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 Core i7 "Nehalem" 920, 940 & 965 XE Processor Review

Status
Not open for further replies.

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,106
Location
Montreal
Feature Test: Memory Scaling

Feature Test: Memory Scaling


Back in mid-2003, Intel introduced the 875P 'Canterwood' chipset, which featured the company's first consumer-oriented implementation of the dual-channel memory interface. This configuration has successfully lasted us until the present day. However, with Nehalem Intel have kicked things up a notch, unveiling a triple-channel memory configuration. Teamed up with the new integrated DDR3 memory controller, we have very high expectations for this new combo.

<table align="center" table border="0" bgcolor="#666666" cellpadding="5" cellspacing="1" width="735px"><tr><td align="center" bgcolor="#cc9999" width="130"><b></b></td><td align="center" bgcolor="#cc9999" width="180"><b>Intel Core i7-965<br> - Triple Channel</b></td><td align="center" bgcolor="#cc9999" width="180"><b>Intel Core i7-965<br> - Dual Channel</b></td><td align="center" bgcolor="#cc9999" width="180"><b>Intel Core i7-965<br> - Single Channel</b></td></tr><tr><td align="center" bgcolor="#ececec" width="100"><b>3DMark Vantage: CPU Score</b></td><td align="center" bgcolor="#ececec" width="100">19546</td><td align="center" bgcolor="#ececec" width="100">19513</td><td align="center" bgcolor="#ececec" width="100">19484</td><tr><td align="center" bgcolor="#ececec" width="100"><b>Valve Particle Simulation Benchmark</b></td><td align="center" bgcolor="#ececec" width="100">153</td><td align="center" bgcolor="#ececec" width="100">153</td><td align="center" bgcolor="#ececec" width="100">153</td><tr><td align="center" bgcolor="#ececec" width="100"><b>Cinebench R10 1-CPU</b></td><td align="center" bgcolor="#ececec" width="100">4252</td><td align="center" bgcolor="#ececec" width="100">4240</td><td align="center" bgcolor="#ececec" width="100">4222</td></tr><tr><td align="center" bgcolor="#ececec" width="100"><b>Cinebench R10 Multi-CPU</b></td><td align="center" bgcolor="#ececec" width="100">18569</td><td align="center" bgcolor="#ececec" width="100">18368</td><td align="center" bgcolor="#ececec" width="100">18284</td><tr><td align="center" bgcolor="#ececec" width="100"><b>x264 HD Benchmark</b></td><td align="center" bgcolor="#ececec" width="100">27.23 fps</td><td align="center" bgcolor="#ececec" width="100">27.21 fps</td><td align="center" bgcolor="#ececec" width="100">27.06 fps</td></tr><tr><td align="center" bgcolor="#ececec" width="100"><b>WinRAR 3.71 Compression</b></td><td align="center" bgcolor="#ececec" width="100">2:39</td><td align="center" bgcolor="#ececec" width="100">2:44</td><td align="center" bgcolor="#ececec" width="100">3:05</td><tr><td align="center" bgcolor="#ececec" width="100"><b>SuperPI 1M</b></td><td align="center" bgcolor="#ececec" width="100">12.960s</td><td align="center" bgcolor="#ececec" width="100">13.061s</td><td align="center" bgcolor="#ececec" width="100">13.207s</td><tr><td align="center" bgcolor="#ececec" width="100"><b>SuperPI 32M</b></td><td align="center" bgcolor="#ececec" width="100">12:04.755</td><td align="center" bgcolor="#ececec" width="100">12:08.356</td><td align="center" bgcolor="#ececec" width="100">12:57.356</td><tr></table>

We re-ran these benchmarks over and over and over again, because we were quite simply puzzled. However, as you can see, generally speaking Nehalem is certainly not memory bandwidth limited. In fact, the only time we noticed a significant performance difference between single and triple-channel is during WinRAR compression, which is a completely memory bandwidth bound workload. There is also an appreciable difference in SuperPI, but even there the difference is nowhere near as large as what we would expect. This is a significant improvement over the Core microarchitecture, which would be hugely bottlenecked by a single-channel memory interface.

Let's see if the synthetic tests can shed any light on this phenomena:


In their native triple-channel memory configuration, the Core i7 processors produce some extremely impressive bandwidth numbers. In fact, the results demonstrate that at DDR3-1066 the Core i7 has nearly 100% more memory bandwidth than the previous Core 2 architecture running at dual-channel DDR2-1066. Even in dual-channel mode, the Core i7 series have almost 90% more bandwidth than previous processors. In single-channel mode, bandwidth drops precipitously, but it still equivalent to a dual-channel DDR2 interface. Clearly, Intel have done a tremendous designing Nehalem's memory architecture.


