Core i5-750 and Core i7-870 Processors Review

[nextpage title=”Introduction”]

Intel is launching a new CPU socket called LGA1156 with a new chipset, P55. The first three CPUs based on this new platform will be Core i5-750 (2.66 GHz), Core i7-860 (2.80 GHz) and Core i7-870 (2.93 GHz), all based on the new “Lynnfield” core. We had the pleasure to receive a Core i5-750 and a Core i7-870 sample from Intel before their launch, so let’s take a look at their performance compared to other CPUs from Intel.

As you may be aware, Intel is adopting new names for their CPU families. Core i3 will be the family targeted to entry-level PCs, Core i5 will be the family targeted to mid-range PCs and Core i7 is the family targeted to high-end PCs. Core i3 products were not announced yet and Core i5-750 that we are reviewing today will be the first Core i5 CPU to be launched.

Some Core i7 processors based on a core called “Bloomfield” were already launched (Core i7-920, Core i7-940, Core i7-950, Core i7-965 and Core i7-975), but using a different socket, LGA1336.

Core i3, Core i5 and Core i7 processors have an integrated memory controller, just like CPUs from AMD have since the very first Athlon 64 (launched in 2003). The memory controller from the models released so far accepts only DDR3 memories, with socket LGA1366 CPUs supporting triple-channel configuration and socket LGA1156 CPUs supporting dual-channel configuration. This is one of the main difference between socket LGA1156 and socket LGA1366. This way, socket LGA1366 is targeted to high-end PCs only.

Under dual channel configuration, two memory modules are accessed at the same time. This doubles the available bandwidth compared when we had only one memory being accessed at a time (single channel). So on PCs using dual-channel configuration memory modules must be installed in pairs so the system can achieve its maximum performance. On triple-channel architecture, three memory modules are accessed at the same time, increasing the available bandwidth by 50% if compared to a dual-channel configuration running at the same clock rate. Thus in systems using Core i7-9xx you must install memory modules in multiples of three in order to achieve the maximum performance the CPU can deliver.

As you can see, it is easy to tell which socket a given Core i7 CPU is based. Models targeted to the new socket LGA1156 start with the number eight, while models targeted to the socket LGA1366 start with the number nine.

Another difference between the “old” socket LGA1366 Core i7 and the new socket LGA1156 Core i7 and Core i5 is that Core i7 processors already launched (Core i7-920, Core i7-940, Core i7-950, Core i7-965 and Core i7-975) officially support DDR3 memories up to DDR3-1066/PC3-8500, while the new socket LGA1156 CPUs support DDR3 memories up to DDR3-1333/PC3-10600.

On the pictured below we compare the physical differences between sockets 775, 1156 and 1366.

Figure 1: Socket LGA775, socket LGA1156 and socket LGA1366 CPUs.

Figure 2: Socket LGA775, socket LGA1156 and socket LGA1366 CPUs.

Figure 3: Socket LGA775.

Figure 4: Socket LGA1156.

Figure 5: Socket LGA1366.

[nextpage title=”Introduction (Cont’d)”]

Another important difference between socket LGA1156 and socket LGA1366 processors is the presence of a x16 PCI Express 2.0 controller inside the CPU. This is the first time a CPU manufacturer does something like this; until now, the PCI Express 2.0 controller was embedded on the chipset. The 16 PCI Express lanes present in the CPU can be connected to one x16 PCI Express 2.0 slot truly working at x16 or to two x16 PCI Express 2.0 slots working at x8 each. This new feature should improve the practical bandwidth achieved by graphics cards.

Socket LGA1366 CPUs  talk to the external world (i.e., the chipset) through a bus called QuickPath Interconnect (QPI), which has the same goal as the HyperTransport bus used with AMD CPUs. For a detailed explanation on how QPI bus works, read our Everything You Need to Know About The QuickPath Interconnect (QPI) tutorial. The new socket LGA1156 CPUs, however, use the DMI (Digital Media Interface) bus to talk to the chipset, which is the interface previously used to make the connection between the north bridge and the south bridge chips on Intel chipsets. At a first look this solution may seem worse than using the QPI bus, because the DMI interface provides a maximum transfer rate of 2 GB/s while QPI provides a maximum transfer rate of 4.8 GB/s or 6.4 GB/s, depending on the CPU. However, since now the CPU talks directly to the main video card without using its external bus and without using the chipset, this solution seems adequate. Both QPI and DMI offer a lower bandwidth compared to Core 2 Duo and Core 2 Quad external bus (see table below). Keep in mind that on these CPUs the external bus is used to make the communication between the CPU, the memory and everything else, while on socket LGA1156 and 1366 CPUs the processor has a separated communication channel with the memory and now on socket LGA1156 the CPU has a separated channel to talk with the main video card as well.

