| Intel Core i7-870 Processor BX80605I7870 |
| Reviews - Featured Reviews: Processors | |||||||||||||||||||||||||||||||||||||||||||||||||
| Written by Mathew Williams | |||||||||||||||||||||||||||||||||||||||||||||||||
| Wednesday, 09 September 2009 | |||||||||||||||||||||||||||||||||||||||||||||||||
Intel Core i7-870 ProcessorWhen it launched last year, the Core i7 platform redefined Intel's high-end processor segment. Unfortunately, the platform was also accompanied by high-end prices. As such, it has seen only a minimal share of the desktop market. However, Intel has released a new platform that it claims will bring the performance of Nehalem to mainstream price points. Referred to as Lynnfield, the new platform shares its core microarchitecture with the i7-900 series, but introduces a completely new chipset and socket design. This means that not only will we see new processors, but new motherboards as well. Benchmark Reviews has been testing several of these new items and will be covering the launch in detail over the next few days. In this article, we examine the Intel Core i7-870 Quad Core Processor model BX80605I7870. Of the three processors launched this week under the Lynnfield platform, the Core i7-870 is considered the high-end part. All three chips are based on the same die, feature 8 MB of shared cache, and are packaged using the new LGA 1156 socket. They also feature a new version of Intel's Turbo Boost Technology that dynamically overclocks the individual processor cores depending on workload. At 2.93 GHz, though, the Core i7-870 boasts the highest clock speed of the three. As an i7-800 series model, it also features Hyper-Threading for a total of 8 threads. It should be a recipe for success, but we'll find out for certain in the benchmarking section.
With a launch-MSRP of $555 for the BX80605I7870 kit, one would hope that the i7-870 lives up to the hype. It's certainly a far cry from mainstream as Intel suggests. Compared to the i7-900 series, however, the i7-870 does have the advantage of using the new P55 chipset. Intel moved the bulk of the Northbridge functions to the processor and combined the rest with the Southbridge to deliver a single chip solution called Ibex Peak. The result should be considerably less expensive motherboards. Still, the processor itself carries a hefty price tag and consumers will want performance to match. About Intel Corporation |
|||||||||||||||||||||||||||||||||||||||||||||||||
|
Core i5-750 |
Core i7-860 |
Core i7-870 | |
| Base Clock Speed |
2.66 GHz |
2.80 GHz |
2.93 GHz |
| Max Turbo Frequency |
3.2 GHz |
3.46 GHz |
3.6 GHz |
| Cores/Threads |
4/4 |
4/8 |
4/8 |
| Shared Cache | 8MB | 8MB | 8MB |
| Memory Support |
DDR3 1333 MHz |
DDR3 1333 MHz |
DDR3 1333 MHz |
| Chipset Support |
P55 |
P55 |
P55 |
| Socket | LGA 1156 | LGA 1156 | LGA 1156 |
| TDP |
95W |
95W |
95W |
| Transistor Count | 774M | 774M | 774M |
| Die Size |
296mm2 |
296mm2 |
296mm2 |
| MSRP | $199 | $285 |
$555 |
Closer Look: Intel Core i7-870
All of the Lynnfield processors launched this week are based on the same Nehalem microarchitecture as the Bloomfield Core i7-900 series. However, as I mentioned in the intro, significant changes have been made to reduce costs. First up is the smaller socket design, 1156 vs 1366. The lower pin-count is primarily the result of a revised memory controller. Triple channel memory support has been dropped in favor of a more economical dual channel configuration. However, the new Lynnfield processors officially support DDR3 1333, a step up from 1066. Unchanged, it seems, are the cache sizes and associativity. Each core gets 64 KB of L1 cache, 256 KB L2 cache, and shares a larger 8 MB L3 cache.
The bus speed is also unchanged and, as you can see in the screenshot above, sits at 133 MHz. With turbo disabled, the maximum multiplier for the i7-870 is 22, for a maximum clock speed of 2.93 GHz. Speed Step, C1E, and C-States are used to independently manage the clock speed and power state of each core. As such, you'll find that under light workloads all four cores often do not have to run at that maximum frequency.