On the latency front, we also see some tremendous improvements. The integrated memory controller and triple-channel memory combine to produce the lowest latency numbers that we have ever seen. Interestingly, there is clearly a latency hit when utilizing the triple-channel mode since in dual-channel mode latency is 15% lower.


What can we say? From a real-life perspective, the triple-channel memory configuration does not really appear to be necessary for the consumer market. Having said that, our sample of benchmarking applications was fairly limited for this test, so it is too early to make a definitive judgement. However, when you consider the mighty Everest results, it is clear that the triple-channel DDR3 memory and integrated memory controller are a tremendously potent combination, and we look forward to seeing some insane new DDR3 bandwidth world records in the coming weeks.
 
Last edited:

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,106
Location
Montreal
PCMark Vantage x64 / WinRAR

PCMark Vantage x64


PCMark Vantage Advanced 64-bit Edition (1.0.0.0)
PCMark Suite / Default Settings
Comparison: Generated Score

The main focus of our General Tasks category lies with the most recent installment of the PCMark series, Vantage. While still classified under the description of a Synthetic benchmark, PCMark Vantage uses many of Vista's (Note - Vantage is Vista-only) built-in programs and features along with its own tests, so it is "real-world" applicable in regards to CPU performance. The following is a general list of the tests in the PCMark suite, very much in line with tasks of an average user: Data encryption, Data compression, CPU image manipulation (compression/decompression/resize), Audio transcoding,Video transcoding,Text editing,Web page rendering, Windows Mail, Windows Contacts, and CPU game test.




As you can see, only the (previously) über-high-end Core 2 Extreme QX9770 3.2Ghz model comes close to competing with the new Core i7 processors, and even then it is beaten by the 2.66Ghz i7-920 model. Our real-life experience with these processors supports these results, as everything does actually feel faster than with a Core 2 Quad.


WinRAR


WinRAR 3.7.1
Test: Compression of 1GB of Assorted Files
Comparison: Time to Finish

One of the most popular file compression tools, we take a 1GB batch of files and archive them, timing the task until completion.



If you compare the i7-965 to the QX9770 on a clock-per-clock basis, the Core i7 chip is a substantial 30% faster. As we mention in all our motherboard reviews, WinRAR is always the first application that we install on a fresh Windows installation, and it is one of the tools we use the most day in, day out. Therefore, witnessing this level of performance improvement is absolutely fantastic. In particular, consider the enormous 63% performance difference between the i7-920 and the Q6600. That is a $284 chip demolishing the most popular quad-core model on the market. Clearly, you don't have to spend big bucks to benefit from this new microarchitecture.
 
Last edited:

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,106
Location
Montreal
Cinebench R10 / Valve Particle Simulation

Cinebench R10


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


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


Cinebench is arguably one of the the most efficient programs at utilizing all available cores, and it perfectly showcases the Core i7's unparalleled multi-threading processing power. In particular, pay close attention the difference between the single-threaded and multi-threaded results. The performance improvement is more than 4X, which signifies that Hyper-Threading is actually contributing to improving multi-threaded performance.


Valve Particle Simulation Benchmark


Valve Particle Simulation Benchmark
Default
Comparison: Particle Performance Metric

Originally intended to demonstrate new processing effects added to Half Life 2: Episode 2 and future projects, the particle benchmark condenses what can be found throughout HL2:EP2 and combines it all into one small but deadly package. This test does not symbolize the performance scale for just Episode Two exclusively, but also for many other games and applications that utilize multi-core processing and particle effects. As you will see the benchmark does not score in FPS but rather in its own "Particle Performance Metric", which is useful for direct CPU comparisons.



Designed by Valve as a special showcase for multi-core processors, Valve Particle Simulation Benchmark is obviously heavily multi-threaded and produces exceptionally reliable results. As a result, the Core i7 series once again demonstrates its multi-threading dominance.
 
Last edited:

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,106
Location
Montreal
iTunes MP3 Conversion / x264 HD Benchmark

iTunes MP3 Conversion



iTunes v7.6.1.9
Test: Convert raw 56 minute .WAV file to .MP3 (192kbps)
Comparison: Time to Finish

Apple iTunes, a necessary tool for millions of iPod users, and used by many who are not. The simple Mac styled GUI for all things audio has a wide range of features which allow users to do all sorts of things to their music. One of the most common tools is converting old and archaic CD Audio into the digitized format of MP3's. iTunes allows users to import songs from their CD while at the same time converting them to MP3 to be put onto an iPod or other MP3 Playing device. This process, while by no means the most difficult challenge in our CPU assault, is a common tool and very much affected by system efficiency.



