What's new
  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

Intel i7-3770K Ivy Bridge CPU Review

Status
Not open for further replies.

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,086
Location
Montreal
We don’t think that it is an exaggeration to say that people have been talking about Ivy Bridge seemingly since the day that Sandy Bridge was launched back in January 2011. All this hype has been understandable though, for a number of reasons. This is Intel’s first new manufacturing process since January 2010, when they launched the innovative Clarkdale processors. Not only were these the first chips to be manufactured on the then cutting-edge 32nm process, but they were also the very first processors to integrate a GPU onto the CPU package. With Sandy Bridge, Intel took things a little further, sticking with the proven 32nm process, but continuing the march towards a true System-on-Chip (SoC) by integrating the GPU portion onto the CPU die itself, which was the logical and inevitable outcome.

Today, Intel is officially unveiling Ivy Bridge, and it is a significant launch. Not only is Ivy Bridge manufactured on the brand new 22nm manufacturing process, which is so technologically innovative that it became its own news story back in May of last year - surely you all heard about Tri-Gate or 3D transistors - but it also features some worthwhile microarchitectural changes. With AMD out of serious contention for the foreseeable future, Intel could have simply started manufacturing Sandy Bridge LGA1155 on the new 22nm process, produced more dies on each 300mm water, reaped in the profits and called it a day. After all, the company’s innovation strategy only called for a ‘tick’ step, which is supposed to merely be a shrinking of the previous microarchitecture. However, they instead decided to address the only part of Sandy Bridge that isn't essentially perfect - the integrated GPU. So while there is no great revolution on the CPU side of Ivy Bridge, Intel claims up to a 15% increase, the IGP has been heavily reworked and enhanced in every way.

The pinnacle of all that work is the chip that we are reviewing today, the flagship Core i7-3770K. This multiplier-unlocked part features a 3.5GHz default clock speed, Turbo capabilities up to 3.9GHz, and 8MB of L3 cache. So at first glance, this new processor doesn’t really distinguish itself from the Core i7-2700K, which has the same 3.5GHz default / 3.9GHz Turbo clock frequencies, and identical L1/L2/L3 cache sizes. However, the slight performance tweaks that Intel have made to IVB's cores can't really be illustrated numerically. Since we really did not have any complaints about any areas of SB's CPU performance to begin with, we we can appreciate the fact that they mostly focused on lowering power consumption, as evidenced by the 77W TDP. This is a significant 23% drop when compared to the 95W quad-core Sandy Bridge parts, and should set a new standard for performance per watt. The default memory speed has been bumped up to DDR3-1600, just like on Sandy Bridge-E, so that already has the potential to provide a nice performance boost, especially on the graphics side.

Speaking of which, the new top-end HD Graphics 4000 IGP has 16 Execution Units (EUs) and a maximum frequency of 1.15GHz. Now that might not sound like much of an improvement when compared to the previous HD Graphics Card 3000 ( 12 EUs – up to 1.35GHz), but these new EU's are about twice as powerful as their predecessors, and altogether Intel is claiming an up to 60% increase in GPU performance. The new Ivy Bridge GPU brings forth compatibility with DirectX 11, OpenCL 1.1, OpenGL 3.1, improved Quick Sync Video performance, and support for three display outputs. Whether it is on-par with AMD have achieved with their Llano A-series APUs is what we are here to find out.


IvyBridge_3770K_1.jpg
 

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,086
Location
Montreal
Ivy Bridge: Intel Core i7-3770K

Ivy Bridge: Intel Core i7-3770K


IvyBridge_3770K_2th.jpg

Ivy Bridge / Sandy Bridge E / Sandy Bridge / Lynnfield / Gulftown - Click on image to enlarge

Initially, Intel presented us with an Ivy Bridge lineup that consisted of 7 regular desktop models and 7 low-wattage variants. However, a few weeks ago that lineup was reduced 5 and 4, respectively. We are assuming that Intel is merely delaying the launch of these now absent models, and that they will be introduced sometime in the future. So far we have gotten zero hints with regard to any Core i3 models, but we have to assume that dual-core/four-threads Ivy Bridge chips are also going to make their appearance sooner rather than later.

With respect to model names, we can't help but feel that Intel are not giving themselves much room to maneuver with Ivy Bridge LGA1155. While they might one day introduce an i7-3780K or 3790K part, they couldn't really go above that, lest it eclipse the "higher-end" Core i7-3820. What we are trying to say is that what you see is what you get with Ivy Bridge on the LGA1155 platform, there will not be much of an upgrade path. Haswell is the future for Intel, which is Socket LGA1150.

