AMD FX-8350 Piledriver “FX-Vishera” Processor Review
In October 2011, AMD made a return to their FX processor platform with the latest Bulldozer architecture that was meant to bring competitive performance against Intel's Sandy Bridge counterparts. Fans around the globe were anticipating the launch of the first generation of FX processor platform codenamed "Zambezi" that would once again bring AMD back in the game of x86 performance leadership with an architecture that's built up from scratch.
When the FX-Bulldozer processors finally launched, people got excited to learn that the processors boast upto 8 cores, impressive clock speeds and great overclocking potential as demonstrated by AMD a few weeks before launch. All that excitement was soon over when reviews hit the internet and consumers learnt that the Bulldozer architecture offers mediocre performance improvement over the Phenom II processors and fails to show any kind of impressive performance results over Intel's Sandy Bridge processors. The lackluster single-thread performance and higher TDP of 125W made matters worse for AMD, so worse that it cost the CEO of the company his job as revealed by AMD themselves a few months ago.
Never again would AMD make such a mistake, though it did pave the way for AMD in a year over year platform refresh. Earlier that year, a roadmap was released by AMD which showed that the successor to Bulldozer code-named "Piledriver" would arrive in 2012 followed by Steamroller in 2013 and Excavator in 2014, each of which bring 15-20% IPC improvement over the previous generation core. So in 2012, All eyes were on Piledriver, the latest architecture by AMD. AMD revealed through slides that Piledriver would be a refresh to Bulldozer offering improvements were Bulldozer architecture lacked and the performance deficit of 15-20% would be gained by optimizing the new core for their 2012 FX CPU lineup. This meant that the CPU would still be based on the 32nm SOI process form globablfoundaries at a time when Intel would launch their Ivy Bridge platform making use of the 22nm process with even lower power consumption than Sandy Bridge.
It was a difficult time for AMD but the decisions they made in 2012 would bring the company back on its knees with a bold roadmap ahead of them. Let's take a look at what the current generation Piledriver core based FX-Vishera processors have to offer in terms of performance.
AMD Piledriver Architecture - Bulldozer Updated
AMD launched their new Piledriver core architecture based FX-Vishera processors on 23rd October 2013. The first question that would pop up in everyone's mind is that, what is new in Piledriver?
Design and architecture wise, Piledriver is technically the same thing that Bulldozer was, its based on the same 32nm SOI process by Global Foundries featuring a die size of 315mm2, 1.2 Billion transistors and upto 8 cores. Each Piledriver module comes with two x86 Vishera cores with an integrated memory controller and shared 2 MB L2 cache. The new Piledriver architecture also features the latest FMA3 and F16C instruction sets. With these small enhancements and optimizations, AMD aimed to deliver a more efficient Bulldozer design which ends up around 10 to 15% performance improvmenet over the Bulldozer core processors.
AMD FX 8-Core Processors
- The industry’s first and only native 8-core desktop processor for unmatched multitasking and pure core performance with all-new CPU architecture.
- New 32 nanometer die shrink designed to reduce leakage for improved efficiency, increased clock rate headroom and better thermals.
AMD Turbo CORE Technology
- The AMD FX Processors come equipped with AMD Turbo CORE Technology. AMD Turbo CORE Technology is a performance boosting technology that helps increase performance on the applications that need it the most.
New Instruction Capabilities
- Advanced Vector Extensions increase parallelism tailored for scientific and 3D applications that use heavy floating point calculations
- FMA4 and XOP
- Floating Point Vector Multiply -Accumulate improves throughput and performance on many vector functions (integer and floating point)
- Advanced Encryption Standard noticeably increase performance on the latest encryption applications like TrueCrypt and benchmarks like PCMark
AMD Balanced Smart Cache
- Shared L3 cache ( up to 8MB)
- Improved scheduling and pre-fetch capabilities
- 64-ways (16-ways/sub-cache)
- Increased data queue sizes
- Coherency for 8-cores
AMD Wide Floating Point Accelerator
- Shared FP Scheduler
- Dual 128-bit Floating point engines – capable of teaming together for 256-bit AVX instructions or operating separately with each core.
