[nextpage title=”Introduction”]
Intel is launching this month their new CPUs based on the “Sandy Bridge” architecture. Let’s check the performance of the new Core i5-2500K (3.3 GHz) and compare it to its main competitor from AMD, the new Phenom II X4 975 Black Edition (3.6 GHz) that is being released today. We also added the Core i5-661 in the mix.
For a detailed explanation of what is new in the Sandy Bridge architecture, please read our Inside the Intel Sandy Bridge Microarchitecture tutorial.
The new Core i5 2500K (3.3 GHz) is a quad-core CPU, coming in two flavors: with its clock multiplier unlocked (“K,” being the equivalent to the “Extreme Edition” CPUs Intel used to carry and to the “Black Edition” CPUs from AMD), giving you an extra way to overclock the CPU, and the standard model with a locked clock multiplier. The “K” model comes with a price tag of USD 216, while the standard model comes priced at USD 205.
The new quad-core Phenom II X4 975 Black Edition comes with a price tag of USD 195.99, being the main competitor to the Core i5-2500, and in this review we will be able to answer the most basic question of all: which one is the fastest?
We also included in our review the Core i5-661, which runs at almost the same clock rate (3.33 GHz) and same price tag as the Core i5-2500. Keep in mind that the Core i5-661 is a dual-core CPU with Hyper-Threading technology, while the Core i5-2500K is a quad-core CPU without Hyper-Threading technology. While both are recognized by the operating system as having “four” cores, on the Core i5-2500K these cores are real, while on the Core i5-661 two are real and two are “simulated.”
We also added the Phenom II X4 970 to our mix, because until yesterday it was the CPU from AMD with the highest clock rate (3.5 GHz), and although it is a little bit cheaper (USD 185.99) than the other CPUs we included, we think users are also curious to see how is the new Phenom II X4 975 compared to the “old” Phenom II X4 970.
In the tables below you can see a brief comparison between the CPUs we included in our review. AMD CPUs do not support SSE4 instructions (they have a proprietary instruction set called SSE4a, which is not the same thing as SSE4).
CPU | Cores | HT | IGP | Internal Clock | Turbo Clock | Base Clock | Core | Technology | TDP | Socket | Price |
Core i5-2500K | 4 | No | Yes | 3.30 GHz | 3.7 GHz | 100 MHz | Sandy Bridge | 32 nm | 95 W | 1155 | USD 216 |
Core i5-661 | 2 | Yes | Yes | 3.33 GHz | 3.6 GHz | 133 MHz | Clarkdale | 32 nm | 87 W | 1156 | USD 210 |
Phenom II X4 975 | 4 | No | No | 3.6 GHz | – | 200 MHz | Deneb | 45 nm | 125 W | AM3 | USD 196 |
Phenom II X4 970 | 4 | No | No | 3.5 GHz | – | 200 MHz | Deneb | 45 nm | 125 W | AM3 | USD 186 |
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.
CPU | L1 Cache | L2 Cache | L3 Cache | Memory Support | Memory Channels |
Core i5-2500K | 32 KB + 32 KB per core | 256 KB per core | 6 MB total | DDR3 up to 1333 MHz | Two |
Core i5-661 | 32 KB + 32 KB per core | 256 KB per core | 4 MB total | DDR3 up to 1333 MHz | Two |
Phenom II X4 975 | 64 KB + 64 KB per core | 512 KB per core | 6 MB total | DDR3 up to 1333 MHz | Two |
Phenom II X4 970 | 64 KB + 64 KB per core | 512 KB per core | 6 MB total | DDR3 up to 1333 MHz | Two |
While all CPUs listed above have an integrated memory controller, only the two Intel CPUs listed have an integrated graphics processor (IGP) and an integrated PCI Express 2.0 controller, handling 16 PCI Express lanes, allowing those CPUs to drive one PCI Express slot at x16 or two PCI Express slots at x8.
Since the Intel CPUs included in our review have an integrated graphics controller, we had to test this aspect of the CPU. We installed the Core i5-661 on an Intel DH55TC motherboard (Intel H55 chipset, USD 90) and the Core i5-2500K on an Intel DH67BL motherboard (Intel H67 chipset, USD 107). Because of that, we installed the AMD CPUs on a motherboard with integrated video on the same price range that we had handy (ASRock 880GXH/USB3, AMD 880G chipset, USD 115).
