Logo   Banner   TopRight
TopUnder
Transparent
AMD FX-8350 Vishera 8-Core CPU Review
Transparent
Date: Oct 23, 2012
Section:Processors
Author: Marco Chiappetta
Transparent
Introduction and Specifications

Even before AMD officially released its Bulldozer-based FX-Series of desktop processors last year, the company was already talking about the follow-on microarchitecture codenamed “Piledriver”. In fact, in the conclusion of our launch article featuring the AMD FX-8150, we posted an AMD-provided slide that showed Piledriver was already on-deck and that it would offer IPC and power improvements over existing architectures, which would result in roughly a 10% to 15% uplift in performance.

We have already shown you what Piledriver could do in mainstream APUs in our coverage of the desktop AMD A10 and A8-Series of products here and the mobile A10-4600M here, but today AMD is finally refreshing its higher-end desktop CPU line-up, which hasn’t seen a new product launch for just over a year. The updated AMD FX-Series of desktop processors featuring the Piledriver microarchitecture was codenamed “Vishera” and we’ve had the flagship variant, the new FX-8350, in the lab for a couple of weeks now. Like other FX-Series processors, the FX-8350 is fully unlocked for easier overclocking, but it’s also been updated with some new features and capabilities courtesy of Piledriver, and AMD has been able to crank the clock speeds way up. In fact, the FX-8350 is the first desktop CPU to feature a base clock of a 4GHz, which can officially Turbo up to 4.2GHz.

We’ve got some quick features and specifications of the new FX-8350 below, along with some related information. Then we’ll move on to the nitty gritty and see just what AMD’s latest flagship desktop processor can do...

AMD FX-Series Processor For 2012
Specifications & Features
CPU Description
Tech/Package 32nm / AM3+
TDP Configs 95W, 125W
Processor Core “Piledriver” (up to 8 Cores), 8MB L2 cache, 256-bit FPUs, 8MB L3 cache
Memory DDR3, 1333-1866, 1.5V
Graphics Core N/A – Pair with AMD Radeon HD 7000 Series Graphics
Performance Management -Core Level: CC6 Power State
-Module Level: L2 Cache power gating via CC6
990FX + SB950
Tech/Package 65nm / FC BGA
TDP Configs 10W for typical configurations
ALink III x4 Gen2
SATA 6 Ports, 6Gbps
USB 12 USB 2.0 Ports, 2 USB 1.1 Internal Ports
PCIe 2.0 Lanes 2x16 or 4x8 + 8x1 + 4x1 + (2x1 on SB)
RAID 0, 1, 5, 10
Software Drivers: Windows 7, Windows 8, Linux


A Side View of the AMD FX-8350 Processor

Before we dive in talk about the new AMD FX-Series, we should call out a few past HotHardware articles which are pertinent to today’s product launch. Although the FX-8350 we’re going to be showing you here today is a new processor, it leverages technologies we’ve previously covered.

The new FX-Series processors, at least for now, will be paired to existing chipsets and motherboards, so we won’t be talking about a totally new platform, like we had to with Trinity and the recently released A-Series APUs. AMD’s latest FX-Series chips use socket AM3+ and are designed for use with previously released AMD 9-Series chipsets, like the high-end 990FX. If you’re unfamiliar with first-gen FX-Series based on the Bulldozer microarchitecture and want to brush up on the features of the 990FX chipset and the newer Piledriver microarchitecture, the list of articles above is a good place to start.

Transparent
Vishera / Piledriver Microarchitecture

Like the recently released A-Series Trinity-based APUs that came before it, the new Vishera-based FX-Series processors leverage AMD's Piledriver microarchotecture. As we've noted before, AMD has either tweaked or revamped many of the primary functional blocks with the Piledriver microarchitecture versus the original Bulldozer which debuted last year.


The AMD Vishera Die

Although much has been changed, Piledriver is essentially an optimization of AMD's Bulldozer core. It shares the same high-level architecture as Bulldozer, but with a number of major enhancements. The same shared fetch, decode, floating point and L2 cache resources per pair of integer units is present in Vishera, however, AMD has improved their branch prediction and L2 efficiency and improved hardware prefetch as well. Piledriver cores also have a larger L1 TLB or Translation Look-aside Buffer.


AMD Made a Number of Enhancements to the Piledriver Microarchitecture

All told, AMD is claiming a combined performance increase of ~14% on the desktop versus their Bulldozer architecture. However, factor in higher Turbo Core 3.0 speed boosts and AMD is claiming larger aggregate performance gains.

