[nextpage title=”Introduction”]
Last year, Intel released its new high-end desktop platform, the X99, which uses the new LGA2011-v3 socket. The CPUs launched for this platform are codenamed “Haswell-E,” and the most high-end CPU lauched for this platform is the Core i7-5960X, which has eight cores (16 threads thanks to the Hyper-Threading technology), 3.0 GHz clock (3.5 GHz maximum clock, through Turbo Boost technology), and the support for quad-channel DDR4 memories. Let’s benchmark this beast and see how well it performs.
The new X99 platform comes to replace the X79 platform, and the main difference between the two is the support to DDR4 memory instead of DDR3. (Both support quad-channel memory access.) This platform uses the new LGA2011-v3 socket, which is not compatible with LGA2011 CPUs from the X79 platform. LGA2011-v3 CPUs can have 40 or 28 PCI Express 3.0 lanes, and the exact way they are connected to the PCI Express slot and devices will depend on the motherboard. Since it is targeted to the high-end user, processors from the X99 platform don’t come with an embedded graphics adapter.
The Core i7-5960X is the most high-end CPU from this family and, so far, the most high-end desktop processor available on the market. It has 40 PCI Express 3.0 lanes, base clock of 3.0 GHz with turbo clock up to 3.5 GHz, 140 W TDP, and supports DDR4 memory modules up to 2,133 MHz, in quad-channel configuration. It supports SSE 4.2, AVX 2.0, and AES instruction sets. The Core i7-5960X has eight processing cores, but thanks to the Hyper-Threading technology, it is seen as a 16-core processor by the operating system.
Figures 1 and 2 unveil the Core i7-5960X CPU.
Figure 1: the Core i7-5960X processor
Figure 2: the Core i7-5960X processor
One of the challenges we have in reviewing this CPU is that there is no direct competitor for it: the most expensive processor from AMD, the FX-9590, costs a quarter of its price. With that in mind, we included an FX-9590 CPU in our review anyway, since users usually want to know what is the performance difference between the most high-end CPU from Intel versus the most high-end CPU from AMD. We also included in our review a Core i7-4770K, one of the most high-end processors from Intel for its mainstream platform, the LGA1150, so users can see what is the performance gain they can expect from picking the most expensive CPU available today.
Figure 3: the Core i7-5960X (left), the FX-9590 (center), and the Core i7-4770K (right)
Let’s compare the main specs of the tested CPUs in the next page.
[nextpage title=”The Reviewed CPUs”]
In the tables below, we compare the main features of the CPUs included in our review.
CPU | Cores | HT | IGP | Internal Clock | Turbo Clock | Base Clock | Core | Tech. | TDP | Socket | Price |
Core i7-5960X | 8 | Yes | No | 3.0 GHz | 3.5 GHz | 100 MHz | Haswell-E | 22 nm | 140 W | LGA2011-v3 | USD 1050 |
Core i7-4770K | 4 | Yes | Yes | 3.5 GHz | 3.9 GHz | 100 MHz | Haswell | 22 nm | 84 W | LGA1150 | USD 345 |
FX-9590 | 8 | No | No | 4.7 GHz | 5.0 GHz | 200 MHz | Vishera | 32 nm | 220 W | AM3+ | USD 240 |
Prices were researched at Newegg.com on the day we published this review. TDP stands for Thermal Design Power and states the maximum amount of heat the CPU can dissipate.
Below you can see the memory configuration for each CPU.
CPU | L2 Cache | L3 Cache | Memory Support | Memory Channels |
Core i7-5960X | 8 x 256 kiB | 20 MiB | Up to DDR4-2133 | Four |
Core i7-4770K | 4 x 256 kiB | 8 MiB | Up to DDR3-1600 | Two |
FX-9590 | 4 x 2 MiB | 8 MiB | Up to DDR3-1866 | Two |
[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. The memory also has to be replaced, since the Core i7-5960X uses DDR4 memory, while the other two CPUs use DDR3.
Hardware Configuration
- Motherboard (socket LGA2011-v3): ASRock Fatal1ty X99M Killer
- Motherboard (socket LGA1150): ASRock Z97 Extreme4
- Motherboard (socket AM3+): ASRock Fatal1ty 990FX Killer
- CPU Cooler: SilverStone TD02
- Memory (DDR3): 16 GiB DDR3 G.Skill Sniper 1866, two F3-1866C10D-16GSR 8 GiB memory modules configured at 1866 MHz
- Memory (DDR4): 16 GiB DDR4 G.Skill Ripjaws 4, four F4-2400C15Q-16GRR 4 GiB memory modules configured at 2400 MHz
- Boot drive: Kingston HyperX Fury 240 GB
- Video Card: Gigabyte GeForce GTX 770 (GV-N770OC-2GD)
- Video Monitor: Philips 236VL
- Power Supply: Corsair CX750
Operating System Configuration
- Windows 7 Home Premium 64-bit
- NTFS
- Video resolution: 1920 x 1080 60 Hz (desktop), 1280 x 720 (games)
Driver Versions
- AMD driver version: 14.12
- Intel Inf chipset driver version: 10.0
- nVIDIA GeForce driver version: 347.25
Software Used
- 3DMark 1.4.828
- Cinebench R15
- DivX 10.2.4
- DVD Shrink 3.2
- Media Espresso 6.7
- PCMark 8 2.2.282
- Assassin’s Creed IV: Black Flag
- Battlefield 4
- Call of Duty: Advanced Warfare
- F1 2013
- Far Cry 4
- Metro Last Light
Error Margin
We adopted a 4% error margin. Thus, differences below 4% cannot be considered relevant. In other words, products with a performance difference below 4% should be considered as having similar performance.