An important difference between Core i5 and Core i7 is the presence of the Hyper-Threading technology on the high-end family. This technology emulates two processing cores per physical core. So with Core i7 CPUs the operating system recognizes eight processors, even though the CPU has “only” four cores.

Both Core i5 and Core i7 support Intel Turbo Boost technology, which is essentially an automatic overclocking made by the CPU when it “feels” that the program needs more processing power, feature not present on Core 2 Duo and Core 2 Quad family of processors. Another difference between the two generations is on the memory cache. Core 2 Duo and Core 2 Quad CPUs have only one L2 memory cache, which is shared by all cores, while Core i5 and Core i7 have individual L2 memory caches for each core plus a L3 memory cache that is shared by all cores.

On the tables below we compare the main specs from the CPUs we included in this review. Due to the short time we had to collect data for this review and publish it as soon as possible, we couldn’t add all the processors we wanted in this comparison – especially competitors from AMD. We included the two socket LGA1366 Core i7 we had and the comparison between Core i5-750 and Core i7-920 should be interesting, as both run int
ernally at 2.66 GHz. For the socket LGA775 selection, we decided to include Core 2 Extreme QX 9770 because it is the fastest (and most expensive) Intel CPU based on Core microarchitecture. A Core 2 Duo E8400 was included to represent a mainstream dual-core CPU and a Core 2 Quad Q6600 was included to represent a mainstream quad-core CPU based on Core architecture.

CPU	Cores	HT	Internal Clock	Turbo Clock	QPI or FSB Clock	Base Clock	Core	Technology	TDP	Socket	Price
Core i5-750	4	No	2.66 GHz	3.20 GHz	2 GB/s	133 MHz	Lynnfield	45 nm	95 W	1156	USD 196
Core i7-870	4	Yes	2.93 GHz	3.60 GHz	2 GB/s	133 MHz	Lynnfield	45 nm	95 W	1156	USD 562
Core i7-920	4	Yes	2.66 GHz	2.93 GHz	4.8 GB/s	133 MHz	Bloomfield	45 nm	130 W	1366	USD 284
Core i7-965	4	Yes	3.20 GHz	3.46 GHz	6.4 GB/s	133 MHz	Bloomfield	45 nm	130 W	1366	USD 999
Core 2 Extreme QX9770	4	No	3.20 GHz	–	12.8 GB/s	400 MHz	Yorkfield	45 nm	136 W	775	USD 1,399
Core 2 Quad Q6600	4	No	2.4 GHz	–	8.5 GB/s	266 MHz	Kentsfield	65 nm	105 W	775	USD 183
Core 2 Duo E8400	2	No	3 GHz	–	10.6 GB/s	333 MHz	Wolfdale	45 nm	65 W	775	USD 163

TDP stands for Thermal Design Power which advises the user of the maximum amount of heat the CPU can dissipate. The CPU cooler must be capable of dissipating at least this amount of heat.

The prices listed are the official prices for distributors based on 1,000 quantities. The end-user price is higher than the prices listed.

CPU	L1 Cache	L2 Cache	L3 Cache	Memory Support	Memory Channels
Core i5-750	32 KB + 32 KB per core	256 KB per core	8 MB total	DDR3 up to 1333 MHz	Two
Core i7-870	32 KB + 32 KB per core	256 KB per core	8 MB total	DDR3 up to 1333 MHz	Two
Core i7-920	32 KB + 32 KB per core	256 KB per core	8 MB total	DDR3 up to 1066 MHz	Three
Core i7-965	32 KB + 32 KB per core	256 KB per core	8 MB total	DDR3 up to 1066 MHz	Three
Core 2 Extreme QX9770	32 KB + 32 KB per core	12 MB total	–	None	Depends on Chipset (Two is the norm)
Core 2 Quad Q6600	32 KB + 32 KB per core	8 MB total	–	None	Depends on Chipset (Two is the norm)
Core 2 Duo E8400	32 KB + 32 KB per core	6 MB total	–	None	Depends on Chipset (Two is the norm)