The new AMD FX-Series processors is manufactured using Global Foundries' 32nm process node and consists of roughly 1.2B transistors. Although 8, 6, and 4-core variants will be available, all of the initial chips are built around the same die, which is approximately 315mm2. If you're keeping track, that makes the chip virtually identical in size to the previous-gen, Bulldozer-based FX-series.


AMD FX-Series Overview

With Vishera, AMD's Turbo Core technology offers more aggressive clock gating and overclocking. Specifically, the FX-8350 that we'll be showing you here can scale up to 4.2GHz, in single-threaded applications, but has a base clock of 4GHz with dynamic scaling as needed in single or multithreaded workloads.

The degree by which each new member of the FX series is able to Turbo varies from model to model. The 200MHz boost available on the flagship FX-8350 is actually the smallest of the initial line-up--there are other models which run at lower base clocks, but boost upwards of 600MHz.
 

Transparent
Vital Signs and Overclocking

The new AMD FX-8350 uses the same packaging and socket as previous-generation FX-series processors, based on the Bulldozer microarchitecture. As such, save for some markings atop the chip’s integrated heat-spreader, the new AMD FX-8350 looks just like its predecessor.

 
AMD FX-8350 8-Core CPU: Top and Bottom

As we’ve already mentioned though, the main differences between the new FX-series and the previous gen, are the new processor’s updated Piledriver microarchitecture, which offers a number of performance and efficiency enhancements over Bulldozer, and the new processor’s more advanced 32nm manufacturing process.

 
AMD FX-8350 CPU-Z Processor and Cache Details

Here is an up-close-and-personal look at the FX-8350 as reported by the latest version of CPU-Z. As you can see, the FX-8350 uses AMD’s existing AM3+ socket infrastructure. If you remember, socket AM3+ was an update to the older socket AM3, and although pin-compatible, AM3+ adds support for CPU voltage loadline, increased ILDT current for higher frequency HyperTransport links, increased DRAM current, and two memory channels with official support for speeds of up to DDR3-1866. All of these features are exploited on the AMD FX-8350.

As you can see in the images above, the AMD FX-8350 has 128K of L1 data cache (16K per core), 256K of L2 instruction cache (64K shared across each dual-core module), 1024K of L2 cache per core, and a total of 8MB od shared L3 cache. The FX-8350 uses the same 200MHz base clock as previous generation FX-series processors as well; it’s peak 4.2GHz Turbo frequency is derived by using a 21x multiplier and the default 200MHz base clock. Note that the chip runs at about 1.4v when hitting max Turbo. Cool ‘n Quiet is also present on the latest FX, however. When idling, the chip clocks down to 1.4GHz at only .9v.

Overclocking the AMD FX-8350
Boosting Beyond Max Turbo

 
We were present when the original FX was overclocked to over 8GHz, achieving a Guinness Book World Record in the process. Since its predecessor was such an adept overclocked, we were eager to see what the new FX-8350 could do. To that end, we also spent some time overclocking the new AMD FX-8350 processor, using a standard air cooler and the UEFI overclocking options available on the Asus CrossHair V Formula. For these overclocking tests, we bumped the CPU voltage up to 1.4125v, disabled Turbo, and increased the CPU's multiplier until our test system was no longer stable.


AMD FX-8350 Overclocked to 4.7GHz

In the end, we were able to take the FX-8350 up to a stable 4.7GHz. Unfortunately, due to time constraints and an incompatibility with AMD OverDrive and our test-bed’s motherboard, we don’t have accurate temperature data to share at this point. But considering how easy it was to take our CPU to 4.7GHz, we suspect that higher clocks will easily be possible with more exotic cooling and more aggressive voltage tweaking.

As an interesting aside, overclocking the FX-8350 to 4.7GHz resulted in a Cinebench R11.5 score of 7.94, and increase of 1.01 or 14.5% over the stock score of 6.93, and enough of a boost to overtake the Core i7-3770K.
 

Transparent
Test Setup and SiSoft SANDRA

Test System Configuration Notes: When configuring our test systems for this article, we first entered their respective system BIOSes and set each board to its "Optimized" or "High performance Defaults". We then saved the settings, re-entered the BIOS and set the memory speed to each platform's highest, officially supported frequency--1866MHz in the case of AMD's newest FX series processors.  The solid state drives in the test systems were then formatted, and Windows 7 Ultimate x64 was installed. When the Windows installation was complete, we fully updated the OS, and installed the drivers necessary for our components. Auto-Updating and Windows Defender were then disabled and we installed all of our benchmarking software, performed a disk clean-up, cleared any prefetch and temp data, and ran the tests.

