AMD’s new flagship processor offers a decent boost in speed over its predecessors but without significantly improved energy efficiency or pricing, it fails to alter the desktop CPU landscape.
November 20, 2012 by Lawrence Lee
| Product | AMD FX-8350 AM3+ Processor |
| Manufacturer | |
| Street Price | US$210 |
The start of 2011 saw Intel release their Sandy Bridge CPU platform to the world to high acclaim. The new chips delivered strong performance, superb energy efficiency, and to ice the cake, each and every processor came equipped with relatively fast integrated graphics. Not content with their sizable performance lead over AMD, Intel raised the bar rather than resting on their laurels, placing even more pressure on the perennial underdog to come up with an answer.
All eyes fell on AMD in the summer when their Bulldozer microprocessor architecture was due to launch, but after a series of delays, it was well into autumn before any of the new silicon made it into outside hands. The result of all this anticipation was only bitter disappointment. The new FX line, codename Zambezi, set an unfortunate precedent — it was actually slower in some tests than its predecessors. The flagship 8-core FX-8150 turned only to be only a minor improvement overall compared to their previous fastest multi-core chip, the Phenom II X6 1100T.
 The AMD FX-8350. |
We’ve already gotten a sneak peak at what the updated CPU cores can do; the recently released Trinity APU platform was based on the same Piledriver architecture AMD is calling Vishera for their high-end desktop AM3+ socket. Unfortunately, it’s an iterative design — the manufacturing process, die layout, and hardware features remain the same, though the inner workings have been tinkered with to achieve higher clock speeds without raising the thermal envelope. Their new headliner, the FX-8350, and the older FX-8150 are both 125W chips but the newcomer has a 400 MHz advantage in base clock speed. With a street price of the US$210, the FX-8350 is also $35 cheaper than the previous flagship was at launch.
| AMD AM3+ CPU Lineup | ||||||
| Model | Cores | Clock (Max Turbo) | L2 Cache | L3 Cache | TDP | Street Price (US) |
| FX-8350 | 8 | 4.0 GHz (4.2 GHz) | 8MB | 8MB | 125W | $210 |
| FX-8150 | 8 | 3.6 GHz (4.2 GHz) | 8MB | 8MB | 125W | $190 |
| FX-8320 | 8 | 3.5 GHz (4.0 GHz) | 8MB | 8MB | 125W | $180 |
| FX-8120 | 8 | 3.1 GHz (4.0 GHz) | 8MB | 8MB | 125W | $160 |
| FX-6300 | 6 | 3.5 GHz (4.1 GHz) | 6MB | 8MB | 95W | $140 |
| FX-6200 | 6 | 3.8 GHz (4.1 GHz) | 6MB | 8MB | 125W | $135 |
| FX-6100 | 6 | 3.3 GHz (3.9 GHz) | 6MB | 8MB | 95W | $115 |
| FX-4300 | 4 | 3.8 GHz (4.0 GHz) | 4MB | 4MB | 95W | $130 |
| FX-4170 | 4 | 4.2 GHz (4.3 GHz) | 4MB | 8MB | 125W | $125 |
| FX-4100 | 4 | 3.6 GHz (3.8 GHz) | 4MB | 8MB | 95W | $105 |
| Vishera processors highlighted in yellow. | ||||||
A smattering of Vishera chips have been released to complement AMD’s current series of processors based on the older Zambezi core. Notable additions include the FX-6300 and FX-4300, which lead their respective hex and quad core lineups in price despite having slower clock speeds than the previous leaders, taking advantage of user preference for more energy efficient 95W parts.
 |
When considering the cost of a system, the CPU is only part of the equation
as the price of motherboards varies greatly from platform to platform. In the
chart above, we added the current street price of the chips compared today to
those of an average compatible motherboard from Newegg.
The following criteria were used for the motherboards: retail versions, Asus/Intel/Gigabyte/MSI
branded, microATX/ATX form factor, SATA 6 Gbps and USB 3.0 controllers (outrageously
priced models were omitted). The average motherboard price turned out to be
US$261 for LGA2011, US$137 for LGA1155, and US$115 for AM3+.