IvyBridge_3770K_3.jpg

By transitioning to the new 22nm manufacturing processor, Intel has managed to increase the transistor count from 1.16 billion on Sandy Bridge up to 1.4 billion Ivy Bridge. This is about a 21% increase, which is not really that much, but what is truly impressive is that the die size has from shrunk 216mm2 down to 160mm2. That is a noteworthy 35% decrease, and as you will see in the later parts of the review that tiny die size caused some problems. Keep in mind that these are the figures for the full-blown quad-core + HD Graphics 4000 die, we fully expect that the dual-core and/or HD Graphics 2500 parts will have a reduced transistor count and die size as what the case with Sandy Bridge.

Despite AMD's Bulldozer being a bit of a flop, Intel is not really trying to take advantage of the situation, and the new Ivy Bridges processors are actually $3-4 cheaper than the equally positioned Sandy Bridge parts were at launch. Consumers are going to have to pay a premium for graphics power though, since the more powerful of the two IGP variants is reserved for the higher-end parts, which is a shame since it's those who buy the more reasonably priced parts that are most likely to forgo a discrete GPU and thus need the most powerful IGP possible. We strongly suspect that Intel will eventually bundle the HD Graphics 4000 with the mainstream processors sometime down the road, just like they did Sandy Bridge and the Core i3-21x5 parts.

As we mentioned in the introduction, the flagship of the new lineup is the Core i7-3770K. It is a K-series model, which means that it has unlocked multipliers, and it features a 3.5GHz default clock speed, Turbo capabilities up to 3.9GHz, and 8MB of L3 cache. All these figures are essentially identical to the Core i7-2700K, which has the same 3.5GHz default / 3.9GHz Turbo clock frequencies, and identical L1/L2/L3 cache sizes. Although the basic cache sizes have remained the same since the Nehalem architecture, the increase in the cache operating frequency and associativity has resulted in much higher bandwidth and lower latencies on Ivy Bridge. Another performance advantage in IVB’s favour is the new DDR3-1600 memory interface, which can provide up 25.6GB/s of memory bandwidth, up from 21.3GB/s on Sandy Bridge. Much like Sandy Bridge-E, the integrated PCI-E controller now supports PCI-E 3.0, but Ivy Bridge chips still only have 16 dedicated lanes for the graphics slots.

The whole lineup features a 77W TDP, which is a very worthwhile reduction when compared to the previous 95W figure that was the standard for most of Intel's recent quad-core mainstream parts. This lower TDP is double edge sword though, at least for performance enthusiasts. On the one hand, it allows for lower power consumption and (theoretically) cooler operating temperatures, but on the other hand it definitely limits the potential core clocks. Releasing a next generation flagship part at the same clock speed as the previous generation's top-end model is slightly underwhelming, especially given the minimal core improvements and the move to a cutting-edge manufacturing process. As we explained at the top of the page, we also aren't really confident that Intel will be releasing much higher clocked models, although that is just speculation on our behalf.

By the way, those of you who put a heavy emphasis on lower consumption or simply want to build a cool-running but very powerful HTPC will be glad to know that Intel is also releasing a couple of low wattage variants too.


IvyBridge_3770K_5th.gif

Click on image to enlarge

There is nothing fundamentally new on the packaging front, Ivy Bridge processors will ship in exactly the same box as Sandy Bridge models. Although Intel did not provide us with a stock cooler, we suspect that they will look this this.

IvyBridge_3770K_6th.gif
IvyBridge_3770K_7th.jpg

Click on image to enlarge

As you might have expected Ivy Bridge LGA1155 processors look exactly like the Sandy Bridge one's did, and are thus also almost indistinguishable from the previous LGA1156 Clarkdale and Lynnfield models.

Based on the digits on the integrated heatspreaders, we can determine that our chip was made in the 6th week of 2012. That is much 'fresher' silicon then we are used to seeing from Intel, which might support the argument that they encountered a few bumps along the way when they started manufacturing on the new 22nm process.


IvyBridge_3770K_8th.gif
IvyBridge_3770K_9th.jpg

Click on image to enlarge

On the voltage front we were very surprised to see that idle voltage was higher than what we have seen on both Sandy Bridge and Sandy Bridge-E. As you will see in the power consumption section, this ultimately proved to be irrelevant, and more importantly the full load voltage was about 0.10V lower.

