[nextpage title=”Introduction”]
Today, AMD is releasing its second-generation of FX CPUs, dubbed “Vishera.” The FX-8350 is currently the highest-end model available, running at 4 GHz and costing USD 195. Let’s see how it fares against its main competitor, the Core i5-3470 (3.2 GHz, USD 200). We also included in our comparison the AMD CPU that used to be the highest-end model, the FX-8150 (3.6 GHz, USD 190 currently, USD 280 at its release), and what was then its main competitor, the Core i5-2500K (3.3 GHz, USD 220). For curiosity sake, we also added to the comparison the Phenom II X6 1100T (3.3 GHz, USD 190 when it was sold), which used to be AMD’s fastest CPU before the release of the FX family.
The new FX-8350 uses the socket AM3+, and is compatible with any socket AM3+ motherboard, although the motherboard may need a BIOS upgrade to correctly recognize the new CPU.
Figure 1: The AMD FX-8350 processor
The second-generation FX processors are based on the new “Piledriver” architecture, which is basically the “Bulldozer” architecture used by the previous generation with a few tweaks.
In the tables below, we compare the main features of the CPUs included in our review. All CPUs support the SSE4 instruction set (both SSE4.1 and SSE4.2) and the AVX instruction set, except for the Phenom II X6 1100T, which doesn’t support these instruction sets. On the Phenom II X6 1100T, the HyperTransport bus (which is used to connect the CPU to the chipset) works at 2 GHz (8 MB/s); the memory controller works at 2 GHz, while on the FX CPUs the HyperTransport bus works at 2.6 GHz (10.4 GB/s) and the memory controller works at 2.2 GHz. All CPUs support a “turbo clock” technology, which increases the CPU internal clock as needed. The Core i5 processors included in our comparison do not support the Hyper-Threading technology.
It is important to understand that with the FX CPUs from AMD, each pair of CPU cores share the same front-end engine (i.e., the fetch unit, the L1 instruction cache, and the instruction decoders). So, each pair of CPU cores is part of the same “module,” and the FX-8350 and the FX-8150 are comprised of four of these modules. With the other CPUs included in our review, every core is a complete CPU, each with its own front-end engine.
CPU | Cores | HT | Internal Clock | Turbo Clock | Max Turbo | Core | Technology | TDP | Socket | Price |
FX-8350 | 8 | No | 4.0 GHz | 4.2 GHz | 4.2 GHz | Vishera | 32 nm | 125 W | AM3+ | USD 195 |
FX-8150 | 8 | No | 3.6 GHz | 3.9 GHz | 4.2 GHz | Zambezi | 32 nm | 125 W | AM3+ | USD 190 |
Phenom II X6 1100T | 6 | No | 3.3 GHz | 3.7 GHz | 3.7 GHz | Thuban | 45 nm | 125 W | AM3 | NA |
Core i5-2500K | 4 | No | 3.3 GHz | 3.7 GHz | 3.7 GHz | Sandy Bridge | 32 nm | 95 W | LGA1155 | USD 220 |
Core i5-3470 | 4 | No | 3.2 GHz | 3.6 GHz | 3.6 GHz | Ivy Bridge | 22 nm | 77 W | LGA1155 | USD 200 |
Prices were researched at Newegg.com on the day we published this review, except for the FX-8350, which is the price provided by AMD. The Phenom II X6 1100T is no longer sold. TDP stands for Thermal Design Power and states the maximum amount of heat the CPU can dissipate. The CPU cooler must be capable of dissipating at least this amount of heat.
Below you can see the memory configuration for each CPU.
CPU | L1 Cache (I + D) | L2 Cache | L3 Cache | Memory Support | Memory Channels |
FX-8350 | 64 kB per module + 16 kB per core | 2 MB x 4 | 8 MB total | Up to DDR3-1866 | Two |
FX-8150 | 64 kB per module + 16 kB per core | 2 MB x 4 | 8 MB total | Up to DDR3-1866 | Two |
Phenom II X6 1100T | 64 kB + 64 kB per core | 512 kB per core | 6 MB total | Up to DDR3-1333 | Two |
Core i5-2500K | 32 kB + 32 kB per core | 256 kB per core | 6 MB total | Up to DDR3-1333 | Two |
Core i5-3470 | 32 kB + 32 kB per core | 256 kB per core | 6 MB total | Up to DDR3-1600 | 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.