TEST METHODOLOGY
Common CPU Test Configuration:
- Asus EN9400GT Silent Edition
graphics card – 512MB - Western Digital VelociRaptor
hard drive – 300GB, 10,000RPM, 16MB cache - Seasonic
SS-400ET ATX power supply - Microsoft
Windows 7 operating system – Ultimate, 64-bit
AMD AM3:
- AMD Phenom II X4 965 Black Edition
processor – 3.4 GHz, 45nm, 125W - AMD Phenom II X6 1100T Black Edition
processor – 3.3 GHz, 45nm, 125W - Asus M4A78T-E motherboard
– 790GX chipset - Kingston HyperX LoVo 4GB memory – 2x2GB, DDR3-1600 @ 1333 MHz, 9-9-9-24
- Scythe Kabuto
CPU cooler – stock fan at 800RPM
AMD AM3+:
- AMD FX-8150
processor – 3.6 GHz, 32nm, 125W - AMD FX-8350
processor – 4.0 GHz, 32nm, 125W - Asus Crosshair V Formula motherboard
– 990FX chipset - Asus Sabertooth 990FX R2.0 motherboard
– 990FX chipset - Kingston HyperX LoVo 4GB memory – 2x2GB, DDR3-1600 @ 1333 MHz, 9-9-9-24
- Scythe Kabuto
CPU cooler – stock fan at 800RPM
Intel LGA1155:
- Intel Core i5-2500K
processor – 3.3 GHz, 32nm, 95W - Intel Core i7-2600K processor
– 3.4 GHz, 32nm, 95W - Intel Core i7-3770K processor – 3.5 GHz, 22nm, 77W
- ASUS P8P67 motherboard
– P67 chipset - Intel DZ77GA-70K motherboard – Z77 chipset
- Kingston HyperX LoVo 4GB memory – 2x2GB, DDR3-1600 @ 1333 MHz, 9-9-9-24
- Scythe Kabuto
CPU cooler – stock fan at 800RPM
Intel LGA2011:
- Intel Core i7-3960X
processor – 3.3 GHz, 32nm, 130W - Intel DX79SI motherboard
– X79 chipset - Kingston HyperX LoVo 4GB memory – 2x2GB, DDR3-1600 @ 1333 MHz, 9-9-9-24
- Corsair
XMS3 memory 2x2GB, DDR3-1600 @ 1333MHz, 9-9-9-24 (combined with Kingston kit for quad channel operation) - Intel RTS2011LC liquid CPU cooler – stock fan/pump at 800RPM
Measurement and Analysis Tools
- CPU-Z
to monitor CPU frequency and voltage. - SpeedFan
to monitor CPU temperatures. - Real Temp
to monitor CPU temperatures. - CPUBurn
CPU stress software. - Prime95
CPU stress software. - FurMark
GPU stress software. - Media Player
Classic – Home Cinema to play H.264/VC-1 video. - Adobe
Photoshop as an image manipulation benchmark. - Eset NOD32 as
an anti-virus benchmark. - WinRAR as an
archiving benchmark. - iTunes
an audio encoding benchmark. - TMPGEnc
Xpress as a video encoding benchmark. - HandBrake as a
video encoding benchmark - Seasonic
Power Angel AC power meter, used to measure the power consumption
of the system. - Custom-built, four-channel variable DC power supply, used to power
and regulate the CPU fan.
Timed Benchmark Test Details
- NOD32: In-depth virus scan of a folder containing 32 files of varying
size with many RAR and ZIP archives. - WinRAR: Archive creation with a folder containing 68 files of varying
size (less than 50MB). - iTunes: Conversion of an MP3 file to AAC.
- TMPGEnc: Encoding a XVID AVI file with VC-1.
- HandBrake: Encoding a XVID AVI file with H.264.
- Photoshop: Image manipulation using a variety of filters, a derivation
of Driver Heaven’s Photoshop
Benchmark V3 (test image resized to 4500×3499).
Video Test Clip
 1080p | 24fps | ~22 mbps | H.264/MKV: A custom 1080p H.264 encoded clip inside an Matroska container. |
Testing Procedures
Our main test procedure is a series of benchmarks, timed tests of real-world applications. System power consumption (AC) is measured with a Seasonic Power Angel during these tests (an average of the first 10~15 seconds) as well as at idle, during playback of a 1080p H.264 encoded clip, and during full CPU load. To stress the CPU we use either Prime95 (large FFTs setting) or CPUBurn depending on which
produces higher system power consumption. The AC system power is then later converted to DC.
Certain services and features like Superfetch and System Restore are disabled
to prevent them from affecting our results. Aero glass is left enabled if supported.
We also make note if energy saving features like Cool’n’Quiet and SpeedStep
do not function properly.
Estimating DC Power
The following power efficiency figures were obtained for the
Seasonic SS-400ET used in our test system:
| Seasonic SS-400ET Test Results | |||||||
| DC Output (W) | 21.2 | 41.6 | 60.2 | 81.9 | 104.7 | 124.1 | 145.2 |
| AC Input (W) | 32.0 | 58.0 | 78.0 | 102.0 | 128.0 | 150.0 | 175.0 |
| Efficiency | 66.3% | 71.7% | 77.1% | 80.3% | 81.8% | 82.8% | 83.0% |
This data is enough to give us a very good estimate of DC demand in our
test system. We extrapolate the DC power output from the measured AC power
input based on this data. We won’t go through the math; it’s easy enough
to figure out for yourself if you really want to.