The bus speed is still 100MHz, and as on Sandy Bridge there is not much overclocking headroom, only about 7-8%. This is because so many of the CPU's different parts are deriving their operating frequencies from this base clock, and since some are very sensitive to frequency changes, they can get out of whack very quickly.

As mentioned above, the cache structure is the same as on Sandy Bridge, although as you will see in the benchmarks it's now even faster and with lower latencies.
 

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,086
Location
Montreal
Tick+ The Ivy Bridge Microarchitecture & Process

Tick+ Ivy Bridge Microarchitecture & Process



IvyBridge_3770K_12.jpg

As mentioned in the introduction, Intel’s development schedule called a ‘Tick’ in 2012, which is generally just a process shrink of the previous microarchitecture, but thankfully Intel haven't been resting on their laurels. At IDF 2011, Intel referred to Ivy Bridge as a Tick+ due to the significant enhancements in the graphics and media portion of the chip, as well as the numerous little new architectural features, microarchitectural improvements, and power savings.

Having said that, the highlight of Ivy Bridge is obviously the new cutting-edge 22nm manufacturing process, and the revolutionary 3-D or Tri-Gate transistor technology that got a lot of publicity a few months ago. In development since first announced in 2002, the characteristics of tri-gate transistors is that they utilize a fin-based, multi-gate transistor design. Instead of a having a flat two-dimensional stream like on 2D planar transistors, the conducting channel is wrapped in three-dimensional fins. This allows for control of the current on all three sides of the fin, instead of just on one side. The end result is a significant reduction in leakage and power consumption, while simultaneously allowing for much higher switching speeds. If you want an visual explanation in layman’s terms, Intel has a neat little video with Mark T. Bohr, a Senior Fellow and Director of the Manufacturing Group.

IvyBridge_3770K_11th.jpg

Click on image to enlarge

The end result of all that advanced manufacturing technology is obviously the Ivy Bridge die itself. As mentioned of the previous page, by transitioning to the new 22nm manufacturing processor, Intel has managed to increase the transistor count from 1.16 billion on Sandy Bridge up to 1.4 billion Ivy Bridge, while simultaneously shrinking the die size has from 216mm2 down to 160mm2.

While the bulk of additional transistors are due to the larger and more powerful IGP, there are a number of other small changes as well.

IvyBridge_3770K_13.jpg


As usual, Intel have been relatively secretive with respect to what tweaks they have done to the cores. However, while we do know that most elements of Ivy Bridge have remained the same as on Sandy Bridge, but they have cleaned up the design a bit to further improve single threaded performance. The floating point unit was given some attention and now throughout was doubled, which is an improvement that should reveal itself in computationally demanding workloads.

Graphics aside, the new parts come in the form of enhancements to security, power management, and memory support. The security enhancements are the addition of Digital Random Number Generator (DRNG) and Supervisory Mode Execute Protection (SMEP). These are hardware-level protection mechanisms that give us a glimpse into what Intel's plans are now that it owns McAfee.

The power management improvements are numerous, but the only one really stands out. Although most areas of the processor can now be power gated, Intel addressed the one area that wasn't, the DDR3 memory interface. Although not a large source of current leak, the DDR3 interface can now be totally shutdown if there's no memory activity.

The memory controller remains basically unchanged, but it does now natively support DDR3-1600 as the default. In dual-channel form that means up to 25.6GB/s of memory bandwidth, up from 21.3GB/s on Sandy Bridge. The maximum supported DDR3 frequency has also increased, from DDR3-2133 on Sandy Bridge to DDR3-2800. Intel have also tweaked the multipliers, so memory frequency can be increased in smaller 200MHz steps (instead of just 266MHz). Although more relevant for the mobile sector, support for 1.35V DDR3L has been added in order to further help reduce system power consumption.

In order to improve overclocking, have implemented real-time multiplier adjustments, thereby eliminating the need to reboot when increasing/decreasing multipliers from within the OS. More importantly they have also increased the CPU multiplier limits from 57X to 63X. This should help alleviate one of the significant overclocking bottlenecks that was discovered on Sandy Bridge.

The integrated PCI-E controller has obviously been updated. It now supports PCI-E 3.0 while also featuring 20 channels. Regrettably, only 16 are enabled on the consumer desktop side. You will need a workstation-class C200 series chipset and Xeon processor to get access to the full 20 lanes. Nevertheless, this new interface operates at 8.0 GT/s, and with 1GB/s of bandwidth per lane, allows for up 32GB/s of aggregate bandwidth to one PCI-E x16 slot.