AMD CPUs talk to the external world (i.e. the chipset) thru a bus called HyperTransport. For a detailed explanation how this bus works, please read our The HyperTransport Bus Used by AMD Processors tutorial.
Socket LGA1156 and 1155 CPUs 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.
One final note. The Core i5-2500K has a base clock of 100 MHz instead of using a 133 MHz base clock as the previous Core i5 generation. This means that it has to multiply 100 MHz by 33 to get to its internal 3.30 GHz. The Core i5-661, on the other hand, has to multiply 133 MHz by 25 to get to its internal 3.33 GHz.
[nextpage title=”How We Tested”]
During our benchmarking sessions we used the configuration listed below. Between our benchmarking sessions the only variable device was the CPU being tested and the motherboard, which had to be replaced to match the different CPU sockets.
Hardware Configuration
- Motherboard (Socket LGA1155): Intel DH67BL (0082 BIOS)
- Motherboard (Socket LGA1156): Intel DH55TC (0040 BIOS)
- Motherboard (Socket AM3): ASRock 880GXH/USB3 (1.20 BIOS)
- CPU Cooler: Intel/AMD stock
- Memory: 4 GB DDR3-1333, two G.Skill F3-10666CL7T memory modules
- Hard Disk Drive: Western Digital Caviar Black 1 TB (WD1001FALS, SATA-300, 7,200 rpm, 32 MB buffer)
- Video Card: GeForce GT 430 (used on some tests only)
- Video Monitor: Samsung Syncmaster 932BW
- Power Supply: OCZ StealthXStream 400 W
Operating System Configuration
- Windows 7 Ultimate 64-bit
- NTFS
- Video resolution: 1440×900 60 Hz
Driver Versions
- NVIDIA video driver version: 260.99
- Intel video driver version (Core i5-2500K): 8.15.10.2266
- Intel video driver version (Core i5-661): 15.17.10.2189
- AMD video driver version: 8.71
- Intel Inf chipset driver version (Intel DH67BL): 9.2.0.1019
- Intel Inf chipset driver version (Intel DH55TC): 9.1.2.1008
- AMD chipset driver version: 8.631
Software Used
- PCMark Vantage Professional 1.0.2
- VirtualDub 1.9.5 + MPEG-2 Plugin 3.1 + DivX 6.9.2
- Adobe Photoshop CS4 Extended + GamingHeaven Photoshop Benchmark V3
- Adobe After Effects CS5
- WinZip 15.0
- iTunes 10
- Cinebench 11.5
- Call of Duty 4 – Patch 1.7
- Starcraft II: Wings of Liberty
- Far Cry 2 – Patch 1.03
- Lost Planet 2
- 3DMark 11 1.0.1.0
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.
On the PCMark overall score the Core i5-2500K (3.30 GHz) was 25% faster than the Core i5-661 (3.33 GHz), 41% faster than the Phenom II X4 975 (3.6 GHz), and 54% faster than the Phenom II X4 970 (3.5 GHz).
The Core i5-2500K (3.30 GHz) achieved a TV and Movies score 28% higher than the one achieved by the Phenom II X4 975 (3.6 GHz) and the Phenom II X4 970 (3.5 GHz), and 31% higher than the one achieved by the Core i5-661 (3.33 GHz).
On the gaming benchmark the Core i5-2500K (3.30 GHz) was 51% faster than the Core i5-661 (3.33 GHz), 83% faster than the Phenom II X4 975 (3.6 GHz), and 87% faster than the Phenom II 970 (3.5 GHz). Keep in mind that with the CPU from AMD this score is actually measuring the chipset (AMD 880G) graphics performance, not the CPU performance.
On the Music benchmark the Core i5-2500K (3.30 GHz) was again the fastest CPU, with a score 21% higher than the one achieved by the Phenom II X4 970 (3.5 GHz), 31% higher than the one achieved by the Phenom II X4 975 (3.6 GHz), and 32% higher than the one achieved by the Core i5-661 (3.33 GHz).
On Communications, the Core i5-2500K (3.30 GHz) was 18% faster than the Core i5-661 (3.33 GHz), 64% faster than the Phenom II X4 975 (3.6 GHz), and 70% faster than the Phenom II X4 970 (3.5 GHz).
On Productivity, the Core i5-2500K (3.30 GHz) achieved a score 19% higher than the one obtained by the Phenom II X4 975 (3.6 GHz), 25% higher than the one achieved by the Core i5-661 (3.33 GHz), and 28% higher than the one achieved by the Phenom II X4 970 (3.5 GHz). [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 (feature not available on the reviewed CPUs).