For heavy workloads, specifically those that do not take advantage of multi-threading, enabling Intel's Turbo Boost Technology can have the opposite effect. It effectively overclocks the cores in use by raising the multiplier and voltage of each core independently until the maximum multiplier or TDP is reached. For the Core i7-870, the maximum turbo multiplier is 27, resulting in a maximum turbo speed of 3.6 GHz. However, this is with only one core in use. For each additional core in use, the maximum multiplier is reduced by one. For example, if all four of the i7-870's cores were in use, all four cores would have a maximum turbo multiplier of 24, or 3.2 GHz.
The Core i7-870 also features Intel's Hyper-Threading Technology, otherwise known as simultaneous multithreading, or SMT. It's the same technology Intel used in their Bloomfield processors and provides each individual core with two processing threads. The result is 8 threads total for the i7-870, as you can see in the screenshot above. We'll examine the performance benefits of Hyper-Threading and Intel's Turbo Boost Technology later on in the review.
Testing and Results
Before I begin any benchmarking or overclocking, I thoroughly stress the CPU and memory by running Prime95 on all available cores for 12 hours. If no errors are found, I move on to a gaming stress test. To do this, I use Prime95 again to stress the processor, while running an instance of FurMark's stability test on top of this. If the computer survives this test for 1 hour without lockup or corruption, I consider it to be stable and ready for benchmarking.
Each benchmark test program begins after a system restart, and the very first result for every test will be ignored since it often only caches the test. This process proved extremely important in the World in Conflict and Supreme Commander benchmarks, as the first run served to cache maps allowing subsequent tests to perform much better than the first. Each test is completed five times, with the average results displayed in our article.
The test systems below have all been carefully configured to reduce any extraneous variables. For this review, we will be examining platform performance. As such, the motherboards and processors will vary. However, with the exception of our LGA1366 test system, all of our test platforms use the same memory, video card, and hard drive. We were unable to test the LGA1366 under these conditions, but felt it worthwhile to include the i7-920 in this review and will only include results from benchmarks that are unaffected by these components.
Intel LGA775 Test System
-
Processors: Intel Core 2 Quad Q9450
-
Motherboard:Asus P5Q3
-
System Memory: 2x2GB DDR3 1066 7-7-7-20 1.5v
-
Video: ASUS EAH4850
-
Disk Drive: Western Digital 6400AAKS 640GB
-
Operating System: Windows XP Professional SP3
Intel LGA 1156 Test System
-
Processors:Intel Core i5-750, Core i7-870
-
Motherboard: Intel DP55KG Extreme
-
System Memory: 2x2GB DDR3 1066 7-7-7-20 1.5v
-
Video: ASUS EAH4850
-
Disk Drive: Western Digital 6400AAKS 640GB
-
Operating System: Windows XP Professional SP3
AMD Socket AM3 Test System
-
Processors:AMD Phenom II 965 BE
-
Motherboard:Asus M4A79T Deluxe
-
System Memory: 2x2GB DDR3 1066 7-7-7-20 1.5v
-
Video: ASUS EAH4850
-
Disk Drive: Western Digital 6400AAKS 640GB
-
Operating System: Windows XP Professional SP3
Intel LGA1366 Test System
-
Processors:Intel Core i7-920
-
Motherboard:Gigabyte GA-EX58-UD4P
-
System Memory: 3x2GB DDR3 1066 7-7-7-20 1.5v
-
Video:ASUS ENGTX285 TOP
-
Disk Drive:OCZ Vertex EX 120GB SLC SSD
-
Operating System: Windows XP Professional SP3
Benchmark Applications
-
EVEREST Ultimate Edition v5.01 by Lavalys
-
Passmark PerformanceTest
-
PCMark05 v1.2.0 by Futurmark Corporation
-
Crysis Benchmark Tool
-
Devil May Cry 4 Benchmark
-
SPECviewperf v10.0 by Standard Performance Evaluation Corporation
-
MAXON CINEBENCH R10
EVEREST Benchmark Tests
EVEREST Ultimate Edition is an industry leading system diagnostics and benchmarking solution for enthusiasts PC users, based on the award-winning EVEREST Technology. During system optimizations and tweaking it provides essential system and overclock information, advanced hardware monitoring and diagnostics capabilities to check the effects of the applied settings. CPU, FPU and memory benchmarks are available to measure the actual system performance and compare it to previous states or other systems. Furthermore, complete software, operating system and security information makes EVEREST Ultimate Edition a comprehensive system diagnostics tool that offers a total of 100 pages of information about your PC.