IvyBridge_3770K_20.jpg


The new IGP is clearly a huge part of Ivy Bridge, both literally and figuratively. The HD Graphics 4000 takes up almost half the die on the Core i7-3770K. This new top-end IGP has 16 Execution Units (EUs) and a maximum frequency of 1.15GHz. Now that might not sound like much of an improvement when compared to the previous HD Graphics Card 3000 ( 12 EUs – up to 1.35GHz), but these new EU's are about twice as powerful as their predecessors, and altogether Intel is claiming an up to 60% increase in GPU performance. To ensure an optimal gaming experience, Intel have also added compatibility with DirectX 11, OpenCL 1.1, OpenGL 3.1, and support for three display outputs. To cap it all off Intel has further improved the performance of their unmatched hardware video encoding/decoding Quick Sync Video technology, and improved media CODEC support and image quality.
 

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,086
Location
Montreal
An In-Depth Look at Intel's HD Graphics 2500 & 4000 IGPs

An In-Depth Look at Intel's HD Graphics 2500 & 4000 IGPs


While we were genuinely impressed with the progress Intel made with Sandy Bridge’s IGP, six months later it got blown out of the water in grandiose fashion by the significantly more powerful Radeon HD 6550D found in AMD’s Llano A-series APUs.

With Ivy Bridge, Intel is trying to regain some of that lost ground, and they are touting the HD Graphics 4000 as being capable of a twofold increase in GPU performance over the previous HD Graphics 3000. They have accomplished this in a number of different ways but as before, their processor graphics is still broken into two categories called GT1 (HD2500) and GT2 (HD4000). GT1 is once again geared towards the lower end of the spectrum while GT2 will typically be attached to higher level solutions.

IvyBridge_3770K_24th.jpg

Click on image to enlarge

When compared to the units contained within Sandy Bridge processors there have been plenty of architectural changes to Intel’s integrated graphics cores this time around. The Processor Graphics Unit is now broken up into three distinct graphics processing stages: the Global Assets containing the fixed function stages along with the Geometry engines, the Slice Common with its Rasterizer, L3 cache setup and pixel back ends and finally the main Slice unit which houses the Execution Units, L1 cache and other rendering pipeline necessities. Separate units have also been included for the Media CODECs and necessary display output features.

As with Sandy Bridge’s architecture, the Execution Units still do the lion’s share of heavy lifting in this core design. Much like NVIDIA’s cores or AMD’s shaders, they are responsible for the day to day multistage processing for both graphics and compute workloads. However, Intel has now added support for Compute Shaders so high levels of parallelism are now possible and shared local memory has been added to increase the performance of Direct Compute applications. As necessitated by the addition of DX11, the architecture also supports Shader Model 5.0.

Speaking of the switch to DX11 compatibility, it has necessitated the modification of the primary rendering stages. A dedicated tessellation unit as well as a pair of programmable stages –the Hull Shader and Domain Shader- has been thrown into the equation. In order to further aid DX11 performance, the architecture now supports BC6H/7 compressed texture formats as well.

While Intel have made plenty of sizeable microarchitectural enhancements to the graphics processor, what’s really interesting is that they have given the IGP its own L3 cache. While the Last Level Cache (LLC) is still shared between CPU and IGP, this small cache has been integrated into the graphics core and slightly reduces the need for the IGP to use power-hungry 256-bit ring bus interface that connects all the elements of the chip. This change, along with the lower GPU frequency and voltage, and of course the switch to the 22nm process has allowed Intel to double GPU’s performance per watt.

IvyBridge_3770K_21th.jpg

Along with the architectural improvements that may not be apparent by looking at the on-paper specifications of Ivy Bridge’s Processor Graphics, the HD4000 series now includes 16 Execution Units, an improvement over the 12 within Sandy Bridge’s higher end layout, resulting in a twofold improvement in certain cases. The HD2500 maintains the six EUs of the previous generation but with the wide range of on-die changes it should still offer a performance bump of between 10-20% in certain graphics intensive workloads. Quick Sync video transcoding and other GPGPU intensive tasks will also see a significant across the board improvement with these new PGUs, regardless of the clock speed differences.