Our tests have a couple of known flaws. Socket LGA1366 Core i7 processors support triple-channel memory configuration, but we used them with dual-channel configuration, because otherwise we wouldn’t be able to match the amount of memory used with the other CPUs (e.g., to run them under triple channel we would need three 1 GB memory modules, making them to run with 3 GB RAM instead of 2 GB like the other CPUs). The second “problem” is that these CPUs can access memory only up to 1,066 MHz and with the other CPUs memory was working at 1,333 MHz. We could make all CPUs to access memory at 1,066 MHz, but then some people could complain that we didn’t run the new Core i5-750 and the new Core i7-870 at their maximum performance. It is hard to find a methodology that pleases everybody.

[nextpage title=”How We Tested”]

During our benchmarking sessions, we used the configuration listed below. Between our benchmarking sessions the only variable was the CPU being tested and the motherboard, which had to be replaced to match the different CPU sockets.

Hardware Configuration

• Motherboard (Socket LGA1156): Intel DP55KG Extreme (86A.3456 BIOS)
• Motherboard (Socket LGA1366): ASUS P6T Deluxe OC Palm Edition (1611 BIOS)
• Motherboard (Socket LGA775): EVGA nForce 790i Ultra SLI (P08 BIOS)
• CPU Cooler (Socket LGA1156): Intel stock
• CPU Cooler (Socket LGA1366): Thermalright Ultra-120 eXtreme 1366 RT
• CPU Cooler (Socket LGA775): Thermaltake TMG i1
• Memory: Two 1 GB Crucial CT12864BA1339 modules (DDR3-1333/PC2-10600, CL9, 1.5 V), configured at 1,333 MHz
• Hard Disk Drive: Western Digital Caviar Black 1 TB (WD1001FALS, SATA-300, 7,200 rpm, 32 MB buffer)
• Video Card: EVGA GeForce GTX 285 FTW
• Video Monitor: Samsung Syncmaster 305T
• Power Supply: OCZ EliteXStream 1000 W 
• Optical Drive: Lite-On LH-20A1L

Operating System Configuration

• Windows Vista Ultimate 32-bit
• Service Pack 2
• NTFS
• Video resolution: 2560×1600 @ 60 Hz

Driver Versions

• NVIDIA video driver version: 190.62
• Intel Inf chipset driver version: 9.1.1.1015
• NVIDIA nForce chipset driver version: 15.25
• Intel network driver version: 14.4
• Realtek audio driver version: 2.80A (6.0.1.5892)

Software Used

• PCMark Vantage Professional 1.1.0
• VirtualDub 1.9.5 + MPEG-2 Plugin 3.1 + DivX 6.8.5
• Adobe Photoshop CS4 Extended + GamingHeaven Photoshop Benchmark V3
• Adobe After Effects CS4
• WinRAR 3.90
• Cinebench 10
• 3DMark Vantage Professional 1.0.1
• Call of Duty 4 – Patch 1.7
• Fallout 3 – Patch 1.7

Error Margin

We adopted a 3% error margin; thus, differences below 3% cannot be considered relevant. In other words, products with a performance difference below 3% should be considered as having similar performance.

[nextpage title=”PCMark Vantage”]

PCMark Vantage simulates the use of real-world applications and gives scores for the following categories:

• PCMark
• Memories
• TV and Movies
• Gaming
• Music
• Communications
• Productivity
• HDD

For a detailed description of each one of these tests, please download and read the PCMark Vantage Reviewer’s Guide.

You can see the results for each category below. We are not going to compare the results for the Memories and HDD suites.



Core i5-750 achieved an overall PCMark score 4.85% higher than Core i7-920 (which runs at the same clock rate), 34.25% higher than Core 2 Duo E8400 and 36.08% higher than Core 2 Quad Q6600. Core i7-870 achieved a score similar to Core i7-965, being 4.40% higher than Core 2 Extreme QX 9770, 7.86% higher than Core i5-750, 13.09% higher than Core i7-920, 44.81% higher than Core 2 Duo E8400 and 46.78% higher than Core 2 Quad Q6600.