HotHardware's Test Systems
Intel and AMD - Head To Head

System 1:
Intel Core i7-3770K
(3.5GHz - Quad-Core)
Intel Core i5-3470
(3.3GHz - Quad-Core)

MSI Z77A-GD65
(Z77 Express Chipset)

2x4GB G.SKILL DDR3-1866
(@ 1600MHz)

GeForce GTX 280
On-Board Ethernet
On-board Audio

OCZ Vertex 3 MaxIOPS

Windows 7 x64
System 2:
Cntel Core i7-3690X
(3.33GHz - Hex-Core)
Intel Core i7-3820
(3.6GHz - Quad-Core)

Asus P9X79 Deluxe
(X79 Express Chipset)

4x4GB G.SKILL DDR3-1866
(@ 1600MHz)

GeForce GTX 280
On-Board Ethernet
On-board Audio

OCZ Vertex 3 MaxIOPS

Windows 7 x64
System 3:
Intel Core i7-990X
(3.43GHz Hex-Core)

Gigabyte EX58-UD4
(X58 Express Chipset)

3x4GB G.SKILL DDR3-1866
(@ 1333MHz)

GeForce GTX 280
On-Board Ethernet
On-board Audio

OCZ Vertex 3 MaxIOPS

Windows 7 x64
System 4:
AMD FX 8350
(4GHz Eight-Core)
AMD FX 8150
(3.6GHz Eight-Core)

Asus CrossHair V Formula
(AMD 990FX Chipset)

2x4GB G.SKILL DDR3-1866
(@ 1866MHz)

GeForce GTX 280
On-Board Ethernet
On-board Audio

OCZ Vertex 3 MaxIOPS

Windows 7 x64

Preliminary Testing with SiSoft SANDRA 2012
Synthetic Benchmarks

We began our testing with SiSoftware's SANDRA 2012, the System ANalyzer, Diagnostic and Reporting Assistant. We ran four of the built-in subsystem tests that partially comprise the SANDRA 2012 suite with Intel's new Core i7 "Ivy Bridge" processor (CPU Arithmetic, Multimedia, Memory Bandwidth, and Cache and Memory). All of the scores reported below were taken with the processor running at its default clock speed of 4GHz (4.2GHz Turbo) with 8GB of DDR3-1866 RAM running in dual-channel mode on the Asus CrossHair V Formula motherboard.


Processor Arithmetic
 AMD FX-8350

 

Processor Multimedia
 AMD FX-8350

Memory Bandwidth
 AMD FX-8350

Cache and Memory
 AMD FX-8350

SiSoft SANDRA didn't reveal any surprises with the AMD FX-8350. As expected, the FX-8350 proved to be faster than the previous-gen FX-8150 across the board, but it couldn't keep pace with the higher-end Intel Core i5 or Core i7 processors. According to SANDRA, the FX-8350 / 990FX combo put up just shy of 19GB/s in the memory bandwidth benchmark and the Cache and Memory test shows relatively high latency until the data set hits about the 512K mark, where the FX-8350 then becomes more competitive with other platforms.

 

Transparent
Futuremark PCMark 7

Futuremark's PCMark 7 is the latest version of the PCMark whole-system benchmarking suite. It has updated application performance measurements targeted for a Windows 7 environment and uses newer metrics to gauge relative performance. Below is what Futuremark says is incorporated into the base PCMark suite and the Entertainment, Creativity, and Productivity suites--the four modules we have benchmark scores for you here.
Futuremark PCMark 7
General Application and Multimedia Performance

The PCMark test is a collection of workloads that measure system performance during typical desktop usage. This is the most important test since it returns the official PCMark score for the system
Storage
  • Windows Defender
  • Importing pictures
  • Gaming

Video Playback and transcoding
Graphics

  • DirectX 9

Image manipulation
Web browsing and decrypting

The Entertainment test is a collection of workloads that measure system performance in entertainment scenarios using mostly application workloads. Individual tests include recording, viewing, streaming and transcoding TV shows and movies, importing, organizing and browsing new music and several gaming related workloads. If the target system is not capable of running DirectX 10 workloads then those tests are skipped. At the end of the benchmark run the system is given an Entertainment test score.

The Creativity test contains a collection of workloads to measure the system performance in typical creativity scenarios. Individual tests include viewing, editing, transcoding and storing photos and videos. At the end of the benchmark run the system is given a Creativity test score.

The Productivity test is a collection of workloads that measure system performance in typical productivity scenarios. Individual workloads include loading web pages and using home office applications. At the end of the benchmark run the system is given a Productivity test score.