- One 16-bit link at up to 5600MT/s
- Up to 8.0GB/s HyperTransport™ I/O bandwidth; Up to 16GB/s in HyperTransport Generation 3.0 mode
- Up to 37GB/s total delivered processor-to-system bandwidth (HyperTransport bus + memory bus)
Benefit: Quick access times to system I/O for better performance.
Integrated DRAM Controller with AMD Memory Optimizer Technology
- A high-bandwidth, low-latency integrated memory controller
- Supports up to DDR3-1866
- Supports new low voltage memories of 1.35V and 1.2V
- Up to 29.9GB/s memory bandwidth for DDR3
- New Pre-Fetcher improvements
- Direct communications to each core in Dual-Core module (APIC registers in each core)
Benefit: Optimized memory controller to feed more cores
AMD Virtualization (AMD-V) Technology with IOMMU
- Silicon feature-set enhancements designed to improve the performance, reliability, and security of existing and future virtualization environments by allowing virtualized applications with direct and rapid access to their allocated memory.
- IOMMU is an extension to AMD64 architecture to support address translation and access protection on DMA transfers
- Security for User Level application and Virtual Machine guest operating system
- Address translation and access control
- Device isolation
- Device assignment in virtualized systems
- Security & trusted boot support Unified interrupt management
- Security for User Level application and Virtual Machine guest operating system
AMD PowerNow! Technology (Cool’n’Quiet Technology)
- Enhanced power management features which automatically and instantaneously adjusts performance states and features based on processor performance requirements
- C6 power state for cache flush, and voltage down individual core
- CC6 power state allows all cores in C6 to power even lower
- For quieter operation and reduced power requirements
- Separate memory controller power control
- IO-based c-state interface
- Works automatically without the need for drivers or BIOS enablement.
- Power can be switched on or off within a single clock cycle, saving energy with no impact to performance.
AMD Turbo Core and Power Management Technologies
AMD has also focused on power efficiency with the Piledriver architecture by minimizing the silicon area through sharing various functions and components between two cores on a single module. Each module will have its own set of functions cross-shared between two cores hence saving space and power. Piledriver also under goes an extensive flip-flop clock gating throughout design and the circuits are power gated dynamically to ensure less leakage within the new processors.
Various other power-saving features are added to the design through software/firmware which includes C6 and Low power C1E state that dynamically reduces clock speed when the processor is not being used. Application power management and DRAM power management is also include with the new processors.
A new feature introduced with AMD’s FX-Piledriver processors is the addition of Turbo Core technology that allows the processor to dynamically boost up the turbo core speed for faster clock functionality. The Tech is already available on Intel’s processors starting with Nehalem generation of processors. AMD now brings the same yet improved Turbo Core tech to their processors and it allows them to run beyond 4 GHz clock speed.
The Turbo Core technology on AMD’s Piedriver processors has improved dramatically since its first yet un-impressive introduction of Phenom II CPUs. With a more robust approach, AMD has now set various points which are configurable by the processor itself to adjust the Turbo Core limit. At base clock, a certain amount of TDP would be available which can be turned into extra clock speed when needed on all eight cores. Now suppose we have an application that doesn’t utilize all eight cores on the processor but instead makes use of four. The CPU would dynamically disable the four extra cores and configure the remaining four cores with the max available turbo core speed which would result in excess amount of performance improvement under demand.