On the TV and Movies benchmark Core i5-750 achieved a performance 23.77% higher than Core 2 Quad Q6600 and 45.81% higher than Core 2 Duo E8400. Core i7-920, which runs at the same clock rate, was 7.42% faster. Core i7-870 achieved the same performance level as Core i7-920, being 5.73% faster than Core 2 Extreme QX 9770, 7.34% faster than Core i5-750, 32.85% faster than Core 2 Quad Q6600 and 56.50% faster than Core 2 Duo E8400.



On the Gaming set Core i5-750 achieve the same performance level as Core i7-920 and was 32.74% faster than Core 2 Extreme QX 9770, 57.32% faster than Core 2 Duo E8400 and 68.32% faster than Core 2 Quad Q6600. Core i7-870 was 5.14% faster than Core i7-920, 7.19% faster than Core i5-750, 42.29% faster than Core 2 Extreme QX 9770, 68.63% faster than Core 2 Duo E8400 and 80.42% faster than Core 2 Quad Q6600. Core i7-965 was 6.14% faster than Core i7-870.



On the Music benchmark Core i5-750 achieved the same performance level as Core i7-920, achieving a performance 9.17% higher than Core 2 Duo E8400 and 10.78% higher than Core 2 Quad Q6600. Core i7-870 achieved a score 3.40% higher than Core i5-750, 4.83% higher than Core i7-920, 12.89% higher than Core 2 Duo E8400 and 14.55% higher than Core 2 Quad Q6600.  Core i7-965 achieved a score 3.05% higher than Core i7-870 and Core 2 Extreme QX 9770 achieved a performance 16.27% higher than Core i7-870.



On the Communications tests Core i5-750 achieved a performance level 7.52% higher than the one achieved by Core i7-920, 20.06% higher than the one achieved by Core 2 Duo E8400 and 30.18% higher than the one achieved by Core 2 Quad Q6600. Core i7-870 achieved a performance 7.38% higher than Core i5-750, a similar performance to Core 2 Extreme QX 9770 but a performance a little bit lower (3.95%) than Core i7-965.



And finally on the Productivity benchmark Core i5-750 achieved a performance 5.54% better than Core i7-920, 19.50% better than Core 2 Duo E8400 and 20.42% better than Core 2 Quad Q6600. Core i7-870 achieved a performance similar to Core 2 Extreme QX 9770, which was 4.53% higher than the one achieved by Core i7-965, 6.06% higher than the one achieved by Core i5-750, 11.93% higher than the one achieved by Core i7-920, 26.74% higher than the one achieved by Core 2 Duo E8400 and 27.71% higher than the one achieved by Core 2 Quad Q6600.

[nextpage title=”VirtualDub + DivX”]

With VirtualDub we converted a full-length DVD movie to DivX format and saw how long it took for this conversion to be completed. The DivX codec is capable of recognizing and using not only more than one CPU (i.e., more than one core), but also the SSE4 instruction set.

The movie we chose to convert was Star Trek – The Motion Picture: Director’s Cut. We copied the movie to our hard disk drive with no compression, so the final original file on our HDD was 6.79 GB. After compressing it with DivX, the final file was only 767.40 MB, which is qui
te remarkable.

The results below are given in seconds, so the lower the better.



On DivX encoding Core i5-750, Core i7-870 and Core i7-920 achieved the same performance level. Core i7-965 was 12.74% faster than them and they were, on average, 7.65% faster than Core 2 Extreme QX 9770, 23.98% faster than Core 2 Duo E8400 and 37.98% faster than Core 2 Quad Q6600.

[nextpage title=”Photoshop CS4″]

The best way to measure performance is by using real programs. The problem, though, is creating a methodology using real software that provides accurate results. For Photoshop CS4, there is a methodology created by the folks at GamingHeaven that is very accurate. Their script applies a series of 15 filters to a sample image, and we wrote down the time taken for each filter to run. At the end, we have the results for each individual filter and we simply added them up to have the total time taken to run the 15 filters from the GamingHeaven batch. The results below are given in seconds, so the lower the number the better.



On Photoshop CS4 Core i5-750 was 4.65% faster than Core 2 Duo E8400 and 23.35% faster than Core 2 Quad Q6600.