Hardware Configuration
- Motherboard (Socket AM3+): ASUS Crosshair V Formula (1605 BIOS)
- Motherboard (Socket LGA1155): Gigabyte Z77X-UP5TH (F11 BIOS)
- CPU Cooler: Intel/AMD stock
- Memory: 4 GB DDR3-1333, two G.Skill F3-10666CL7T memory modules
- Hard Disk Drive: Western Digital Black Caviar 1 TB (WD1001FALS, SATA-300, 7,200 rpm, 32 MB buffer)
- Video Card: EVGA GeForce GTX 660 SuperClocked
- Video Monitor: AOC e3343Fwk
- Power Supply: Antec TruePower New 750 W
Operating System Configuration
- Windows 7 Ultimate 64-bit
- NTFS
- Video resolution: 1920 x 1080 60 Hz
Driver Versions
- NVIDIA video driver version: 306.97
- Intel Inf chipset driver version: 9.3.0.1021
- AMD chipset driver version: 3.0.825.0
Software Used
- 3DMark 11 Professional 1.0.3
- Adobe After Effects CS4
- Adobe Photoshop CS5 Extended + Retouch Artist Speed Test 1.0
- Battlefield 3
- Cinebench 11.529
- Far Cry 2 – Patch 1.03
- iTunes 10.7
- Media Espresso 6.5
- PCMark 7 1.0.4
- Starcraft II: Wings of Liberty – Patch 1.5
- VirtualDub 1.9.5 + MPEG-2 Plugin 3.1 + DivX 6.9.2
- WinZip 17
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 7″]
PCMark 7 performs a series of tests and gives scores in the following categories: An overall score called PCMark; a Productivity score, which is the system performance when using applications such as web browsing and home office applications; a Creativity score, which is the system performance when viewing, editing, converting, and storing photos and videos; an Entertainment score, which is the system performance when recording, viewing, streaming, and converting TV shows and movies, importing, organizing, and browsing music, and gaming; and a Computation score, which indicates the processing performance of the system. Let’s analyze the results.
The Core i5-3470 achieved an overall score 28% higher than the FX-8350, which achieved the same performance level as the FX-8150. The Core i5-2500K achieved a score 13% higher than the FX-8350.
On the productivity benchmark, the Core i5-3470 achieved a score similar to the Core i5-2500K, which was 16% higher than the FX-8350. The FX-8350 achieved the same score as the FX-8150.
On the creativity benchmark, the Core i5-3470 achieved a score 59% higher than the FX-8350. The Core i5-2500K achieved a score 12% higher than the FX-8350, which achieved the same performance level as the FX-8150.
On the entertainment benchmark, all CPUs achieved the same performance level. This was expected, as this benchmark takes mostly into consideration the performance of the video card.
On the computation benchmark, the Core i5-3470 achieved a performance level 198% higher than the FX-8350. The performance level of the Core i5-2500K was 30% higher than the FX-8350. Here the FX-8350 achieved a score 10% higher than the FX-8150.
[nextpage title=”VirtualDub”]
With VirtualDub, we converted a full-length DVD movie to DivX format and saw how long it took for this conversion to be completed. 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 quite remarkable.
The results below are given in seconds, so the lower the better.
On DivX encoding, the Core i5-3470 was 29% faster than the FX-8350. The new CPU
from AMD was 8% faster than the FX-8150 and 10% faster than the Core i5-2500K.
[nextpage title=”Photoshop CS5″]
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 CS5, we used a script called “Retouch Artist Speed Test,” which applies a series of filters to a sample image and gives us the time that Photoshop took to run all the filters. The results below are given in seconds, so the lower the number the better.