The new HD graphics architecture isn’t completely focused upon offering a preset specification layout either. It is able to easily scale upwards or downwards, creating a nearly infinite list of derivatives. We likely won’t see any of these offshoots in this generation but expect higher performance from an expanded layout when Haswell hits sometime in 2013.

IvyBridge_3770K_23.jpg

The HD Graphics on Ivy Bridge can dynamically adjust its frequency in order to automatically increase the clock speeds of the graphics controller when higher loads are detected. Much like the Turbo Boost technology on the CPU itself, this acts as a way to conserve power when high speeds aren’t needed and yet allows for on-the-call performance in demanding situations. And as you will see in our IGP gaming benchmarks section, the HD Graphics 4000 needs every bit of extra performance to compete with the Llano APUs.
 

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,086
Location
Montreal
Diving into Intel's Z77 Express Chipset

Diving into Intel's Z77 Express Chipset


A few weeks ago, in preparation for the Ivy Bridge launch, Intel’s partners introduced a lineup of supporting motherboards with the new Panther Point platform controller hub. When combined with an Ivy Bridge processor, this PCH allows for a wide variety of connectivity options but the general layout hasn’t changed all that much from the Z6x-series. For clarity’s sake, we’ll be looking at the Z77 layout here since it represents the Ivy Bridge platform in its highest end form.

IvyBridge_3770K_36.jpg

While the basic layout may be the same here, several new functionalities have been added. Starting with the CPU itself, it is now paired up with the memory at speeds of up to DDR3-1600, an improvement over Sandy Bridge’s 1333MHz.

While the processor still acts as a controller for the main external graphics interface, it now boasts up to 20 native PCI-E 3.0 lanes, though only 16 can be enabled on consumer chipsets. Back in the heyday of Z68, several motherboard vendors included PCI-E 3.0 support with their products but this was typically done through inefficient means and without the official support from Intel and the PCI-SIG association. Native support brings heightened interface protocol speeds and the lanes can be configured in a number of different ways. Either one external graphics card can communicate with the CPU through 16 dedicated lanes or they can be split into a pair of Gen 3 x8 interfaces (each with the bandwidth of a single Gen 2 x16 slot) for SLI or Crossfire support. Alternately, eight of the lanes can be directed towards other purposes like providing bandwidth to a secondary Thunderbolt controller but this will also sacrifice multi GPU compatibility unless a PLX switch is used.

As with Z68, the Processor Graphics communicates with and ultimately outputs its display signals to the PCH via the FDI or Flexible Display Interface. This runs in parallel with the link between the CPU and the PCH called the DMI interface which features four lanes in each direction that can operate at speeds of up to 2 GB/s. This results in 4 GB/s of aggregate bandwidth if both upstream and downstream lanes are used to their theoretical maximum.

IvyBridge_3770K_37.jpg

The Z77 Express PCH act as a hub for all of the system’s I/O needs and is based off of Intel’s 65nm manufacturing process. Among other things there is a very important addition here: native support for up to four USB 3.0 port so third party controllers are no longer necessary for this functionality. And additional ten USB 2.0 ports are also available.

The PCH also houses Intel’s High Definition Audio output for the onboard GPU alongside an additional eight PCI-E 2.0 lanes for any periphery add in boards or secondary system controllers. Like the processor itself, a number of these PCI-E 2.0 lanes can be used to power a Thunderbolt controller.

As with Z68 chipsets, native support for up to six SATA 6Gbps ports has been included and all are capable of running Intel’s Smart Response Technology for decreased system load times.


IvyBridge_3770K_39th.jpg


Another interesting feature is the PCH’s ability to output up to three discrete display signals from the Processor Graphics engine. These signals are split towards three distinct ports, each controlling a HDMI, DisplayPort or DVI connector, giving the end user multiple connectivity options.

IvyBridge_3770K_40th.jpg

Click on image to enlarge

The Z7x series consumer motherboard lineup will consist of the usual combination of mainstream and performance oriented boards, headlined by the Z77 we’ve been talking about. The Z75 is a nearly identical sibling to the Z77 with the only major change being the elimination of some PCI-E lane adaptability for secondary controllers. We’re guessing that the vast majority of Intel’s motherboard partners will likely avoid this chipset since there just isn’t enough to differentiate it from the Z77 Express.

Intel’s H77 meanwhile doesn’t include the broad feature sets and performance tuning abilities of the Z-series since it is generally geared towards the entry level market. The ability to run two graphics cards in SLI or Crossfire has also been nixed. We expect to see this chipset as the backbone of many mATX and ITX boards along with sub-$100 ATX products. The other Panther Point chipsets are marketed towards the business market so we won’t cover them here.
 