Although iTunes is supposed to support two threads, it is very lackluster when it comes to properly utilizing more than one core. As a result, pure clock speed and L2 cache size are what matter in this benchmark, and the Penryn-based processors are able to put on a strong showing. From a clock-per-clock point-of-view, both microarchitectures are effectively equal in this seemingly single-threaded benchmark.


x264 HD Benchmark


x264 HD Benchmark v1.0
Test: MPEG-2 HD 720P Video Clip Conversion to x264
DVD Video Length: 30 Seconds
Comparison: FPS of Second Pass

x264 is quickly becoming the new codec of choice for encoding a growing number of H.264/MPEG-4 AVC videos. Think of it as the new Divx of HD and you can understand why we felt it critical to include. Tech Arp's recent development of the x264 HD Benchmark takes a 30 second HD video clip and encodes it into the x264 codec with the intention of little to no quality loss. The test is measured using the average frames per second achieved during encoding, which scales with processor speed and efficiency. The benchmark also allows the use of multi-core processors so it gives a very accurate depiction of what to expect when using encoding application on a typical full length video.



The x264 HD Benchmark is multi-threaded and very impressively so. It effectively places 100% load across all eight logical cores. As a result, the Core i7 processors further demonstrate their multi-threaded power and post a 40%+ clock-per-clock performance boost over the other processors. If you frequently encode high-quality x264 video files, the Core i7 is the processor that you have been waiting for. There is no doubt about it, with properly coded software this new microarchitecture is absolutely dominant when it comes to multimedia tasks.
 
Last edited by a moderator:

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,106
Location
Montreal
3DMark06 / 3DMark Vantage

3DMark06


3DMark06 v1.1.0
Graphic Settings: Default
Resolution: 1280X1024

Test: Specific CPU Score and Full Run 3Dmarks
Comparison: Generated Score

The Futuremark 3DMark series has been a part of the backbone in computer and hardware reviews since its conception. The trend continues today as 3DMark06 provides consumers with a solid synthetic benchmark geared for performance and comparison in the 3D gaming realm. This remains one of the most sought after statistics, as well as an excellent tool for accurate CPU comparison, and it will undoubtedly be used for years to come.



With its high clock speed and huge L2 cache, the $1500+ QX9770 barely competes with the $284 i7-920 when it comes CPU score. In the overall score, there is a mere 10% difference performance difference between the high-end i7-965 and the more budget-friendly i7-920.


3DMark Vantage


3DMark Vantage v1.0.1
Graphic Settings: Performance Preset
Resolution: 1280X1024

Test: Specific CPU Score and Full Run 3Dmarks
Comparison: Generated Score

3DMark Vantage is the follow-up to the highly successful 3DMark06. It uses DirectX 10 exclusively so if you are running Windows XP, you can forget about this benchmark. Along with being a very capable graphics card testing application, it also has very heavily multi-threaded CPU tests, such Physics Simulation and Artificial Intelligence (AI), which makes it a good all-around gaming benchmark.



Unlike its predecessors, 3DMark Vantage was clearly designed with multi-core processors in mind, and it has very technically impressive and heavily multi-threaded CPU tests. Therefore, as you can see, the Core i7's really distinguish themselves once again with their CPU score. The overall score is fairly stagnant across all the tested processors, and that is due to the fact that our GeForce 9800GTX is a bottleneck in this graphics-intensive benchmark.
 
Last edited:

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,106
Location
Montreal
Crysis / Half-Life 2: Episode 2

Crysis


Crysis v1.2
Graphic Settings: Medium Settings
Resolution: 800X600
Anti Aliasing: 0
Quality Settings: Medium
Global Settings: DX10 Enabled/ 64Bit Enabled

Test 1: Ice benchmark_CPU2 demo
Comparison: FPS (Frames per Second)

Arguably the most hardware intensive game on the market today, Crysis has been chosen for its obvious ability to be able to showcase the differences between processors and to showcase just how far one will need to go in the quest for maximum performance. The game also features the renowned CryEngine, the power behind the incredible graphics, which is expected to be foundation of future titles.



Crysis is known as being particularly graphics-intensive, but it does feature a multi-threaded engine As a result, by reducing resolution and quality settings, we were able to showcase the multi-core advantage. However, once again, Core i7 processors do not really distinguish themselves from their quad-core Penryn predecessor. They are still very fast at gaming, but we aren't seeing the huge gains that were achieved in the productivity and multimedia applications.