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 quite remarkable.
The results below are given in seconds, so the lower the better.
On DivX encoding the Core i5-2500K (3.30 GHz) was 45% faster than the Core i5-661 (3.33 GHz), 50% faster than the Phenom II X4 975 (3.6 GHz), and 55% faster than the Phenom II X4 970 (3.5 GHz).
[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, the Core i5-2500K (3.30 GHz) was 19% faster than the Core i5-661 (3.33 GHz), 36% faster than the Phenom II X4 975 (3.6 GHz), and 44% faster than the Phenom II X4 970 (3.5 GHz).
[nextpage title=”After Effects CS5″]
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 several 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 CS5, the Core i5-2500K (3.30 GHz) was 48% faster than the Phenom X4 975 (3.6 GHz), 51% faster than the Phenom II X4 970 (3.5 GHz), and 68% faster than the Core i5-661 (3.33 GHz).
[nextpage title=”WinZip”]
We used WinZip not only to measure compression time, but also decryption time. We measured the time each CPU took to decompress and decrypt 200 JPEG images, 125 of them at 10 megapixels and 75 of them at six megapixels. The total size of all the photos was around 830 MB. The results below are given in seconds, so the lower the number the better.
Decompressing and decrypting files, the Core i5-2500K (3.30 GHz) was, again, the fastest CPU: 18% faster than the Core i5-661 (3.33 GHz), 23% faster than the Phenom II X4 970 (3.5 GHz), and 32% faster than the Phenom II X4 975 (3.6 GHz).
[nextpage title=”iTunes”]
We used iTunes to convert an uncompressed .wav file into a high-quality (160 Kbps) MP3 file, and checked how many seconds each CPU took to perform this operation. Therefore, the results below are given in seconds, so the lower the number the better.
The Core i5-2500K was the fastest CPU for MP3 converting, being 36% faster than the Phenom II X4 975 (3.6 GHz) and 38% faster than the Core i5-661 (3.33 GHz) and the Phenom II X4 970 (3.5 GHz).
[nextpage title=”Cinebench 11.5″]
Cinebench 11.5 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, the Core i5-2500K (3.30 GHz) achieved the highest score, 29% higher than the one achieved by the Phenom II X4 975 (3.6 GHz), 32% higher than the one achieved by the Phenom II X4 970 (3.5 GHz), and 88% higher than the one achieved by the Core i5-661 (3.33 GHz).
[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. We ran this game at 1440×900 with all image quality settings at their minimum values (no anti-aliasing and no anisotropic filtering). 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 benchmarking. 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).
We ran this game twice with each CPU. First, we used the CPU or chipset integrated graphics. Then we added a GeForce GT 430 video card, which is an entry-level DirectX 11 video card. We used an entry-level video card because we wanted to see the impact each CPU had in the performance achieved (when using high-end video cards, the CPU role in gaming performance is reduced).
Let’s compare the results of the CPU and chipset integrated graphics first. The integrated graphics processor of the Core i5-2500K (3.30 GHz) proved to be 65% faster than the one embedded in the Core i5-661 (3.33 GHz), 82-84% faster than the one used in the AMD 880G chipset. In fact, lowering all image quality settings we finally have a playable frame rate on Call of Duty 4.
When we installed a GeForce GT 430, the Core i5-2500K (3.30 GHz) and the Phenom II X4 975 (3.6 GHz) achieved the same performance level, with the Core i5-2500K being 8% faster than the Phenom II X4 970 (3.5 GHz) and 12% faster than the Core i5-661 (3.33 GHz).
[nextpage title=”Starcraft II: Wings of Liberty”]
StarCraft II: Wings of Liberty is a very popular DirectX 9 game that was released in 2010. Though this game uses an old version of DirectX, the number of textures that can be represented on one screen can push most of the top-end graphics cards to their limits (especially when the graphics settings are set at “Ultra”). StarCraft II: Wings of Liberty uses its own physics engine that is bound to the CPU and thus does not benefit from PhysX.
We tested this game at 1440×900. The quality of the game was set to the “low” preset, disabling both anti-aliasing and anisotropic filtering. We then used FRAPS to collect the frame rate of a replay on the “Unit Testing” custom map. We used a battle between very large armies to stress the video cards.