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,086
Location
Montreal
Clock-per-clock: Ivy vs. SB-E vs. SB

Clock-per-clock: Ivy vs. SB-E vs. SB


Now as we have already discussed, although Intel didn't really do anything drastic to the CPU portion of Ivy Bridge they did do a few small obscure tweaks to the cores, so we are interested in finding out the effect of these changes on the instructions per clock (IPC) and how IVB compares on a clock-per-clock basis with Sandy Bridge and Sandy Bridge-E.


IvyBridge_3770K_31.gif

3.5GHz - Ivy Bridge vs. Sandy Bridge-E vs. Sandy Bridge

Now as you can see we clocked all three chips to as close to 3.50GHz as possible and we set identical memory frequencies (DDR3-1600) and timings (8-8-8) for all, although the i7-3820 does retain its quad-channel interface (not really an advantage as we have proven before). We selected a nice mix of applications in order to get a good idea of single and multi-threaded performance.

IvyBridge_3770K_32.jpg

As you can see, it's good that we kept our expectations low. On average, Ivy Bridge is about 4% faster than the two Sandy Bridge variants on a clock-per-clock basis. Neither the single or multi-threaded performance really stands out, both are about equally improved. 3D modelling and media creation are the areas that see the biggest boost, and although not demonstrated above minimum frame rates while gaming are consistently higher thanks to IVB's super-low latency L3 cache.

Those who were hoping for a giant leap forward performance-wise are obviously going to be disappointed, but keep in mind that Intel has achieved this level of performance with a much smaller TDP than on Sandy Bridge (77W vs. 95W/130W), so it is an impressive feat.

Don't let this clock-per-clock comparison turn you off Ivy Bridge though, because as you will on the following page, one of IVB's strengths is its Turbo Boost implementation.
 

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,086
Location
Montreal
Feature Test: Turbo Boost 2.0

Feature Test: Turbo Boost 2.0



For those of you who aren't familiar with it, let's recap what Turbo Boost is and what it does. Turbo Mode is a performance feature that automatically unlocks additional speed bins (multipliers) and allows the processor to self-overclock based on thermal conditions and workload. For example, if the Power Control Unit (PCU) senses that only one core is active and the other three are in an idle state, it will use the unused power and thermal headroom to overclock that single active core to ensure superior single-threaded performance. Conversely, if you are running a multi-threaded application, the PCU will measure the thermal headroom and if the processor is running cool enough it will overclock all six cores. On Ivy Bridge processors, or at least the i7-3770K, Turbo can provide a 400Mhz speed boost in single and dual-threaded workloads, 300Mhz in triple-threaded workloads, and 300Mhz in applications that utilize four threads or more. Ivy Bridge also engages the Turbo Boost modes faster, and holds onto it longer, which also helps improve performance.


IvyBridge_3770K_33th.jpg
IvyBridge_3770K_34th.gif

Turbo Boost Off - Click on image to enlarge - Turbo Boost On


As we mentioned previously, with this implementation of Turbo Boost we never ever saw the default 3.5GHz clock speed, no matter how fully the cores were loaded. As a result, with Turbo Boost enabled owners of the i7-3770K can realistically consider 3.7GHz to be the stock frequency. It is basically a free and automatic 6% overclock, not too shabby. In order to unlock this extra performance, Intel's allows the chip to exceed its 77W TDP if there is enough thermal headroom. So if it's running cool enough, the i7-3770K will operate in burst mode at up to 95W for one second.

Let's check out the performance gains that Turbo Boost can provide on this flagship part. We selected a nice mix of benchmarks with both light and multi-threaded workloads.


IvyBridge_3770K_35.jpg


Because of the more aggressive Turbo Boost implementation, the performance improvements in multi-threaded workloads can be twice as high as what we recorded on Sandy Bridge. The single-threaded performance gain is about the same as on Sandy Bridge, which is to say a solid 9-11% boost.
 