On Photoshop CS5, the Core i5-3470 was 30% faster than the FX-8350, while the Core i5-2500K was 7% faster than the new CPU from AMD. The FX-8350 was 8% faster than the FX-8150 on this program.
[nextpage title=”After Effects CS4″]
After Effects is a very well-known program for video post-production, to add animation and visual effects in videos. To evaluate the performance of each CPU running this program, we ran a workload consisting of a number of 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 times each CPU took to finish the whole batch, given in seconds, so the lower the number the better.
On After Effects CS4, the Core i5-3470 was once again the fastest CPU, with a 16% advantage over the FX-8350. The Core i5-2500K was 9% faster than the FX-8350, which was 10% faster than the FX-8150.
[nextpage title=”Media Espresso 6.5″]
Media Espresso is a video conversion program that uses the graphics processing unit of the video card to speed up the conversion process. It is also capable of using Intel’s QuickSync technology available in the CPUs from this company. We converted a 449 MB, 1920x1080i, 18,884 kbps, MPG2 video file to a smaller 640×360, H.264, .MP4 file for viewing on a portable device such as an iPhone or iPod Touch.
Here the Core i5-3470 was again the fastest CPU, with an 11% advantage over the FX-8350, which achieved the same performance level as the Core i5-2500K. The new FX-8350 was 5% faster than the FX-8150.
[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 i5-3470 was 27% faster than the FX-8350. The Core i5-2500K was 8% faster than the FX-8350, which achieved the same performance level as the FX-8150.
[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.
Here the Core i5-3470 and the FX-8350 achieved the same performance level. The new CPU from AMD was 9% faster than the FX-8150 and 11% faster than the Core i5-2500K.
[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 – to speed up the process. (On CPUs with Hyper-Threading technology, each core is recognized as two cores by the operating system.)
Here the advantage switched to the CPUs from AMD, since they are eight- or six-core parts, while the CPUs from Intel are four-core products. The FX-8350 achieved a score 15% higher than the FX-8150, 22% higher than the Core i5-3470, and 30% higher than the Core i5-2500K.
[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 amount of textures that can be represented on one screen can push most of the top-end graphics cards to their limits. 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 1920×1080, configuring “Texture Quality” and “Graphics Quality” to “Medium.” When benchmarking CPUs, we can’t increase image quality settings; otherwise, we will measure the performance of the video card, not the performance of the CPU. 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 card.
The advantage of the Core i5-3470 over the FX-8350 on StarCraft II was marginal (4%), with the FX-8350, the FX-8150, and the Core i5-2500K achieving the same performance level.
[nextpage title=”Far Cry 2″]
Released in 2008, Far Cry 2 is based on a game engine called Dunia, which is DirectX 10. We used the benchmarking utility that comes with this game at 1920×1080, configuring “Overall Quality” to “High” (which is the minimum value for this setting), disabling anti-aliasing, and running the “Ranch Long” demo three times. When benchmarking CPUs, we need to lower the image quality settings in order to get a very high number of frames per second; othe
rwise, we will be benchmarking the video card. The results below are expressed in frames per second and are an arithmetic average of the three results collected.
All CPUs achieved the same performance level here, except for the Phenom II X6 1100T, which was 3.3% slower than the FX-8350.
[nextpage title=”DiRT3″]
DiRT3 is a DirectX 11 game. We measured performance using this game by running a race and then playing it back using FRAPS. We tested this game at 1920×1080, configuring all image quality settings at their lowest values, disabling anti-aliasing and anisotropic filtering. As already explained, when benchmarking CPUs, we need to lower the image quality settings in order to get a very high number of frames per second; otherwise, we will be benchmarking the video card.
On DiRT3, the Core i5-3470 was 4% faster than the FX-8350, which achieved the same performance level as the Core i5-2500K and the FX-8150, and was 4% faster than the Phenom II X6 1100T.