AMD FX-Vishera CPU Lineup
AMD’s FX-Vishera lineup consists of four FX processors all of which fall in three separate tiers. The high-end FX 8 Core series are labelled as “Black Edition” which feature 8 native CPU cores, enhanced gaming and faster task achieving. Next up are the FX 6 core processors followed by the entry level FX 4 core offerings. All of AMD FX-Processors come with an unlocked design allowing for unprecedented overclocking. The complete lineup is listed below:
|CPU Base||4.0 GHz||3.5 GHz||3.5 GHz||3.8 GHz|
|CPU Turbo Core||4.2 GHz||4.0 GHz||4.1 GHz||4.0 GHz|
|CPU Turbo MAX||4.2 GHz||4.0 GHz||4.1 GHz||4.0 GHz|
|Cores||8 Cores||8 Cores||6 Cores||4 Cores|
|TDP||125 W||125 W||95 W||95W|
|L3 Cache||8 MB L3||8 MB L3||8 MB L3||8 MB L3|
AMD has once again bumped up the clock speeds to absolute heights securing the GHz dominance if not IPC. This means that while the architecture remains almost similar as Bulldozer, the Piledriver based processors deliver far more better efficiency and performance within the same package.
AMD FX-8350 – Pumping Up The Clocks
Today, we will be testing the flagship FX-8350 processor that comes with an Octa-core design and features impressive clock speeds of 4.0 GHz base and 4.2 GHz Turbo. The chip is compatible with the latest AM3+ socket motherboards which will be detailed later on. The CPU comes with 8 MB of L3 cache and has a 125 W TDP. The AMD FX-8350 had a launch price of $195 which offered competetive performance against the Core i5-3570K at much lower price range. For comparison, the Core i5-3570K retailed at around $220 US when FX-8350 launched but it comes with a 22nm design as opposed to the 32nm architecture on the Piledriver processor.
AMD 990FX Chipset and AM3+ Socket
Rather than releasing a new socket and chipset for their FX-Vishera platform, AMD stood loyal with their already released AM3+ platform featuring the 990FX chipset. Both series processors are pin-to-pin compatible with their new socket and support the same set of features as enabled on the FX-Bulldozer processors.
AM3+ boards still come with a southbridge (SB950) and northbridge (990FX). The AMD 990FX chipset is the latest in the stack replacing the 890FX chipset though is alot similar to it. Both chips are directly connected to the AM3+ socket processor that offers a DDR3 Dual channel memory controller supporting memory with speeds of upto 1866 MHz. The chipset adopts the PCI-e 2.0 interface (3.0 isn’t available on 990FX chipset). The SB950 offers up to 14 USB 2.0 ports (USB 3.0 ports are offered through an separate chip), 6 SATA 6 GB/s ports, HD audio, Gigabit LAN and PCI interface.
The AMD 990FX chipset also features SLI multi-gpu functionality on AM3+ motherboards which wasn’t available on 890FX chipset.
|Power Supply:||Xigmatek NRP-MC1002 1000 Watt|
|Hard Disk:||Intel SSD 520 Series 256 GB (OS)Seagate Barracuda 500GB 7200.12|
|Memory:||4 x 4 GB Kingston HyperX 2133 MHz10th Anniversary Edition Memory Kit|
|Case:||Cooler Master HAF 932|
|Video Cards:||ASUS GeForce GTX 680|
|Cooling Solutions:||Phantek PH-TC14PE Triple Fan|
|OS:||Windows 8 Ultimate 64-bit|
Overclocking the FX-8350
Overclocking the FX-8350 was a fairly easy job, we set the bus speed to 218.00 MHz (22x Multiplier) with a voltage adjustment of 1.48V and turbo core disabled across all cores so the CPU reaches 4.8 GHz clock speed. This is a good improvement over the base speed of 4.0 GHz. The FX-8350 can easily reach around 5 GHz overclocked frequencies with better coolers.
We have included the overclock scores in the charts which are provided in the performance section. Before we move into our on benchmarking session, we would like to show you some slides made by AMD themselves which show the FX-8350's competitive performance against the Ivy Bridge i5-3570K processors:
X264 HD Encode Benchmark
This benchmark measures the encoding performance of the processor. It offers a standardized benchmark as the clip as well as the encoder used is uniform.
Cinebench is based on Maxon’s Cinema 4D. It is used to compare graphics as well as processor performance. We are using the CPU performance numbers for our comparison.