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,086
Location
Montreal
Test Setups & Methodology

Test Setups & Methodology


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

Intel Core i7 LGA1155 Ivy Bridge Test Setup​
IvyBridge_3770K_46.jpg

Intel Core i7 LGA2011 Test Setup​
IvyBridge_3770K_47.jpg

Intel Core i5/i7 LGA1155 Test Setup​
IvyBridge_3770K_48.jpg

Intel Core i3/i5/i7 LGA1156 Test Setup​
IvyBridge_3770K_49.jpg

Intel Core i7 LGA1366 Test Setup​
IvyBridge_3770K_50.jpg

AMD Llano FM1 Test Setup​
IvyBridge_3770K_51.jpg

AMD Zambezi AM3+ Test Setup​
IvyBridge_3770K_52.jpg

AMD Phenom II AM3 Test Setup​
IvyBridge_3770K_53.jpg


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

A) Windows is installed using a full format.

B) Chipset drivers and accessory hardware drivers (audio, network, GPU) are installed.

C)To ensure consistent results, a few tweaks are applied to Windows 7 and the NVIDIA control panel:
  • UAC – Disabled
  • Indexing – Disabled
  • Superfetch – Disabled
  • System Protection/Restore – Disabled
  • Problem & Error Reporting – Disabled
  • Remote Desktop/Assistance - Disabled
  • Windows Security Center Alerts – Disabled
  • Windows Defender – Disabled
  • Screensaver – Disabled
  • Power Plan – High Performance
  • V-Sync – Off

D) Windows updates are then completed installing all available updates

E) All programs are installed and then updated, followed by a defragment.

F) Benchmarks are each run three to eight times, and unless otherwise stated, the results are then averaged..

Here is a full list of the applications that we utilized in our benchmarking suite:
  • 3DMark Vantage Professional Edition v1.1.0
  • 3DMark11 Professional Edition v1.0.2
  • 7-Zip 9.22 beta 64-bit
  • AIDA64 Extreme Edition v1.85.1641 Beta / v2.00.1719 Beta / v2.30.1900 Beta
  • Cinebench R10 64-bit
  • Cinebench R11.529 64-bit
  • Civilization V 1.0.1.383
  • Crysis v1.2.1 64-bit
  • Crysis 2 v1.9 + DX11 Pack + HiRes Texture Pack
  • Deep-Fritz 12
  • DiRT 3 v1.2.0
  • Far Cry 2 v1.03
  • HyperPI 0.99b
  • Lame Front-End 1.0 (LAME 3.97 32-bit codec)
  • Left 4 Dead 2 v2.0.8.9
  • LuxMark v1.0
  • MaxxMEM² - PreView v1.90
  • PCMark 7 Professional Edition v1.0.4
  • Photoshop CS4 64-bit
  • POV-Ray v3.7 RC3 64-bit
  • Street Fighter IV Benchmark V1.0.0.1
  • Team Fortress 2 v1.1.7.6
  • TrueCrypt 7.1
  • Valve Particle Simulation Benchmark v1.0.0.0
  • WinRAR 4.0.1 64-bit
  • World in Conflict Demo v1.0.0.0
  • wPRIME version 2.05
  • x264 HD Benchmark 4.0
  • X3: Terran Conflict Demo v1.0

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

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,086
Location
Montreal
Synthetic Benchmarks: AIDA64 / MaxxMEM² / SiSoft

Synthetic Benchmarks: AIDA64 / MaxxMEM² / SiSoft




AIDA64 Extreme Edition 1.85 - CPU & FPU Benchmarks


IvyBridge_3770K_54.jpg
IvyBridge_3770K_55.jpg

IvyBridge_3770K_56.jpg


AIDA64 Extreme Edition 1.85 - Cache Benchmark


IvyBridge_3770K_57.jpg


AIDA64 Extreme Edition 1.85 - Memory Benchmarks


IvyBridge_3770K_58.jpg

IvyBridge_3770K_59.jpg


MaxxMEM² - Memory Benchmarks


IvyBridge_3770K_60.jpg

IvyBridge_3770K_61.jpg


Sisoft Sandra 2011.SP5 - Memory Benchmarks


IvyBridge_3770K_62.jpg

IvyBridge_3770K_63.jpg


Sisoft Sandra 2011.SP5 - Cache Benchmarks


IvyBridge_3770K_65.jpg

 

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,086
Location
Montreal
Synthetic Benchmarks: SuperPI 32M / wPRIME 1024M

Synthetic Benchmarks: SuperPI 32M / wPRIME 1024M



SuperPi Mod v1.5


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

IvyBridge_3770K_66.jpg


wPRIME 2.03


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

IvyBridge_3770K_67.jpg
 
Status
Not open for further replies.

Latest posts

Top