The results of EVEREST's memory tests are a bit unexpected. Despite being reduced to dual channel memory, it appears that the new Lynnfield processors can still keep up with the i7-920. In fact, the i7-870 even pulls ahead in memory read bandwidth. What's interesting is that it actually falls a little behind the i5-750 in the other tests.
Moving on to integer tests, the rankings are much closer together, with the Core i7-870 taking the lead in the Queen and AES benchmarks. The Photoworxx test, however, is a bit more memory dependent and the i7-920's triple channel memory configuration helps it pull ahead.
In the floating point benchmarks, the trend continues. The Core i7-870 takes the lead, the i5-750 comes in second, and the AMD X4 965 and Intel Q9450 battle it out for last place. The extended precision Sin Julia benchmark is particularly interesting: the Core i7-870's Hyper-Threading helps it achieve almost twice the performance of the i5-750.
Passmark PerformanceTest
PassMark PerformanceTest is a PC hardware benchmark utility that allows a user to quickly assess the performance of their computer and compare it to a number of standard 'baseline' computer systems. The Passmark PerformanceTest CPU tests all benchmark the mathematical operations, compression, encryption, SSE, and 3DNow! instructions of modern processors.
In our tests there were several areas of concentration for each benchmark, which are combined into one compound score. This score is referred to as the CPU Mark, and is a composite of the following tests: Integer Math, Floating Point Math, Find Prime Numbers, SSE/3DNow!, Compression, Encryption, Image Rotation, and String Sorting. For this review, we've also decided to run the memory benchmark, which results in a composite score based on the following tests: small block allocation, cached read, uncached read, write performance, and large block allocation.
Passmark's composite scores confirm what we saw in the raw EVEREST benchmarks. The Core i7-870 actually outpaces the i7-920 in both memory and CPU scores. Without hyper-threading, though, the Core i5-750 can't quite keep up in the CPU tests. It does, however, post a decent memory score.
PCMark05 Benchmark Suite
Using synthetic benchmarks to compare one product to another has some distinct advantages when testing similar hardware, yet I have never found myself completely satisfied by the process. I have come to understand that they're important for comparing "apples to apples", and that the results are usually very consistent. But as with any synthetic benchmark, the numbers can often mean very little more than just numbers. We don't take a high score on a synthetic benchmark to mean that a product will/should perform well, and neither should you. The difference between projected performance and actual performance is the difference between fire and the fire-fly.
PCMark is a series of computer benchmark tools developed by Futuremark. The tools are designed to test the performance of the user's CPU, read/write speeds of RAM and hard drives. We have used these tests to simulate a battery of applications and tasks, which will produce results we can compare to other systems using similar hardware.
In PCMark05, the Core i7-870 again sits at the top of the charts. However, the performance difference compared to the i5-750 is likely due to the difference in clock speed. The majority of PCMark05's benchmarks do not take advantage of all of the i7-870's 8 processing threads.
Crysis Gaming Tests
Crysis uses a new graphics engine: the CryENGINE2, which is the successor to Far Cry's CryENGINE. CryENGINE2 is among the first engines to use the Direct3D 10 (DirectX10) framework of Windows Vista, but can also run using DirectX9, both on Vista and Windows XP.
Roy Taylor, Vice President of Content Relations at NVIDIA, has spoken on the subject of the engine's complexity, stating that Crysis has over a million lines of code, 1GB of texture data, and 85,000 shaders. To get the most out of modern multicore processor architectures, CPU intensive subsystems of CryENGINE 2 such as physics, networking and sound, have been re-written to support multi-threading.
Crysis offers an in-game benchmark tool, which is similar to World in Conflict. This short test does place some high amounts of stress on a graphics card, since there are so many landscape features rendered. For benchmarking purposes, Crysis can mean trouble as it places a high demand on both GPU and CPU resources. Benchmark Reviews uses the Crysis Benchmark Tool by Mad Boris to test frame rates in batches, which allows the results of many tests to be averaged.
When testing a CPU, the settings we choose are a bit different than a typical video card review. As you'll see in the charts below, modern games can easily max out the processing abilities of the video card and mask any differences between the CPUs. For this reason, we generally start at low resolutions and quality settings and slowly ramp them up until we hit the limit of the video card. This allows us to quickly distinguish differences between CPUs and identify any other limiting hardware.
