips taking the next three spots. In the case of FX-8350 and, to a greater extent, A10-5800K, lower host processor performance brings down the Overall suite score, represented by the red bar.

Back when I published Core i7-4770K: Haswell's Performance, Previewed, a lot of folks (even some from within Intel) tried claiming our sample wasn’t representative of final performance. A quick comparison shows that, while we’re seeing very similar SSE3-based numbers, the Core i7-4770K’s advantage over -3770K is actually smaller in this latest version of Sandra, despite Haswell benefiting from AVX2 support.

Sandra’s Multimedia benchmark generates an image of the Mandelbrot Set fractal using 255 iterations for each pixel, representing vectorised code that runs as close to perfectly parallel as possible.

Our AVX2-based results from Core i7-4770K almost match the preview piece’s exactly, while the AVX-accelerated Ivy and Sandy Bridge numbers are close too. We now see that AVX2 helps a four-core Haswell part outperform Sandy Bridge-E’s six cores in AVX-optimized code.

When it comes to floating-point performance, the AVX code that runs on Core i7-3770K and -2700K matches pretty closely. But -4770K’s FMA3 path isn’t as consistent. Floats are faster, while doubles slow down quite a bit. The scaling is much closer to what we were expecting when we were testing for the preview.

Sandy Bridge-E takes off like a bat out of hell in the Encryption/Decryption module due to the extreme memory bandwidth you’ll see in the next chart, which feeds the CPU cores data quickly.

The low memory bandwidth issues affecting our early motherboard are mostly worked out, and we now see Core i7-4770K delivering competitive throughput in the AES-NI-accelerated workload, along with strong hashing performance.

These bars could have been predicted based on the AES256 results above, but here they are.

L1D, L2, and L3 cache bandwidth are all up on our Core i7-4770K compared to my preview piece. However, only the L1D yields the doubling of bandwidth we were expecting. Given 64 bytes/cycle (compared to 32 for Ivy Bridge), this number should still be much higher than it is. There’s still no good explanation for that outcome.

13. Results: Adobe CS6

We use two distinct Photoshop benchmarks, one of which fully taxes each processor’s x86 cores using well-threaded filters, and another that is OpenCL-optimized to leverage graphics resources. Don’t compare the black and red bars above—they’re only together to save space (and your scrolling finger).

Beginning with the CPU-only benchmark, according to which this chart is ordered, Intel’s Core i7-4770K slots in behind the Core i7-3930K and just ahead of the FX-8350. Though, the Core i7-3770K and -2700K are within two seconds of AMD’s flagship. In comparison, the company’s fastest A10-5800K trails far behind.

Calling this the OpenCL version of our benchmark is a little disingenuous, since we already ran our benchmarks based on each processor's on-die graphics engine using this same workload. Here, all platforms are accompanied by Nvidia’s GeForce GTX Titan, so the differences are wholly attributable to CPU interaction.

Now we see that the Core i7-4770K, -3770K, and 2700K take the top three spots, followed by Sandy Bridge-E. Perhaps it’s a utilization issue, but the FX-8350 and A10-5800K simply do not pair as well to a Titan in this test as Intel’s newer desktop architectures.

Our Premiere Pro CS6 benchmark is likewise optimized to take advantage of as many cores as we throw at it, and that’s why the almost 17-month-old Core i7-3930K maintains its first-place position. Yes, this is a $570 CPU, but there’s a reason we gave it our highest honor in Intel Core i7-3930K And Core i7-3820: Sandy Bridge-E, Cheaper.

Meanwhile, Core i7-4770K is only about 5% faster than Core i7-3770K. Hardly a reason to upgrade, taken on its own.

A first-place finish for Intel’s Core i7-4770K only puts it about six percent in front of its predecessor. Meanwhile, the Core i7-3930K we’d expect to be out front in a threaded title falls to third place. Consistently, this application demonstrates lower performance on platforms with higher thread counts if memory capacity doesn’t increase concurrently. Our 16 GB memory kit remains constant, helping explain why Sandy Bridge-E and AMD’s FX-8350 drop in the standings.