[nextpage title=”PCMark 8″]
PCMark 8 is a benchmarking software that uses real-world applications to measure the computer performance. We ran three tests: Home, which includes web browsing, writing, light gaming, photo editing, and video chat tests; Creative, which includes web browsing, photo editing, video editing, group video chat, media transcoding, and gaming; and Work, which runs tasks such as writing documents, web browsing, spreadsheets, editing, and video chatting. Let’s analyze the results.
The Core i7-5960X achieved a score 8.6% higher than the FX-9590 and 6.3% lower than the Core i7-4770K in the Home test. It seems PCMark doesn’t make use of all sixteen threads available.
On the Creative benchmark, the Core i7-5960X achieved a score 17% higher than the FX-9590, but 5.8% lower than the Core i7-4770K. It seems PCMark doesn’t make use of all sixteen threads available.
On the Work benchmark, the Core i7-5960X achieved the same performance level of the FX-9590 and was 11% slower than the Core i7-4770K. It seems PCMark doesn’t make use of all sixteen threads available.
[nextpage title=”3DMark”]
3DMark is a program with a set of benchmarks that create 3D scenarios and simulations.
The 3DMark Fire Strike benchmark measures DirectX 11 performance, and is aimed at high-end gamer PCs. The Sky Diver benchmark also measures DirectX 11 performance, but is more suitable to mainstream computers. The 3DMark Cloud Gate benchmark measures DirectX 10 performance, running at 1280×720 resolution. The 3DMark Ice Storm benchmark measures DirectX 9 performance and it is aimed at low-end computers, so we did not include it in our tests.
On the Fire Strike benchmark, the Core i7-5960X was not significantly faster than the Core i7-4770K, but it was 9% faster than the FX-9590.
On the Sky Diver benchmark, the Core i7-5960X was 15% faster than the Core i7-4770K, and 26% faster than the FX-9590.
On the Cloud Gate benchmark, the Core i7-5960X was 35% faster than the Core i7-4770K, and 57% faster than the FX-9590.
[nextpage title=”Cinebench R15″]
Cinebench R15 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 256 CPUs.)
Since we were interested in measuring the rendering performance, we ran the CPU test, which renders a “heavy” sample image using all available CPUs or “cores” – either real or virtual – to speed up the process. The resut is given as a score.
Here the Core i7-5960X really takes benefit from its sixteen threads: it was 79% faster than the Core i7-4770K and 90% faster than the FX-9590.
[nextpage title=”DivX”]
We used the DivX converter, a tool included in the DivX package, in order to measure the encoding performance using this codec. The DivX codec is, at least theorethically, capable of recognizing and using all available cores and the SSE4 instruction set.
We converted a Full HD, six-minute long .mov video file into a .avi file, using the “HD 1080p” output profile.
The results below are given in seconds, so the lower the better.
On DivX encoding, the Core i7-5960X was 35% faster than the FX-9590, but 6% slower than the Core i7-4770K.
This means that the software was unable to recognize all sixteen threads. We asked Intel about it and they answered that they are working along with DivX developers to provide this support soon.
[nextpage title=”DVDShrink”]
DVDShrink is an old but still very useful program to “shrink” video DVDs that have more than 4.7 GiB of data to fit single-layer DVD media. We used it to compress the DVD of “The Lord of the Rings: The Fellowship of the Ring” DVD to 4.7 GiB. The results below are given in seconds, so the lower the better.
In this test, for our surprise, the other two CPUs included in our review were about 206% faster than the Core i7-5960X. We made some experiments and discovered that, with more than eight processing threads available, DVDShrink uses only one of them. As we mentioned before, DVDShrink is an old program and is not prepared to run on a CPU with a high core count. So, the very bad performance here is due to a bug in this particular software.
[nextpage title=”Media Espresso 6.7″]
Media Espresso is a video conversion program that uses the graphics processing unit of the video engine to speed up the conversion process. We converted a 1 GiB, 1920x1080i, 23,738 kbps, .mov video file to a smaller 320×200, H.264, .MP4 file for viewing on a smartphone. The results below are given in seconds, so the lower the better.
Here the Core i7-5960X was 14% faster than the FX-9590, but 7,8% slower than the Core i7-4770K.