We ran this game twice with each CPU. First, we used the CPU or chipset integrated graphics. Then we added a GeForce GT 430 video card, which is an entry-level DirectX 11 video card. We used an entry-level video card because we wanted to see the impact each CPU had in the performance achieved (when using high-end video cards, the CPU role in gaming perfo
rmance is reduced).
Let’s compare the results of the CPU and chipset integrated graphics first. The integrated graphics processor of the Core i5-2500K (3.30 GHz) proved to be 104% faster than the one embedded in the Core i5-661 (3.33 GHz), which is outstanding. It was 86-90% faster than the one used in the AMD 880G chipset. In fact, lowering all image quality settings we finally have a playable frame rate on StarCraft II.
When we installed a GeForce GT 430, the Phenom II X4 975 (3.6 GHz) proved to be the fastest CPU, being 6% faster than the Phenom II X4 970 (3.5 GHz), 7% faster than the Core i5-2500K (3.30 GHz), and 9% faster than the Core i5-661 (3.33 GHz).
[nextpage title=”Far Cry 2″]
Far Cry 2 is based on an entire new game engine called Dunia, which is based on DirectX 10 when played under Windows 7 or Windows Vista with a DirectX 10-compatible video card. We used the benchmarking utility that comes with this game, setting video resolution to 1440 x 900, overall image quality to “high” (this is the lowest possible setting if you want to run this game at DirectX 10), setting all “Performance” options to “low,” disabling both anti-aliasing and anisotropic filtering, and running the demo “Ranch Long.” The results below are expressed in frames per second.
We ran this game twice with each CPU. First, we used the CPU or chipset integrated graphics. Then we added a GeForce GT 430 video card, which is an entry-level DirectX 11 video card. We used an entry-level video card because we wanted to see the impact each CPU had in the performance achieved (when using high-end video cards, the CPU role in gaming performance is reduced).
Let’s compare the results of the CPU and chipset integrated graphics first. The integrated graphics processor of the Core i5-2500K (3.30 GHz) proved to be 90% faster than the one embedded in the Core i5-661 (3.33 GHz). It was 75-76% faster than the one used in the AMD 880G chipset. However, don’t keep high expectations, because you won’t be able to play Far Cry 2 with any integrated video solution.
When we installed a GeForce GT 430, all four CPUs achieved the same performance level.
[nextpage title=”Lost Planet 2″]
Lost Planet 2 is a game that uses a lot of DirectX 11 features, like tessellation (to round out the edges of polygonal models), displacement maps (added to the tessellated mesh to add fine grain details), DirectCompute soft body simulation (to introduce more realism in the “boss” monsters), and DirectCompute wave simulation (to introduce more realism in the physics calculations in water surfaces; when you move or when gunshots and explosions hit the water, it moves accordingly). We reviewed the CPUs using Lost Planet 2 internal benchmarking features, choosing the “Benchmark A” (we know that “Benchmark B” is the one recommended for reviewing video cards, however, at least with us, results were inconsistent). We ran this game at 1440×900 with graphics set at “low,” with no anti-aliasing and no anisotropic filtering. The results below are the number of frames per second generated by each system.
Since the integrated graphics engine of the Intel CPUs and AMD 880G chipset aren’t DirectX 11, we had to run this game in DirectX 9 mode in order to compare the performance of the integrated video. Under this scenario, the Core i5-2500K (3.30 GHz) was the fastest CPU, being 72% faster than the Core i5-661 (3.33 GHz) and 76%-79% faster than the integrated video produced by the AMD 880G chipset.
Installing a GeForce GT 430 video card, we were able to run this game in DirectX 11 mode. Here the Phenom II X4 975 (3.6 GHz) was 12% faster than the Core i5-2500K (3.30 GHz), which achieved the same performance level of the Core i5-661 (3.33 GHz).
[nextpage title=”3DMark 11 Professional”]
3DMark 11 Professional measures Shader 5.0 (i.e., DirectX 11) performance. Since the integrated graphics engine of the Intel CPUs and AMD 880G chipset aren’t DirectX 11, the only way to run this game was by installing an add-on video card. We used a GeForce GT 430 video card, which is an entry-level DirectX 11 video card. We ran this program at 1440×900 using the “Performance” profile.
This program provides three different scores: graphics, physics and combined.
All CPUs achieved comparable graphics scores. This was expected, since this number measures exclusively the video card performance, not taking the CPU into account.