Core i7-870 was 5.53% faster than Core i7-920 and 9.58% faster than Core i5-750.

Core i7-965 was 7.04% faster than Core i7-870 and Core 2 Extreme QX 9770 was 3.25% faster than Core i7-870.

Core i7-920, which runs at the same clock rate as Core i5-750, was 4.29% faster than this processor.

 

[nextpage title=”After Effects CS4″]

After Effects is a very well-known program for video post-production that is used to add animation and visual effects in videos. To evaluate the performance of each CPU running this program, we ran a workload consisting of 25 compositions that applied several filters and effects to a variety of input file types such as PSD (Photoshop), AI (Illustrator), EPS, and TIF. After each filter was applied, the composition was rendered to an uncompressed AVI file with the same resolution as the input files. The results below are the time each CPU took to finish the whole batch, given in seconds, so the lower the number the better.



On After Effects CS4 Core i5-750 was 30.99% faster than Core 2 Quad Q6600 and 47.82% faster than Core 2 Duo E8400.

Core i7-870 was 6.14% faster than Core i7-920, 16.20% faster than Core 2 Extreme QX 9770, 21.05% faster than Core i5-750, 45.52% faster than Core 2 Quad Q6600 and 58.80% faster than Core 2 Duo E8400.

Core i7-920 was 15.88% faster than Core i5-750, and they run at the same clock rate.

Core i7-965 was the fastest CPU here, being 7.43% faster than Core i7-870 and 26.92% faster than Core i5-750.[nextpage title=”WinRAR”]

We measured the time each CPU took to compress five high-resolution 48-bit uncompressed TIF images, each one with around 70 MB, to RAR format with the popular WinRAR application. The results below are given in seconds, so the lower the number the better.



Core i5-750 was 6.87% faster than Core 2 Extreme QX 9770, 27.29% faster than Core 2 Duo E8400 and 31.65% faster than Core 2 Quad Q6600. Core i7-920 was 14.81% faster than Core i5-750 (and both run at the same clock rate!), Core i7-870 was 20.53% faster than Core i5-750 and Core i7-965 was 23.94% faster than Core i5-750. Core i7-965 was 4.29% faster than Core i7-870 and Core i7-870 was 6.71% faster than Core i7-920.

[nextpage title=”Cinebench 10″]Cinebench 10 is based on the 3D software, Cinema 4d. It is very useful to measure the performance gain given by having more than one CPU installed on the system when rendering heavy 3D images. Rendering is one area in which having more than one CPU helps considerably, because usually, rendering software recognizes several CPUs. (Cinebench, for instance, can use up to 16 CPUs.)

Since we were interested in measuring the rendering performance, we ran the test called “Rendering x CPUs,” which renders a “heavy” sample image using all available CPUs (or cores – either real or virtual, as on CPUs with Hyper-Threading technology, each core is recognized as two cores by the operating system) to speed up the process.



On Cinebench we could see a big difference in performance between Core i5-750 and Core 2 Quad Q6600 (36.79%) and Core 2 Duo E8400 (88.05%). Core 2 Extreme QX 9770 was still faster than Core i5-750 (18.01%), probably due its higher clock rate. The new Core i7-870 was 32.14% faster than Core i5-750 and a little bit faster (6.36%) than Core i7-920, probably due to its higher clock.  Core i7-965 was 8.07% faster than Core i7-870, probably due to its higher clock. Core i7-920, which runs at the same clock rate as Core i5-750, was 24.24% faster than this processor.

[nextpage title=”3DMark Vantage Professional”]

3DMark Vantage measures Shader 4.0 (i.e., DirectX 10) gaming performance. We ran this program under the “Extreme” profile, which sets resolution to 1920×1200, Anisotropic filtering to 16x and max out all quality settings to the maximum allowed. We are going to analyze two results. The 3DMark score and the CPU score. Currently with so much processing in 3D games being migrated from the CPU to the GPU (graphics processing unit; the graphics processor present on the video card), replacing the CPU with a more powerful one doesn’t impact gaming performance as much as it used to happen several years ago.



Under the “Extreme” profile from 3DMark Vantage all CPUs achieved the same performance level. This shows how for high-end gaming the video card is the component that most influences performance, not the CPU, as mentioned above.