[nextpage title=”Assassin’s Creed IV: Black Flag”]
Released in 2013, Assassin’s Creed IV Black Flag uses the AnvilNext DirectX 11 engine. We adjusted the resolution at 1280 x 720, setting enviroment quality to “very high”, disabled anti-aliasing, and disabling “vsync,” then played the first mission of the game, measuring three times the number of frames per second using FRAPS. The results below are expressed in frames per second and are an arithmetic average of the three results collected.
On Assassin’s Creed IV Black Flag, all the three CPUs provided the same performance level (altough the given result are averages, and maximum and minimum framerates were not the same).
[nextpage title=”Battlefield 4″]
Battlefield 4 is the latest installment in the Battlefield franchise, released in 2013. It is based on the Frostbite 3 engine, which is DirectX 11. In order to measure performance using this game, we walked our way through the first mission, measuring the number of frames per second three times using FRAPS. We ran this game at 1280 x 720, setting overall image quality at “high.”
The results below are expressed in frames per second and are an arithmetic average of the three results collected.
On Battlefield 4, the Core i7-5960X performed the same way of the Core i7-4770K, and was 11% faster than the FX-9590.
[nextpage title=”Call of Duty: Advanced Warfare”]
Call of Duty: Advanced Warfare is the most recent game in the Call of Duty franchise, and it was released in november 2014. It uses a new In-house engine. In order to measure performance using this game, we walked our way through the first mission, measuring the number of frames per second three times using FRAPS. We ran this game at 1280 x 720, overall image quality at “high.”
The results below are expressed in frames per second and are an arithmetic average of the three results collected.
On Call of Duty: Advanced Warfare, the Core i7-5960X reached the same performance level of the Core i7-4770K, and was 39% faster than the FX-9590.
[nextpage title=”F1 2013″]
F1 2013 is a car racing game released in 2013, based on the Formula One championship and using the EGO 3.0 engine.
We tested the performance in this game using the built-in performance test, using “medium” quality settings and 1280 x 720 screen resolution.
The results below are expressed in frames per second.
On F1 2013, the Core i7-5960X performed just like the Core i7-4770K, and was 11% faster than the FX-9590.
[nextpage title=”Far Cry 4″]
Far Cry 4 is based on the Dunia 2 engine, which is DirectX 11. In order to measure performance using this game, we played the same mission three times, measuring the number of frames per second using FRAPS. We ran this game at 1280 x 720 resolution, and overall image quality at “high.”
The results below are expressed in frames per second and are an arithmetic average of the three results collected.
On Far Cry 4, the Core i7-5960X performed 6% faster than the FX-9590, but 12% slower than the Core i7-4770K.
[nextpage title=”Metro Last Light”]
Metro Last Light is a first-person shooter horror game that uses the 4A Engine, launched in 2013.
In order to measure performance using this game, we played the introduction of the game, measuring the number of frames per second three times using FRAPS. We ran this game at 1280 x 720, setting overall image quality at “high”, with SSAA off.
The results below are expressed in frames per second and are an arithmetic average of the three results collected.
On Metro Last Light, the Core i7-5960X achieved the same performance level as the Core i7-4770K, and was 12% faster than the FX-9590.
[nextpage title=”Overclocking”]
The Core i7-5960X has an unlocked clock multiplier, which means you can overclock it by simply changing the core multiplier.
We were only able to tun it stable at 3.7 GHz, mantaining the cache memory clock at 3.0 GHz. At 3.8 GHz, we began to experience random crashes.
However, keep in mind that we did not change voltage or more advanced options, so with a little more patience, you will be able to reach higher overclocking levels.
[nextpage title=”Conclusions”]
We can say the Core i7-5960X is the “most high-end” desktop CPU available today based on its price and specifications. After all, it has sixteen processing threads (eight real and eight virtual, through the Hyper-Threading technology), which is a new record for desktop processors, and DDR4 memory access with quad-channel technology.
However, to answer the question whether or not it is the faster processor in the market today, you must specify what kind of tasks you will run.
If you need a faster CPU to run applications that can actually use sixteen cores, probably the Core i7-5960X is the faster CPU available, as the result with Cinebench R15 suggests.
However, if you run a task that does not use all the available cores, the Core i7-5960X may be slower than the Core i7-4770K, which uses a similar architecture but has a higher clock rate. This was very clear on several of the programs we used.
We also must keep in mind the cost: the Core i7-5960X costs about USD 1,050, three times the price of the Core i7-4770K (and its sucessor, the Core i7-4790K) and more than four times the cost of the FX-9590 from AMD.
Therefore, we can only recommend the Core i7-5960X for users who work professionaly with movie editing, 3D rendering, and similar tasks, where the addition of more processing threads directly increases performance, meaning less waiting time, and more work that can be done.
For any other kind of user (including hardcore gamers), it is a better deal to buy a less expensive CPU and save the money to buy a better video card or additional video cards. You will bring home more performance for less money.
Leave a Reply