Crysis is a heavily CPU-bound game, but only if the video card can keep up. The benchmark settings were specifically chosen to illustrate this. At 1024x768, we see a clear difference in framerates, with the Core i7-870 securing a commanding lead over the rest of the processors. However, as we tax the video card more with increased resolutions and settings, our HD 4850 becomes the limiting factor. That said, you'll want to pair these processors with a high end video card to get the most out of them in gaming.
Devil May Cry 4 Gaming Tests
Devil May Cry 4 is one of the newest additions to Benchmark Reviews' testing suite. Fortunately for us, Capcom recognized the community's interest in hardware testing and included a built in benchmarking tool with this game. In fact, it is even included it in the demo. The benchmarking tool runs through four different scene, all of which I highly recommend watching. However, for the purposes of our review, we only report the scores of the most challenging tests: scenes 2 and 4.
Similar to the Crysis gaming benchmarks, we will start testing DMC 4 at lower settings to reduce the impact of GPU limitations. From there, we'll slowly increase resolution and quality settings until we reach the limit of our HD 4850.
Similar to Crysis, scene two of Devil May Cry 4 places Core i7-870 in the lead in the low resolution test. However, as resolution and settings are increase, it actually falls behind the Phenom II X4 965 and the Intel Q9450. As I mentioned in the i5-750 review, though, we did run into a few software issues during benchmarking. At times, our Radeon HD 4850 would revert to PCIe 1.0 speeds when used with the Intel DP55KG motherboard.
In scene 4 of Devil May Cry 4, things look a little better for the Core i7-870. Once again it's at the front of the pack in the low resolution test, and essentially matches X4 965 and Q9450 in the higher resolutions as we approach the limit of our HD 4850.
SPECperfview 10
SPECviewperf is a portable OpenGL performance benchmark program written in C. It was developed by IBM. Later updates and significant contributions were made by SGI, Digital (Compaq, HP), 3Dlabs (Creative Labs) and other SPECopc project group members. SPECviewperf provides a vast amount of flexibility in benchmarking OpenGL performance. Currently, the program runs on most implementations of UNIX, Windows XP, Windows 2000, and Linux.
SPECviewperf parses command lines and data files, sets the rendering state, and converts data sets to a format that can be traversed using OpenGL rendering calls. It renders the data set for a pre-specified amount of time or number of frames with animation between frames. Finally, it outputs the results. SPECviewperf reports performance in frames per second. Other information about the system under test - all the rendering states, the time to build display lists (if applicable), and the data set used - are also output in a standardized report.
A "benchmark" using SPECviewperf is really a single invocation of SPECviewperf with command-line options telling the SPECviewperf program which data set to read in, which texture file to use, what OpenGL primitive to use to render the data set, which attributes to apply and how frequently, whether or not to use display lists, and so on. One quickly realizes that there are an infinite number of SPECviewperf "benchmarks" (an infinite number of data sets multiplied by an almost infinite number of command-line states).
The SPECviewperf CATIA results are somewhat perplexing. As a single-threaded benchmark, the only notable difference between the i5-750 and the i7-870 is clock speed. Even factoring in Turbo, that's only an 11% difference, while the results here suggest a 25% difference in performance. In all likelihood, it's the same software issue we discovered in Devil May Cry 4. Nevertheless, these numbers are good news for the Core i7-870.
MAXON CINEBENCH R10
CINEBENCH comes to us from MAXON, the developers of CINEMA 4D animation software. This benchmark runs several tests to measure performance of the processor and the graphics card under real-world circumstances. It can make use of up to 16 CPUs or CPU cores, but also includes a single-threaded test to provide an accurate comparison of efficiency and performance across a wide range of CPU types.
The resulting values among different operating systems are 100% comparable and therefore very useful with regard to purchasing decision making. It can also be used as a marketing tool for hardware vendors or simply to compare hardware among colleagues or friends.
One of the reasons we include CINEBENCH is to analyze the effects of multiple threads. The Core i7-870 and i7-920 easily top the charts thanks to Intel's hyper-threading. What's interesting is that the i7-870 actually takes the lead. The higher clock speed and the improved Turbo Boost clearly help out here.