We could see some performance differences comparing exclusively the CPU benchmark results from 3DM
ark Vantage. Core i5-750, Core i7-920, Core 2 Extreme QX 9770 and Core i7-870 achieved the same performance level here. Core i5-750 achieved a performance 26.28% higher than Core 2 Duo E8400 and 28.44% higher than Core 2 Quad Q600 here. Core i7-965 achieved a performance 8.32% higher than Core i5-750 and 10.21% higher than Core i7-870.

[nextpage title=”Call of Duty 4″]Call of Duty 4 is a DirectX 9 game implementing high-dynamic range (HDR) and its own physics engine, which is used to calculate how objects interact. For example, if you shoot, what exactly will happen to the object when the bullet hits it? Will it break? Will it move? Will the bullet bounce back? It gives a more realistic experience to the user.

We ran this game under 2560×1600, maxing out all image quality controls (i.e., everything was put on the maximum values on the Graphics and Texture menus). We used the game internal benchmarking feature, running a demo provided by NVIDIA called “wetwork.” We are putting this demo for downloading here if you want to run your own benchmarks. The game was updated to version 1.7. We ran this test five times, discarding the lowest and the highest scores. The results below are an arithmetic average of the three remaining values, given in frames per second (FPS).



The performance for most CPUs was within the same level, with the exception of Core 2 Quad Q6600, which achieved a performance a little bit lower (6%) than the other CPUs tested. This shows us that at least for this game picking the video card is more critical than picking the CPU.

[nextpage title=”Fallout 3″]

Fallout 3 is based on the same engine used by The Elder Scrolls IV: Oblivion, and it is a DirectX 9.0c (Shader 3.0) game. We configured the game with “ultra” image quality settings, maxing out all image quality settings, at 2560×1600. To measure performance, we used the FRAPS utility running an outdoor scene at God mode, running through enemy fire, triggering post processing effects, and ending with a big explosion in front of Dupont Circle.



On Fallout 3 Core i5-750 achieved the same performance level as Core i7-870 and Core 2 Extreme QX 9770, being 7.71% faster than Core 2 Duo E8400 and 22.65% faster than Core 2 Quad Q6600.

Core i7-965 was 3.40% faster than Core i7-870 and 5.61% faster than Core i5-750. Core i7-920 achieved the same performance level as Core i7-870, being 3.25% faster than Core i5-750.

Core i7-870 was 3.44% faster than Core 2 Extreme QX 9770, 10.01% faster than Core 2 Duo E8400 and 25.27% faster than Core 2 Quad Q6600.

Core i7-920 and Core i5-750, which run at the same clock rate, achieved a practically similar performance level (Core i7-920 was 3.25% faster).

[nextpage title=”Conclusions”]

The new Core i5 is definitely a major step for the user that likes mid-range CPUs. Core i5-750 is going to arrive on the same price range of current mainstream CPUs but with a far higher performance for most applications. This is great news. The not so great news is that you will need a new motherboard to use it, so you won’t be able to upgrade your PC by just replacing the CPU: you will need to replace the motherboard and possible your memory modules as well, if you are not using DDR3 modules at the moment.

Core i7 will continue to be the CPU of choice for the professional user working with rendering and image processing and the new socket LGA1156 promises to give far more options of CPU and motherboard models to choose from in the future.

For the average user, Core i5 provides a better cost/benefit ratio than Core i7. It is true that Core i7 is faster than Core i5 even when we compare a Core i7 and a Core i5 running at the same clock rate as we did in our review, thanks to the Hyper-Threading technology that makes programs to “see” eight CPU cores instead of “only” four, but this difference in performance will be only worthwhile to professionals because a faster processing means more work done in less time, and thus more jobs for their clients and hence more money. We don’t see why a regular user would pay at least 45% more on a CPU that won’t give that much performance increase in return.

And this also applies for gamers. Gaming performance nowadays depends more on the video card than on the CPU – at least with the latest titles maximizing video quality. In summary, you will be better off buying a cheaper CPU and investing the difference on a more expensive video card than trying to have the most expensive CPU that you can afford. Thus if you are a gamer and want to have the latest Intel CPU, our recommendation falls on Core i5 once again. Unless, of course, you have a bottomless pocket and want to have the fastest PC in town: in this case buy a Core i7-975 with 12 GB RAM and two SSD’s in RAID0 and be happy!