14. Results: Content Creation

We run a couple of different 3ds Max-based tests using this year’s and last year’s versions of the software. The outcome is pretty consistent from one to the other, though. Clearly-visualized utilization of all available physical and logical cores maps over to a result chart that rewards the CPUs armed with the most parallelized architectures. Core i7-3930K takes first place, with -4770K not far behind. The quad-module/octa-core FX-8350 takes third when we sort by 3ds Max 2013. However, it essentially ties Core i7-3770K. When you consider that’s a $330 Intel chip against a $200 AMD CPU, the Piledriver-based offering looks pretty good.

Our Blender workload favors Intel’s Core i7-3930K by more than 30 seconds compared to the second-place -4770K. Quad-core Ivy Bridge and Sandy Bridge processors file in behind, with AMD’s FX-8350 trailing closely.

Based on Maxon’s Cinema 4D software, our scripted Cinebench test measures single- and multi-core processor performance. We’re simply not concerned with OpenGL-based graphics frame rates in this piece.

Haswell turns in the fastest single-core time slip, just as we expected. Ivy Bridge is just behind, and both processors based on Intel’s Sandy Bridge architecture nearly tie. FX-8350 and A10-5800K are based on AMD’s Piledriver architecture, explaining why they’re so close, too.

Once we crank up the threading, Intel’s hexa-core Core i7-3930K screams into the lead, trailed by the Haswell-based -4770K and the Ivy Bridge-based -3770K. Core i7-2700K and FX-8350 nearly tie yet again.

15. Results: Productivity

Three generations of Intel CPUS are separated by roughly 10 percentage points, with AMD’s FX-8350 tying the Core i7-2700K at the slower end of that range. Given its ability to scale across at least 12 threads, FineReader puts Intel’s Core i7-3930K in first place.

In sharp contrast, printing a PowerPoint presentation to PDF only involves a single thread. So, the CPU with the highest clock rate and best IPC wins. That processor is Intel’s new Core i7-4770K, which outperforms the Core i7-3770K that, in turn, out-maneuvers the Core i7-2700K.

This is where AMD’s Piledriver architecture especially feels pain. Despite throttling up under the effects of Turbo Core technology, the FX-8350 and A10-5800K fall quite a bit behind.

Prior compile projects wrapped up a lot faster. So, you asked us for something more demanding. This Google Chrome workload is well-threaded, which is why the Core i7-3930K places first. Intel’s new Core i7-4770K isn’t far behind, though. Meanwhile, Core i7-3770K and -2700K maintain a sizeable advantage over AMD’s quad-module FX-8350.

Fritz isn’t really a productivity app (unless you consider playing chess productive), but we’re putting it here anyway. The results from each processor are reflected in kilonodes per second. A node is a position on the chessboard. So, in the case of Core i7-4770K, Fritz evaluates more than 14,000 thousand nodes per second, or 14+ million. If you give the engine enough time to “think”, you’re going to get a pretty competitive computer opponent. Hope you brought your A-game.

16. Results: Compression Apps

Once upon a time, we were able to get decent scaling out of WinRAR. Now the Core i7-3930K, -4770K, and -3770K all pile up on top of each other, trailed closely by the -2700K. FX-8350 and A10-5800K are the laggards.

7-Zip typically responds much better to the addition of cores, and we certainly see the Core i7-3930K’s six cores distance themselves from Core i7-2700K’s four, both chips  based on Sandy Bridge. Of course, Ivy Bridge and Haswell chip away at that advantage with greater IPC throughput. However, neither newer architecture is able to overcome the raw performance of Sandy Bridge-E. If you spent $500-something on a Core i7-3930K more than a year and a half ago, you’re still very happy with it today.

Our WinZip chart includes several results, since we first test using the CPU cores, and then follow that up by enabling OpenCL acceleration to offload some of the workload. Of course, we know from talks with Corel that the GPU only kicks in on files larger than 8 MB. Because our 1.3 GB archive is a mix of different sizes and types, only some of this benchmark is aided by turning on OpenCL.

The red CPU bar shows us that Core i7-4770K turns in the fastest result, outpacing the six-core -3930K (the assumption there is that WinZip doesn’t scale much beyond four cores, since the Sandy Bridge-E part also loses to Ivy and Sandy Bridge).

Adding OpenCL support brings down the time of all solutions by at least a few seconds, since we’re maintaining the same GeForce GTX Titan graphics card through our suite’s benchmark runs.

17. Results: Media Encoding

Once we get into the media-oriented apps, any relative weakness the six-core -3930K showed against newer quad-core CPUs should evaporate.