[nextpage title=”Battlefield 3″]
Battlefield 3 is the latest installment in the Battlefield franchise released in 2011. It is based on the Frostbite 2 engine, which is DirectX 11. In order to measure performance using this game, we walked our way through the first half of the “Operation Swordbreaker” mission, measuring the number of frames per second using FRAPS. We tested this game at 1920×1080, configuring all image quality settings at their lowest values. As mentioned before, when benchmarking CPUs, we need to lower the image quality settings in order to get a very high number of frames per second; otherwise, we will be benchmarking the video card.
On Battlefield 3, all CPUs achieved the same performance level, except for the Phenom II X6 1100T, which was 3.5% slower than the FX-8350.
[nextpage title=”Borderlands 2″]
Borderlands 2 is a very recent game, released in 2012, supporting NVIDIA’s PhysX technology. We used the in-game benchmarking utility at 1920×1080, disabling PhysX, anti-aliasing, and anisotropic filtering, and configuring all image quality settings at their lowest values. As mentioned before, when benchmarking CPUs, we need to lower the image quality settings in order to get a very high number of frames per second; otherwise, we will be benchmarking the video card.
On this game, all CPUs achieved the same performance level, except for the Phenom II X6 1100T, which was 4.1% slower than the FX-8350.
[nextpage title=”3DMark 11 Professional”]
3DMark 11 Professional measures Shader 5.0 (i.e., DirectX 11) performance. We ran this program at 1920×1080 using the “Entry” profile.
This program provides three different scores: graphics, physics and combined.
All CPUs achieved approximately the same graphics score. This was expected, since this score measures the performance of the video card, not that of the CPU.
The physics score measures exclusively the physics performance of the system, a process that is typically done on the CPU. Here the Core i5-3470 achieved a score 42% higher than the FX-8350, and the Core i5-2500K achieved a score 32% higher than the FX-8350. The FX-8350 achieved a score 10% higher than the FX-8150.
The combined score shows a balance between the graphics and the physics performance achieved by each system being tested. Here the Core i5-3470 achieved a score 15% higher than the FX-8350. The Core i5-2500K achieved a score 10% higher than the FX-8350. The FX-8350 achieved a score 4% higher than the FX-8150.
[nextpage title=”Overclocking”]
The FX processors have their clock multiplier unlocked, allowing you to overclock them by changing this parameter. With the FX-8350, which runs internally at 4 GHz multiplying a base clock of 200 MHz by 20, we could increase the clock multiplier to 22 and the base clock to 212 MHz, resulting in an internal clock rate of 4,664 MHz, a 16.6% increase over the CPU default clock rate.
Then we replaced the stock cooler with AMD’s liquid cooling solution, and we were able to increase the CPU multiplier to 22.5 and the base clock to 213 MHz, resulting in an internal clock rate of 4,792 MHz, a 19.8% increase over the CPU default clock rate.
In both cases, we increased the CPU voltage to 1.4000 V (from 1.348 V), the CPU-NB voltage to 1.4000 V (from 1.187 V), and the CPU VDDA voltage to 2.8000 V (from 2.490 V).
With more time and patience, you may be able to achieve even better results than we did.
[nextpage title=”Conclusions”]
The FX-8350 provided a slight performance improvement over the FX-8150. However, we can’t tell if this improvement was caused by the use of the new “Piledriver” architecture, by the use of a higher clock rate, or a combination of both.
While AMD has the lead on the USD 100 price segment, it is way behind Intel on the USD 200 price segment. Also, the company doesn’t have any product to compete against the Core i7 at the high-end segment.
Costing the same and providing up to 30% performance advantage over the FX-8350, the Core i5-3470 is a far better choice. We simply can’t recommend the new FX-8350.
For games, the performance is dictated mostly by the video card, not by the CPU. Therefore, some could claim that for games it doesn’t matter which CPU you choose. However, you will use the computer for other tasks, and the CPU from Intel will be faster for those.
The only scenario where the FX-8350 was faster was on 3D rendering with Cinebench. This could mean that the FX-8350 is a better choice for professionals rendering 3D images. However, we believe this kind of user will prefer to buy a Core i7 processor instead.
Leave a Reply