Half-Life 2: Episode 2


Half-Life 2: Episode 2
Graphic Settings: High
Resolution: 1024x768
Filtering: Bilinear / 0AF
Visual Settings: High (All)
Test 1: HWC Custom Timedemo
Comparison: FPS (Frames per Second)

Released years ago in 2004, the latest chapter in the Gordon Freeman story makes it to our list just because its so dang popular. With over a dozen titles spawned off the legendary Valve Source Engine, chances are one of the games you have played recently has been a part of the family, games such as Counter Strike: Source, Team Fortress 2 and Dark Messiah. While aging in relation to many of the other titles featured in reviews, we believe the games popularity will allow you to get an accurate idea of exactly what type of performance gains can be expected with the hardware in games you are sure to play.



The Core i7 processors once again put on a respectable showing, but it is clear that HL2:EP2 is partial towards the Penryn architecture's larger L2 cache. Even the popular 3.0Ghz E8400 is able to compete with the mid-range 2.93Ghz i7-940.
 
Last edited:

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,106
Location
Montreal
Supreme Commander / Unreal Tournament 3

Supreme Commander


Supreme Commander v1.1.3280
Graphic Settings: High
Resolution: 1280X1024
Anti Aliasing: 0
Fidelity Presets: High (All)

Test 1: Custom Bench Replay
Comparison: Time Comparison. Timed till completion of replay at 10X speed.

While not frequently found in many mainstream hardware reviews, we felt it beneficial to bring the unique style of Supreme Commander to our benchmark suite. Upon release, SupCom was heralded for packing an extreme punch when it came to CPU requirements, so much so, that many people recommended not even running it without a dual core CPU. The Moho Engine, developed by Gas Powered Games is unique in the fact that it was one of the first and continues to be one of the few that is designed with true out of the box multi threading capabilities. This allows the game to divide up tasks such as AI Calculations, Rendering, Sound, Drivers and others to separate CPU cores, resulting in far better utilization of multi-core system and a more accurate assessment of CPU performance for multi threaded gaming.

The test we designed takes a large scale 8 player map with 7 AI and records a 20 minute demo. This replay is then sped up to the maximum playback speed (10x realtime) and timed until completion. The significance of running the test in this fashion, rather than the traditional method of an FPS comparison is that the CPU is so heavily utilized as armies grow that the gameplay itself is forced to decrease, and at times may eventually lead to play slower real time. Take note of the CPU's that complete the replay with faster times, these will generally allow the game to last longer at an acceptable speed rather than finding yourself waiting many more real seconds for every second of game operation.



Although it is multi-threaded, Supreme Commander can only form 4 threads. As a result, the Core i7's hyper-threading capabilities go unutilized in this benchmark. Nevertheless, the i7-965 posts a respectable 7% performance improvement over the identically-clocked QX9770.

Unreal Tournament 3


Unreal Tournament 3 v1.2
Graphic Settings: Low
Resolution: 1024X768
Visual Settings: Texture Detail: 1/5 - Level Detail 1/5

Test 1: UT3 Benchmark DM-Shangrila Bot Match (200 Seconds)
Bots: 32
Comparison: FPS (Frames per Second)

The latest in the Epic saga, Unreal Tournament 3 is the beginning of a whole new breed of games. Using the redeveloped Unreal Engine 3, the game uses numerous advanced rendering techniques and physics processing effects not to mention a whole slew of new visual developments. As you can imagine, that makes it a worthy candidate for CPU comparison and allows some of the latest processors to showoff their readiness for the games of tomorrow.



The Core i7 processors once again performs quite well, but the QX9770 is able to squeak by and take a 2% lead. This is well within the margin of error between runs, and it is fair to state that both Nehalem and Penryn perform equally in Unreal Tournament 3.
 
Last edited:

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,106
Location
Montreal
Power Consumption/Temperature Testing

Although the Core i7 series is manufactured with the same 45nm process as current Core 2 processors, each microarchitecture has unique thermal characteristics and power consumption figures. There is no doubt that Intel has focused on energy-effiency this new microarchitecture, but they have also added a significant number of new performance improvements features that generally increase power consumption.
With this in mind, let's take a closer look at the Core i7 series power consumption and thermal output.


Power Consumption

For our power consumption test, we enabled all the Intel energy saving features in the BIOS, then let the system idle for 15 minutes and measured the peak wattage through our UPM EM100 power meter. We repeated the same procedure for 15 minutes of Prime 95 Blend load. Here are the results:


The results are pretty much what we expected. Under load, the Nehalem architecture has higher power consumption than Penryn does, and this despite having over 10% less transistors. The reasons for this higher power consumption are the integrated memory controller and built-in QPI Link. Both of these new core components are quite power-hungry, especially under full load scenarios. Having said that, you must take into account the simple fact that Nehalem does more work per clock than Penryn does, and when viewed from that perspective, Nehalem is in fact a more efficient design when it comes to Performance per Watt.