Indeed, the Sandy Bridge-E-based processor is the fastest in our TotalCode Studio benchmark, while the quad-core chips pile up on top of each other. Core i7-4770K is only marginally quicker than -3770K.

The Core i7-4770K comes close to catching the two generation-old Core i7-3930K, but falls just short. AMD’s FX-8350 manages to outperform the Core i7-2700K, approaching the -3770K’s performance level. In sharp contrast, the A10 trails way behind, serving as an example of what you don’t want from a media-oriented machine.

Great single-threaded performance makes it easy for the Haswell architecture to walk away with the win in iTunes. Not surprisingly, Ivy Bridge and Sandy Bridge trail in second and third position. Sandy Bridge-E, which sacrifices peak Turbo Boost clock rates in favor of a more complex six-core configuration, follows close behind.

Similar to iTunes, LAME is single-threaded, and partial to the processor with the fastest clock rate and highest IPC rate. No surprise—Haswell takes an easy first-place finish, trailed by Ivy Bridge and variations of Sandy Bridge.

18. Power Consumption

As an architecture, Haswell is being lauded for enabling the “biggest battery life increase in Intel’s history.” But of course, battery life isn’t a consideration on the desktop. Larger form factors that aren’t power-constrained give Intel more headroom to offer higher thermal ceilings (which is why we have Sandy Bridge-E-based chips rated for 130 W).

Core i7-4770K’s TDP is 84 W, 7 W higher than the 77 W Core i7-3770K. But it also employs on-package voltage regulation, moving components previously on the motherboard into the processor itself. How does this translate to efficiency?

First, let’s look at our complete benchmark run, graphed out over time.

We’re truncating the end of the chart for readability, but essentially, AMD’s A10-5800K trails off at the end, taking far longer to finish all of our workloads than any other processor. The big spikes are indicative of where 3DMark is running, so we can see Core i7-3930K got there first, followed by Core i7-4770K, with Core i7-3770K and -2700K not far behind.

Other landmarks are more difficult to identify, though we do see the FX-8350- and Core i7-3930K-powered systems averaging much higher draw from the wall than other platforms. The difference, of course, is that Sandy Bridge-E also finishes the suite notably faster than Piledriver.

The script we use for testing builds in short idle periods between benchmarks, which, more than anything, is necessary for slower systems so that they don’t fail taking too long to shut down or launch certain applications. Even still, if we averaged power consumption during the entirety of our run, you’d come away thinking that these platforms were hogs, since most of the time they’d be active. We add 1,800 seconds (30 minutes) of idle time to the back end before shutting each system off automatically.

With this idle time factored in, each GTX Titan-equipped platform draws what we’d consider to be a reasonable amount of power for high-end hardware performing a number of demanding tasks. The Core i7-3770 predictably achieves the lowest consumption result thanks to its 77 W TDP. Next is the 84 W Core i7-4770K, followed by Intel’s 95 W Core i7-2700K.

Because we also know that each sample represents two seconds, we also know exactly how long each system is powered on. So, we can take that figure, multiply it by the average power consumption, and get an idea of how much power was used during each run. Again, it appears that Intel’s Core i7-3770K leads the way, with -4770K behind.

Update (6/4): Special thanks to Tom's Hardware reader mikitd, who noticed the -4770K's Wh result seemed high. We revisited the logs, and found one test that was left enabled on the Haswell-based system, extending its run time while generating a result we weren't using. We've since re-run the test and now show Haswell and Ivy Bridge consuming roughly the same power. Although -4770K still averages higher consumption, its performance improvement helps compensate.

A look at the logs suggests that the Haswell-based chip isn’t idling as low as the -3770K, even though all power-savings features are enabled in both motherboard UEFIs.

19. Core i7-4770K: Did I Shave My Legs For This?

AMD introduced us to its Kabini and Temash SoCs one week ago. Naturally, we were excited to learn more about the Jaguar architecture, to see GCN rolled into a truly low-power configuration, and most of all, to get our hands on the devices AMD was promising. Tablets with big graphics performance. Convertibles that’d invoke Intel’s Ultrabook initiative, but better. Detachable form factors unlike anything ever seen with an AMD APU inside. Oh, we couldn’t wait.

And then we returned home with a reference-class laptop. It wasn’t even touch-enabled. As a performance demonstration, it worked well enough, but that was hardly what we were hoping for after all of the build-up. Frankly, we were disappointed.