The new AMD FX-8350 offered clearly better performance than AMD's previous flagship desktop CPU, the FX-8150, in PCMark 7. Unfortunately for AMD, even the Core i5-3470 has no trouble outpacing the FX-8350 here, in every one of the PCMark 7 tests and the overall score.

Transparent
LAME MT and SunSpider

In our custom LAME MT MP3 encoding test, we convert a large WAV file to the MP3 format, which is a popular scenario that many end users work with on a day-to-day basis to provide portability and storage of their digital audio content. LAME is an open-source MP3 audio encoder that is used widely in a multitude of third party applications.

LAME MT
Audio Encoding

In this test, we created our own 223MB WAV file (a hallucinogenically-induced Grateful Dead jam) and converted it to the MP3 format using the multi-thread capable LAME MT application, in both single and multi-thread modes. Processing times are recorded below, listed in seconds. Shorter times equate to better performance.

 

The new AMD FX-8350 performs much better than its predecessor in our custom LAME MT benchmark, but it gets crushed by every one of the Intel processors we tested.

SunSpider JavaScript Benchmark
JavsScript Performance Testing

Next up, we have some numbers from the SunSpoder JavaScript benchmark. According to the SunSpider website:

This benchmark tests the core JavaScript language only, not the DOM or other browser APIs. It is designed to compare different versions of the same browser, and different browsers to each other. Unlike many widely available JavaScript benchmarks, this test is:

Real World - This test mostly avoids microbenchmarks, and tries to focus on the kinds of actual problems developers solve with JavaScript today, and the problems they may want to tackle in the future as the language gets faster. This includes tests to generate a tagcloud from JSON input, a 3D raytracer, cryptography tests, code decompression, and many more examples. There are a few microbenchmarkish things, but they mostly represent real performance problems that developers have encountered.

Balanced - This test is balanced between different areas of the language and different types of code. It's not all math, all string processing, or all timing simple loops. In addition to having tests in many categories, the individual tests were balanced to take similar amounts of time on currently shipping versions of popular browsers.

Statistically Sound - One of the challenges of benchmarking is knowing how much noise you have in your measurements. This benchmark runs each test multiple times and determines an error range (technically, a 95% confidence interval). In addition, in comparison mode it tells you if you have enough data to determine if the difference is statistically significant.

All of the systems were tested using the latest version of Internet Explorer 9, with default browser settings, on a clean install of Windows 7 Ultimate x64.

The SunSpider benchmark results essentially mirror those of the LAME MT benchmark above. The new FX-8350 offers much improved performance over the previous flagship FX-8150, but the 8350's performance still lags behind anything from Intel here.

Transparent
Cinebench R11.5 and POV-Ray

Cinebench R11.5 is a 3D rendering performance test based on Cinema 4D from Maxon. Cinema 4D is a 3D rendering and animation suite used by animation houses and producers like Sony Animation and many others. It's very demanding of processor resources and is an excellent gauge of pure computational throughput.
Cinebench R11.5
3D Rendering

This is a multi-threaded, multi-processor aware benchmark that renders a photorealistic 3D scene (from the viral "No Keyframes" animation by AixSponza). This scene makes use of various algorithms to stress all available processor cores. The rate at which each test system was able to render the entire scene is represented in the graph below.

 

The AMD FX-8350 offers improved performance over the previous-gen FX-8150 in both single and multi-threaded workloads according to Cinebench R11.5. The FX-8350 is also able to overtake the Core i5-3470 in this test and nip at the Core i7's heels.

POV-Ray Performance
Ray Tracing

POV-Ray, or the Persistence of Vision Ray-Tracer, is an open source tool for creating realistically lit 3D graphics artwork. We tested with POV-Ray's standard 'one-CPU' and 'all-CPU' benchmarking tools on all of our test machines, and recorded the scores reported for each. Results are measured in pixels-per-second throughput; higher scores equate to better performance.

POV-Ray proved to be somewhat of a strong point for the new FX-8350. In this benchmark, the FX-8350 once again offers improved single and multi-threaded performance over the FX-8150 and is also able to overtake all of the Intel quad-core processors. Only the 6-core, and much more expensive, Core i7-3960X was faster.
 

Transparent
Low-Res Gaming: Crysis and ETQW

For our next set of tests, we moved on to some in-game benchmarking with Crysis (DirectX) and Enemy Territory: Quake Wars (OpenGL). When testing processors with Crysis or ET:QW, we drop the resolution to 1024x768, and reduce all of the in-game graphical options to their minimum values to isolate CPU and memory performance as much as possible. However, the in-game effects, which control the level of detail for the games' physics engines and particle systems, are left at their maximum values, since these actually place some load on the CPU rather than GPU.