Performance Features and Overclocking
In this section, we examine the performance features and overclocking potential of the Core i7-870. As i mentioned earlier, all of the Lynnfield processors launched this week have the advantage of Intel's revised Turbo Boost Technology. As this is an i7-800 series chip, it also has Hyper-Threading enabled. Both of these technologies are enabled by default and were used in all of our benchmarking so far. However, to illustrate impact each has on overall performance, I ran a selection of the benchmarks with both technologies disabled. To further isolate the two, I also ran the benchmarks with only Turbo Boost enabled and only Hyper-Threading enabled. You'll find the results in the charts below.
In addition to the included performance features, the charts also reflect the results of manual overclocking. As we saw with the i5-750, one of the first decisions you'll have to make is whether to enable or disabled Turbo Boost. With it enabled, I was able to increase the bus speed to 160 MHz for a base clock speed of 3.52 GHz and a maximum turbo clock speed of 4.32 GHz. This is roughly the same result I achieved with the i5-750. Disabling Turbo Boost would most likely net a much higher base clock speed for all four cores, but at the cost high single-threaded performance. To stay consistent with previous, I decided to run the benchmarks at the highest Turbo Boost clock speed.
Looking at the EVEREST integer results, we find that in the PhotoWorxx and AES benchmarks, Turbo Boost yielded a higher performance increase than Hyper-Threading compared to the baseline. Of course, enabling both results in the best scores at stock speeds. As is to be expected, though, overclocking had an even more significant effect on performance.
Moving on to Crysis, we find that once again Turbo Boost has more of an impact on performance than Hyper-Threading. Clearly, clock speed is more valuable in Crysis than the jump from four processing threads to eight. Of course, it does have some effect and when both technologies are combined we observe a synergistic effect. It's also worth pointing out that neither technology, nor our attempts at overclocking altered the Core i7-870's performance in the last two tests. No increase to clock speed or processing threads can remove the limitation of our Radeon D 4850.
CINEBENCH finally give the i7-870's Hyper-Threading time to shine. In the multi-threaded benchmark, it's clear that the extra threads help out quite a bit. Quite obviously, however, it has no effect in the single-threaded benchmark. It's here that Turbo Boost comes out on top. Not surprisingly, enabling both technologies and overclocking produces an even more significant increase in performance.
Power Consumption
Life is not as affordable as it used to be, and items such as fuel and electrical energy top the list of resources that have exploded in price over the past few years. Add to this the limit of non-renewable resources compared to demand and you can see that the prices are only going to get worse. Planet Earth is needs our help, and needs it badly. With forests becoming barren of vegetation and snow capped poles quickly turning brown, the technology industry has a new attitude towards suddenly becoming "green". Motherboard manufacturers, in particular, have been touting their new energy saving features. How effective these power management system are, is exactly what we intend to measure in our power consumption tests.
Before we get to the results, let me quickly explain the procedures. We begin with a completely idle system (Windows XP desktop with 0-2% processor utilization), and measure the power draw from the wall with a Kill-a-Watt power meter. Then we restart and disable CnQ, Speed Step, C1E, C-States, and Turbo Boost in the BIOS. This allows us to analyze efficiency of the different power saving technologies. We then max out the available CPU cores/threads using Prime95 to establish a maximum 2D power draw. Finally, as an additional test for the Core i5 and Core i7 processors, we run the 2D load test once again, but with Turbo Boost enabled to determine the impact it has on power consumption.
As I mentioned in the Core i5-750 reviews, Intel has a made quite a bit of progress on their power saving technologies since the Core 2 Quad generation. At idle, the Core i7-870 actually consumes 15 less watts than the Q9450. At load, it's a bit higher, but the i7-870 also performs better, which means it will compete the tasks faster. Not to mention, it's still much less than the power-hungry Phenom II X4 965.
Intel Lynnfield Final Thoughts
Lynnfield has been rumored and speculated over for months. Now that it's finally here, I have to say I'm extremely impressed with the performance I've seen today. Not only does it surpass the previous Core 2 Quad generation, but it also challenges the Core i7-900 series. Thanks to Intel's Turbo Boost technology, the Lynnfield processors should also compare well to higher-clocked dual core processors in single-threaded applications, and in multi-threaded applications the hyper-threading of the i7-800 series should help out significantly.