A week later, Intel has a potentially great story to tell. Its Haswell architecture is expected to dramatically stretch out what you can get from a notebook battery. It’s going to drop into innovative products that fill a gap between tablets and notebooks. We’re expecting certain models to boast graphics performance to rival mid-range mobile GPUs. However, you don’t get a sense of any of that from Intel’s Core i7-4770K, the implementation of Haswell Intel chose to lead off with.

The Core i7-4770K, specifically, is a bit faster than the -3770K it replaces—but only because of IPC improvements. It runs at the same 3.5 GHz and sports the same four cores otherwise. HD Graphics 4600 are a small step up, but not significant enough to overtake AMD’s $130 A10-5800K APU in any meaningful way. The vaunted Iris Pro Graphics 5200, with eDRAM, is currently reserved for BGA-based SKUs. And although it appears we received fairly overclockable samples of the -4770K, industry consensus amongst the companies with hundreds of these chips on-hand is that, at safe input voltages, 4.3 or 4.4 GHz should be OK. The luckiest enthusiasts might get 4.5 or 4.6 GHz. Skill won’t get you far; Haswell is all about luck of the draw due to its integrated voltage regulator.

So, for the second time in a week, we’re disappointed. Haswell has a lot to offer, just not to desktop enthusiasts. Intel’s attention is fully in the mobile space, and we can tell.

Remember back to December of 2011, when we published Intel Core i7-3930K And Core i7-3820: Sandy Bridge-E, Cheaper? I gave the -3930K our Best of Tom’s Hardware award. Although the Sandy Bridge-E-based part was $600 at the time, power users who bought one have been enjoying it for the last year and a half—and, at its stock clock rate, it’s still faster than a Core i7-4770K in threaded workloads. That might have saved you a $300+ upgrade on Ivy Bridge and now a complete platform overhaul for Haswell.

For those of you on Core i7-2700K or older, Core i7-4770K makes sense as part of a two- or three-year upgrade cycle. Otherwise, I see little reason to spend money on a desktop processor upgrade, a new motherboard, and a compliant power supply. Save those few hundred dollars and put them toward a Haswell-based convertible, perhaps (or something based on Temash, if AMD’s partners can show us a compelling platform). In the meantime, we’ll be waiting on a manifestation of Haswell that more accurately shows off the spirit of Intel’s efforts.

For a chance at winning your own Core i7-4770K-based PC, please click this link to enter our CyberPower PC/Tom's Hardware sweepstakes. The system's specs are as follows:

You may enter the sweepstakes only one time. If you enter more than once, duplicate entries will be deleted. Entries from contest entry sites will be deleted.

The Sweepstakes opens on June 1, 2013 7:00 AM PDT and closes June 14, 2013 7:00 AM PDT.

1 Winner Will Be Chosen Randomly; the prize will be One (1) CyberPower PC as configured below; approximate retail value: $2,400.00.

  • Intel Core i7-4770K 3.50 GHz 8 MB Intel Smart Cache LGA-1150
  • Asus Z87-A w/ Intel Z87 Express Chipset Mainboard
  • AMD Radeon HD 7990 6 GB 16X PCIe 3.0 Dual GPU Video Card 
  • CoolIT LCS ECO II-120 mm FB Liquid Cooling 
  • Corsair 60 GB solid-state drive SATA 6Gb/s (Boot Drive)
  • Corsair Vengeance 16 GB DDR3-1600 Dual Channel Memory 
  • 1 TB SATA 6Gb/s 32 MB Cache 7,200 RPM HDD
  • LEPA G850-MAS 80 PLUS Gold-Certified Modular Power Supply 
  • Enermax Hoplite ST Mid-Tower Gaming Case w/ Hot-swap HDD Dock
  • LG 12x Internal Blu-ray Drive & DVD-RW, 3D Playback Combo Drive
  • Microsoft Windows 8 (64-bit Edition) 
  • Crysis 3 Game, BioShock Infinite Game, and Far Cry 3 Blood Dragon Game Coupon


DUE TO LEGAL REQUIREMENTS, THIS SWEEPSTAKES IS LIMITED TO LEGAL RESIDENTS OF THE USA (EXCLUDING RI) AND 18 YEARS OF AGE OR OLDER. UNLESS OTHERWISE NOTED, ALL PERSONAL INFORMATION WILL ONLY BE USED TO QUALIFY AND CONTACT THE WINNER.