Low-Resolution Gaming: Crysis and ET: Quake Wars
Taking the GPU out of the Equation

The performance trend in our low-res gaming tests is a repeat of most of the tests from the previous pages. In both Crysis and ET: Quake Wars, the new AMD FX-8350 offers significantly better performance than the previous-generation FX-8150 due to its architectural and frequency advantages. However, the FX-8350 still trails even the "slowest" of the Intel processors by a large margin in these game tests.
 

Transparent
Total System Power Consumption

Throughout all of our benchmarking and testing, we also monitored how much power our test systems consumed using a power meter. Our goal was to give you all an idea as to how much power each configuration used while idling and while under a heavy workload. Please keep in mind that we were testing total system power consumption at the outlet here, not just the power being drawn by the processors alone.

Total System Power Consumption
Tested at the Outlet

The new Piledriver-based AMD FX-8350 is clearly more efficient than the previous-generation FX-8150. Not only did the FX-8350 offer clearly better performance than its predecessor, but it did so with lower idle and peak power consumption. While it's an improvement over the previous-gen FX though, the new 8350 still uses much more power than the higher performing, Ivy Bridge-based Core i7-3770K. Intel's current processors are simply in another league in terms of power efficiency due to their architectural and manufacturing advantages.

We also monitored the FX-8350's power consumption while we had the chip overclocked. As you can see, boosting the chip up to 4.7GHz (with a model +.03v bump in voltage) resulted in a huge increase in power consumption, to the tune of almost 100 watts. Although there is some headroom left in the chips, AMD is obviously pushing the envelope with the FX-8350's default boost clock of 4.2GHz.  A less than 10% increase in frequency and a tiny bump in voltage resulted in a 41% increase in power consumption.

Transparent
Performance Summary and Conclusion

Performance Summary: The FX-8350 is the most powerful desktop processor released by AMD to date. The improvements afforded by its updated Piledriver microarchitecture, in addition to significantly increased frequencies, have resulted in a processor that clearly outperforms its Bulldozer-based predecessors across the board. And it does so while consuming less power, which hints at its improved efficiency as well. Unfortunately for AMD, while they have eaten into Intel’s lead in the desktop CPU space somewhat, they have not changed the performance landscape all that much. Intel’s higher-end Core i5 and Core i7 processors, are still the highest performing and most efficient desktop processors available.

The new AMD FX-Series of desktop processor is still a clear step forward for AMD. The FX-8350 we tested offered better performance and improved power efficiency than its predecessor in every test we ran. Forward progress is something that AMD needed and that is what they got with the new FX-Series of desktop processors, even if the amount of forward progress isn’t enough to significantly shift the balance of power in the desktop CPU space. We would have liked to have seen a new chipset accompany this CPU release that would bring things like native USB 3.0 and PCI Express 3.0 support to AMD’s high-end desktop platform, but that’s not in the cards just yet. The former can be added via third-party controllers and the latter offers minimal benefit to end-users at this time.  Regardless, it still would have been nice to have feature parity between Intel and AMD in that respect. Compatibility with current platforms does make it easy for existing AMD owners to upgrade their processors, however, which is something we always like to see.

The initial Piledriver-based FX-Series product line-up is going to consist of four processors, ranging from the high-end FX-8350 at $195 to the more mainstream FX-4300 at $122. The processors will differ in their core counts, Northbridge frequencies, base and Turbo CPU clocks, cache compliments and TDP. Please note, however, that all of these initial FX-Series processors will feature the same die; quad and hex-cores listed here will feature the same 8-core die with a core module or two disabled. At the FX-8350’s expected price point, it goes head-to-head with Intel processors like the Core i5-3470 ($199) and Core i5-3570 ($214). Unfortunately, the performance comparisons between those chips don’t favor AMD. The FX-8350 outpaced the i5-3470 in a couple of tests, but overall, we’d still consider the Core i5-3470 the “faster” CPU. The Core i5 is more power friendly too. What the AMD FX-8350 does offer over mainstream Core i5 chips though is an easy upgrade path for existing AMD owners and more flexibility for overclocking, due to its unlocked multipliers.

Ultimately, the new FX-Series of desktop processors from AMD is an evolutionary step for the company that advances their products forward. If you’re an AMD fan looking for a new CPU, the FX-8350 is the fastest processor the company has released to date. It is not, however, a threat to Intel’s high end CPU line-up.

  • Good Performance
  • Decent Overclocker
  • Improved Efficiency
  • Easy Upgrade For Existing AMD Owners

 

  • Dated Platform
  • Intel CPUs Still Clearly Faster

 



Content Property of HotHardware.com