On the idle front, the situation is quite promising. For example, the Core i7-965 uses 13% less power at idle than the identically-clock QX9770, which is a noteworthy achievement by any measure. All three Core i7 series processors recorded identical idle power consumption figures because they all downclocked to approximately 1536Mhz (128Mhz x 12).

Clearly, the Power Control Unit (PCU) is doing a good job of dynamically lowering the voltage and frequency of the CPU cores, or perhaps even using the Power Gates technology to completely shut down and place in a C6 sleep mode some of the idle cores.


Temperature Testing

For the temperature testing, we had both the stock Intel Core i7 CPU cooler and a Thermalright Ultra-120 eXtreme with an LGA1366 mounting kit to test out. OCCT v2.0.0a was run for 15 minutes and the temperatures were recorded at 5 minute intervals and the results averaged out. We relied on the new CoreTemp 0.99.3 utility to accurately report temperatures. The ambient temperature was 21°C/71.6°F.


As you can see, and Intel effectively told us, the stock Core i7 cooler is not intended for the high-end 965 model. However, it is capable of keeping the other two models at fairly respectable load temperatures. The venerable Thermalright Ultra-120 eXtreme was much better at cooling these new processors, and even the high-end 965 is reigned-in a great deal. As usual, those interested in serious manual overclocking will want to invest in a good quality cooler.
 
Last edited by a moderator:

MAC

Associate Review Editor
Joined
Nov 8, 2006
Messages
1,106
Location
Montreal
Initial Impressions

Initial Impressions



And now, on with the show...

With everything we have so seen so far there sure is a lot to say, but to be honest with you it is hard to wrap one's mind around what Intel has accomplished with this new architecture. The Core i7 has had to follow in the tremendous footsteps of the Penryn microarchitecture, which not only provided awesome performance but unparalleled overclocking capabilities as well. As a result, the stakes were high, and the possibility for disappointment even higher. The level of hype and the sheer volume of misinformation did nothing to bolster pre-launch public opinion. However, in our opinion, Core i7's is a great success and improves upon the last generation in nearly every way. It provides performance where needed and it absolutely blows away the competition from AMD to the point that Team Green's position in this review barely warrants a mention.

When it comes to innovation in the processor market, consumers are usually left wanting since vast design changes are few and far between. This is because companies such as Intel and AMD take massive financial gambles betting that their new architectures will strike a cord with the buying public. Naturally, some like the Core 2 series succeed while others flounder. Intel took this same gamble with the i7 but in the high stakes poker game that is the current processor marketplace, it seems they stacked the deck and came up with a Royal Flush.

As they say; with all good things comes a bit of bad. We are a little disappointed that the single-threaded performance has not improved much (if at all) over Penryn but for us, the staggering multi-threaded performance more than makes up for this shortcoming. However, we can't forget the simple fact is that there are few resource heavy and time-consuming single-threaded applications that we use on a daily basis. On the other hand, most applications like 3D rendering, image editing, video encoding, WinRAR, Office 2007 suite, etc., are multi-threaded and all benefit substantially from Nehalem’s capabilities. Specifically, as we demonstrated in the feature test, the addition of Hyper-Threading was a brilliant move by Intel and the performance improvements are clearly night and day. Having said that, we hesitate to give the Core i7 series our unconditional thumbs up simply because the performance improvements are so very focused on this one aspect. Not to mention that the platform as a whole (CPU, memory and motherboard) will be quite expensive at first but this is the price that early adopters have always had to pay.

We fully understand that the majority gamers will not be overly impressed with Core i7 right now.Yes, there are more multi-threaded games than ever, but many are poorly coded and the performance improvements are just not visible at the higher resolutions that most gamers use. Therefore, if you're a gamer and you currently have a dual or quad-core 45nm Core 2 processor, you probably should not rush out to buy a Core i7 system. However, we can assure that once games highly multi-threaded games like Alan Wake are released, your opinions of this platform will change immediately.

If you are a ‘general purpose’ power user or simply a ‘Tech Head’ who likes fiddling with the latest and greatest, the Core i7 platform will give you a lot to smile about….and I can tell you that we are smiling from ear to ear right about now.


Please visit the comment thread in our forums to post your questions and opinions

Our utmost thanks goes out to Intel for including us in this latest product launch and for making this review possible!

 
Last edited:
Status
Not open for further replies.

Latest posts

Twitter

Top