That said, the platform does have a few drawbacks. First, and most obvious, is the new socket. This will require purchasing a new motherboard, and in most cases, a new heatsink and fan. If your still running DDR2, you'll also need to purchase DDR3 modules as DDR2 is not supported. The other issue at hand is the platform upgrade path. Intel's LGA1366 socket isn't simply disappearing. In fact, they intend to use both sockets for quite some time. In all likelihood, Intel's high-end processors will be released under LGA1366, while LGA 1156 will be platform of choice for Intel's mainstream and performance segment. That said, it's definitely worth considering your future needs before investing in either platform.
One other item worthy of note is the compatibility issue we had with our reference Radeon HD 4850. While we're still working with Intel and AMD to isolate the problem, we have determined that it only occurs when the HD 4850 is paired with a P55 motherboard under Windows XP. Shortly after completing the benchmarking for this article, I installed Vista SP2 and the issues were eliminated. At this point, though, it still isn't clear whether this is an isolated incident or a more widespread problem. If it does turn out to be a widespread problem, a software or driver update will likely be issued to address it.
Core i7-870 Conclusion
The Core i7-870 represents the best Lynnfield currently has to offer. Needless to say, my expectations were quite high. When it comes to performance, though Intel delivered. The i7-870 consistently topped the charts, even outpacing the Core i7-920 in several benchmarks. Without a doubt, this is the Nehalem-level performance Intel promised.
As with all of our systems, we also tested the Core i7-870 for stability and construction quality. While it passed our torture tests just fine, I'm still concerned about the compatibility issues we observed with our HD 4850. Once again, though, I prefer to hold my judgment until we can track down the source of the problem.
Moving on functionality, the Intel Core i7-870 easily earns top marks. As a Lynnfield processor, the i7-870 includes Intel's revised Turbo Boost Technology, which allows it to independently overclock each processing core up to maximum of 3.6 GHz. As an i7-800 series part, it also features Intel's Hyper-Threading technology, which, as we saw in the benchmarks, helps out significantly in highly-threaded applications. Throw in DDR3-1333 support and improved power saving features, and the i7-870 should prove to be very functional chip.
If you still want to get more out of the i7-870, though, our overclocking results indicate that it has plenty of headroom. While our base clock speed was limited by enabling Turbo Boost, we were able to achieve a base clock speed of 3.52 GHz and a maximum turbo clock speed of 4.32 GHz. If, however, you're willing to sacrifice single-threaded performance, disabling Turbo Boost will allow all four cores to reach even higher speeds.
Based purely on performance and features, it would be easy to recommend the Core i7-870. Unfortunately, that's not the complete story. As the top of the line Lynnfield chip, Intel also attached a premium price tag. As of December 2009, the BX80605I7870 kit sells for $549.99 at NewEgg. That's far from Intel's claim of mainstream price points and a real disappointment in terms of value. The single-chip P55 motherboards should bring the platform cost down somewhat. At $279.99, though, the 2.8 GHz Core i7-860 looks like a much better deal.
Pros:
+ Incredible Performance
+ Turbo Boost Technology
+ Hyper-Threading Technology
+ Low Power Consumption
+ DDR3-1333 Support
+ 8 MB Shared Cache
Cons:
- Requires New LGA 1156 Motherboard
- Premium Price
Ratings:
-
Performance: 9.75
-
Construction: 9.0
-
Functionality: 9.75
-
Overclock: 9.75
-
Value: 8.0
Final Score: 9.25 out of 10.
Excellence Achievement: Benchmark Reviews Golden Tachometer Award.
Questions? Comments? Benchmark Reviews really wants your feedback. We invite you to leave your remarks in our Discussion Forum.
Related Articles:
- Best CPU Cooler Performance - Q2 2008
- Cooler Master V6GT CPU Cooler
- Xigmatek HDT-S1283 Exposed Copper Heatpipe CPU Cooler
- Best CPU Cooler Performance LGA1366 - Q1 2009
- ASUS P7P55D EVO P55 LGA1156 Motherboard
- AMD Phenom II X4 810 AM3/AM2+ Processor
- NVIDIA APEX PhysX: CPU vs GPU Efficiency
- Intel Sandy Bridge CPUs Chill Aftermarket Cooling
- Thermaltake V14Pro CPU Cooler CL-P0471
- AMD Athlon-II X2-255 CPU ADX255OCGQBOX