The physics score measures exclusively the physics performance of the system, in our case the CPU performance, as the video card was always the same. Here the Core i5-2500K (3.30 GHz) was 60% faster than the Phenom II X4 975 (3.6 GHz), 63% faster than the Phenom II X4 970 (3.5 GHz), and 79% faster than the Core i5-661 (3.33 GHz).
The combined score shows a balance between the graphics and the physics performance achieved by each system being tested. Here the Phenom II X4 970 (3.5 GHz) and the Phenom II X4 975 (3.6 GHz) achieved the same performance, which was 5% higher than Core i5-2500K’s.
[nextpage title=”Overclocking”]
We did some overclocking with both CPUs. Since they come with their clock multiplier unlocked, you can achieve very high overclocking levels with both of them. In both cases we used a different motherboard, as the boards we were using don’t provide good overclocking options. All the other components were the same, using a GeForce GT 430 video card. We ran Far Cry 2 to test the system stability (i.e., to make sure the system wasn’t crashing).
We used an ASUS Crosshair IV Formula motherboard (1102 BIOS) with the Phenom II X4 975 (3.6 GHz). Without playing with any advanced options, we could increase the clock multiplier of this CPU from 18x to 20x, and increase the base clock up to 206 MHz, making this CPU to run internally at 4.12 GHz. Then we decided to play with voltages a little bit. Increasing the CPU main voltage from 1.4020 V to 1.5375 V made us able to increase the base clock up to 213 MHz, making our CPU to run internally at 4.26 GHz, a good 18.3% overclocking.
To overclock the Core i5-2500K (3.30 GHz) we used an MSI P67A-GD65 motherboard (1.0 BIOS). Without playing with any advanced options, we could increase the clock multiplier of this CPU from x33 to x42, and increase the base clock up to 110 MHz, making this CPU to run internally at 4.62 GHz. Then we decided to play with voltages. We increased the CPU main voltage from 1.208 V to 1.400 V, and we were able to increase the clock multiplier even further to x44, keeping the base clock at 110 MHz. This made our CPU to run stable at 4.84 GHz, a 46.7% increase in the CPU internal clock rate.
It is very important to remember that we got an engineering sample from Intel, which may have a higher overclocking capability than the final product that will arrive on the market.
We didn’t play a lot with all overclocking options, and you may achieve even better results.
[nextpage title=”Conclusions”]
The new Core i5-2500K simply crushes its main rival, the Phenom II X4 975 (3.6 GHz), in almost all applications. In some games with an entry-level add-on video card installed, however, the Phenom II X4 975 was faster than the Core i5-2500K (it is important to remember that with high-end video cards the CPU usually doesn’t play a significant role in performance). The Core i5-2500K is also way faster than the Core i5-661, partly because of the new microarchitecture, partly because it is a true quad-core CPU (the Core i5-661 is a dual-core CPU with Hyper-Threading technology). The truth is that AMD is very delayed with the launch of the Phenom II X4 975, a CPU that should be released months ago.
Intel was saying that the integrated video of “Sandy Bridge” CPUs would be twice as fast as the integrated video of “Clarkdale” CPUs. Giving the right circumstances, they were right. In fact, this is the first time that you can play DirectX 9 games with an integrated video solution at a decent frame rate, if you minimize all graphics settings. With the DirectX 10 game we ran (Far Cry 2), the integrated video of the Core i5-2500K achieved very low frame rate, even with all image quality settings at their lowest values, meaning that you won’t be able to play this game without installing a “real” video card.
There is still one important question to be considered. Who buys a USD 200 CPU to run integrated video? Usually people that need processing power but won’t be playing games anyway.
We were satisfied with the overclocking capability of the Phenom II X4 975: we could easily put it to run at 4.26 GHz, an 18.3% increase above the CPU’s standard clock, which is a very good mark for an AMD CPU.
Now, the new Core i5-2500K creates a new chapter in overclocking history. We could put it to run at 4.84 GHz, a 46.7% increase in the CPU internal clock rate, which is unheard of – and keep in mind that we are not into overclocking! Reviewers from other websites could put this CPU to run at 5.1 GHz! It is very important to keep in mind that we got an engineering sample, and the final product may not have such a fantastic overclocking potential. Only time will tell.
In summary, the new Core i5-2500 is the best CPU today in the USD 200 range, bringing an impressive performance gain over the competition and the previous-generation Core i5. If you like overclocking, you should spend USD 11 more and buy the “K” version, which has the best overclocking potential of any CPU we’ve ever reviewed.
Leave a Reply