Performance
|
|
|
|
|
|
|
|
|
|
|
|
The FX-8350 finished quicker than the FX-8150 in all of our tests by an average of 11%. As both chips have the same maximum Turbo Core speed of 4.2 GHz, the improvements in single-threaded applications suggests the updated cores might be inherently faster. We saw more variation in the power consumption numbers but the overall energy efficiency was similar.
While an 11% boost in speed is nothing to sneeze at, it’s still not enough to be competitive with Intel’s offerings, at least not across the board. In applications that don’t use all eight cores, the FX chips were noticeably slower than its Sandy and Ivy Bridge counterparts. There’s also a caveat regarding the multi-threaded programs they do excel at — poor power efficiency.
Energy Efficiency
|
|
|
|
On light load, the FX-8350’s power draw was essentially the same as its predecessor. On heavy load, the higher base clock speed resulted in a noticeable increase. When stressed, it eclipsed even the Sandy Bridge Extreme i7-3960X equipped with quad channel memory.
|
|
For users with balanced workloads, we’ve determined what we call the “average power consumption” which assumes the system is used half the time for light load activities (an average of idle and H.264 playback) and the remaining half for heavy load (an average of the power consumption used running our five benchmarks). We believe this is a very common usage pattern for an average PC — they are often left on for long periods of time, doing little to no work.
In this scenario, the difference between the FX-8150 and FX-8350 is marginal with both chips flirting with the 100W mark.
|
|
For users with heavy workloads, the total power consumed while running our benchmark suite is of pertinent interest. The total power takes into account the energy efficiency of each CPU while running our benchmark tests as well as how quickly they complete each task. This simulates the power draw of a machine that is purely for doing work and shuts down when its job is finished.
As the FX-8350’s uses a similar amount of power as the FX-8150 but touts stronger performance, it held a power advantage of 4.2 Watt-hours. It’s a significant amount, at least compared to AMD’s previous offerings. As usual, Intel’s lineup was still well ahead.
Performance Analysis
|
|
We arrived at our overall performance figures by giving each CPU a proportional
score in each real world benchmark with each test having an equal weighting.
The scale has been adjusted so that the FX-8350 is the reference point with
a score of 100.
The FX-8350’s performance leap over the FX-8150 is considerably more than the FX-8150 had over its predecessor, the X6 1100T, but it still lags behind popular Intel chips like the i5-2500K. It’s disappointing to see AMD’s best still trail Intel chips released almost two years prior.
|
|
Dividing the overall performance by the platform street cost (CPU plus an average
priced motherboard) gives us the performance per dollar, re-weighted with
the FX-8350 at 100 points.
Competitive pricing has been an important strategy for AMD’s viability throughout most of its history and here it saves the FX-8350 from domination. With regards to initial platform costs, the FX-8350 delivers almost the exact same bang for your buck as the i5-2500K.
|
|
To determine performance per watt, we divided the overall performance score
by the average power consumption calculated earlier and again re-scaled with the FX-8350
as our reference.
If energy efficiency is your biggest concern, the results are overwhelmingly in favor Intel. The i7-3770K is a particularly adept specimen touting twice as much performance per Watt than the FX-8150 while the rest of Intel’s lineup have an advantage of at least 45%.
FINAL THOUGHTS
The FX-8350 is faster across the board than the FX-8150 but like its predecessor, it’s hampered by uninspiring single-threaded performance and crummy energy efficiency. In both regards, there’s a lot of ground to make up if AMD want to catch Intel. Sandy Bridge chips, which are approaching their second anniversary, hold a noticeable lead in all areas except for price. The one place FX-8350 might have the advantage is in performance per dollar, but only against high-end models like the i7-3770K. In this metric, it effectively ties the Core i5-2500K, a processor which is much closer to the FX-8350’s price-point.
AMD’s new flagship offers a decent boost in speed over its predecessor but the upgrade isn’t substantial enough to alter the desktop CPU landscape in any meaningful way. Like the FX-8150, the FX-8350 is only a suitable choice for a small subsection of desktop users: those with heavy, multi-threaded workloads and complete ambivalence toward power consumption. If for example, you primarily use your PC for video encoding/rendering and your utilities are included with your rent, Piledriver/Vishera might be for you. For the rest of us, it’s hard to stomach and impossible to rationalize.
Our thanks to AMD
and ASUS for the FX-8350 and Sabertooth 990FX R2.0 samples used in this review.
* * *
Articles of Related Interest
AMD A10-5700 APU: Trinity at 65W
AMD Trinity: A10-5800K & A8-5600K 2nd Gen APUs
Intel Core i7-3770 Ivy Bridge CPU
Intel Sandy Bridge Extreme: Core i7-3960X LGA2011 Processor
AMD FX-8150 8-Core Bulldozer Processor
AMD A8-3850 Quad Core Desktop APU (updated July 10)
* * *