PCMark 7 is a complete PC benchmarking solution for Windows 7 and Windows 8. It includes 7 tests combining more than 25 individual workloads covering storage, computation, image and video manipulation, web browsing and gaming. Specifically designed for the full range of PC hardware from netbooks and tablets to notebooks and desktops, PCMark 7 offers complete Windows PC performance testing for home and business use.
3DMark Vantage CPU Performance
3DMark Vantage is a DirectX 10 video card benchmark test for Windows that is designed to measure your PC’s gaming performance. While the overall benchmark is great, the utility also provides a good indication of the CPU performance.
Super PI is used by many overclockers to test the performance and stability of their computers. In the overclocking community, the standard program provides a benchmark for enthusiasts to compare “world record” pi calculation times and demonstrate their overclocking abilities. The program can also be used to test the stability of a certain overclock speed.
FX-8350 – Battlefield 3 Performance
Battlefield 3 is the award winning mutliplayer shooter from EA’s studio DICE. The game features an enhanced Frostbite 2.0 engine that implements DirectX 11 features such as tessellation, dynamic lightning and sub-surface scattering with per particle motion blur. Battlefield 3 relies heavily on graphics performance which is the reason we are looking at minimal differences in performance from each processor in the chart below:
FX-8350 – Elder Scroll Skyrim Performance
Bethesda released their Elder Scrolls Skyrim in 2011 and the hype level shattered the roofs. The game was met with wide success and is built with the latest Creation engine which meant a farewell to the much older game bryo engine from Bethesda. The game has one of the most biggest and well crafted world ever created. The game requires a good CPU for stable performance, most GPUs can handle the game well. It wasn’t until users started pouring in their custom content and mods for the PC version such as high-resolution textures and more which required GPUs with higher VRAM for adequate performance. Skyrim is known as one of the reasons by PC gamers for GPUs with higher VRAM.
FX-8350 – CRYSIS 3 Performance
Crysis franchise is the crown jewel for the PC gamers. Even after going multi-plat, the game is considered to be the most brilliant and gorgeous game ever brilliant, a visual marvel in short. The third and most probably the last title in the Crysis trilogy blends in the most deepest story line of the franchise with graphics that would make any other game cry in shame. Powering the game is the beastly and much user friendly CryEngine 3 that adopts all the features of DirectX 11 API. Simply speaking, the game demands a high performance graphics card than a high performance CPU to play well on higher settings.
The FX-8250 is built on the same 32nm process as its predecessor, the FX-8150. It comes in a 125W TDP package which is higher than the mainstream processors from Intel. While AMD pushed for efficiency in its new Piledriver architecture which did improve the power consumption to some extent, Intel on the other hand toned down to a 22nm process with a 77W TDP package which is far better than AMD's offerings hence AMD did take a hit by not going for a die shrink.
By the end of this review, all i have to say is that AMD now has a product that is competitive against Intel processors in terms of value. Intel Core i5 series are aimed by budget PC builders and gamers since they offer about the same performance minus the multi-threading which to be honest doesn't matter much in gaming and even in the most commonly used applications.
AMD priced their FX-Vishera lineup at such a competitive price range that those going for Intel processors would have to second guess their choice. At $195 US, the FX-8350 provides performance that's competetive against Intel's Core i5 and enough to run even the most demanding games and apps released without hurting your wallet. And if you're looking to gain performance, than it would be great for you to hear that my testing proved the Piledriver processors to be great and stable clockers. These chips already run at 4.4 GHz but if you want to push them up to 5 GHz, then its achievable with a mid-tier cooler. Although i would suggest a high-end motherboard such as the Gigabyte 990FX-UD7 or ASUS Crosshair IV Formula for overclocking just so you get the right amount of features and a stable overclocking motherboard.
So, the AMD FX-8350 is in short Bulldozer done right. If this was Bulldozer in the first place than AMD would have been in a far better position than they were with the FX-8150.