• Home
  • blog
  • Athlon II X4 610e & Phenom II X4 910e: 45W & 65W Quad Cores

Athlon II X4 610e & Phenom II X4 910e: 45W & 65W Quad Cores

blog image

The Athlon II X4 610e and Phenom II X4 910e are quad core processors from AMD with very low thermal design power of 45W and 65W, respectively. Are they as energy efficient as they claim and are the higher price tags justified?

AMD X4 610e & 910e: 45W & 65W Quad Cores Processors

June 27, 2010 by Lawrence Lee

Product Athlon II X4 610e
AM3 Processor
Phenom II X4 910e
AM3 Processor
Manufacturer AMD AMD
List Price US$145 US$175

It has been 4 years since Intel took the reins from AMD with Core 2, claiming
the performance crown and never looking back. To this day, AMD remains dominated
by their Santa Clara rivals, particularly in the high-end processor market.
While the glory days of Athlon 64 are long over, recently it seems they have
returned to their roots, delivering tremendous value with their budget CPUs,
the kind of chips that made them the popular alternative to Intel in the first
place. It is simply amazing that today you can purchase a reasonably fast triple
and quad core AMD desktop processor for as little as US$70 and US$100 respectively.

AMD is also known for producing energy efficient chips, models that are easily
recognizable by a lower case ‘E’ at the end. These CPUs are typically expensive
compared to the standard models and do not offer a great performance to cost
ratio. They do run cooler though and are thus ideal in smaller cases where heat
and the noise associated with dissipating it are problematic. The two main question
are A. are they as energy efficient as AMD claims, and B. are they worth the
extra cost?

AM3/LGA1156 4-thread Processors under US$200
Model
Clock Speed
Cache
(L2+L3)
TDP
Street Price†
Athlon II X4 610e
2.4 GHz
2 MB
45W
$145*
Athlon II X4 630
2.8 GHz
2 MB
95W
$100
Athlon II X4 635
2.9 GHz
2 MB
95W
$100
Athlon II X4 640
3.0 GHz
2MB
95W
$120
Phenom II X4 810
2.6 GHz
2+4 MB
95W
$140
Phenom II X4 910e
2.6 GHz
2+6 MB
65W
$175*
Phenom II X4 925
2.8 GHz
2+6 MB
95W
$140
Phenom II X4 945
3.0 GHz
2+6 MB
95W
$160
Phenom II X4 955 BE
3.2 GHz
2+6 MB
125W
$160
Phenom II X4 965 BE
3.4 GHz
2+6 MB
125W
$180
Core i3-530
2.93 GHz
0.5+4 MB
73W
$115
Core i3-540
3.06 GHz
0.5+4 MB
73W
$140
Core i5-650
3.2 GHz
0.5+4 MB
73W
$180
Core i5-660
3.33 GHz
0.5+4 MB
73W
$200
Core i5-661
3.33 GHz
0.5+4 MB
87W
$200
Core i5-750
2.66 GHz
1+8 MB
95W
$200
prices rounded to nearest US$5
*retail 1000-unit tray price
Tested processors are in bold.

Enter the Athlon II X4 610e and the Phenom II X4 910e, a pair of quad core
chips with thermal design powers of 45W and 65W respectively. The downside is
they are clocked much slower than the competition. We will do our best to quantify
the performance and power difference.


CPU-Z screenshot: X4 610e at load.


CPU-Z screenshot: X4 610e at idle.

The Athlon II X4 610e is a US$145 processor with a TDP of only 45W, a clock
speed of 2.4 GHz, and 2MB (4 x 512KB) of L2 cache. Our sample had very low operating
voltages, 0.853V when idle, and 1.059V on load according to CPU-Z.


CPU-Z screenshot: X4 910e at load.


CPU-Z screenshot: X4 910e at idle.

The Phenom II X4 910e is a US$175 processor with a TDP of 65W, a clock speed
of 2.6 GHz, 2MB (4 x 512KB) of L2 cache, 6MB of L3 cache (L3 cache is the main
difference between the Phenom II and Athlon II). Our sample had operating voltages
of 0.982V when idle and 1.147V on load according to CPU-Z.

TEST METHODOLOGY

Common Test Platform:

Intel LGA1156:

AMD AM3:

  • AMD Phenom II X4 965 Black Edition processor – 3.4 GHz, 45nm, 125W
  • AMD Phenom II X4 955 Black Edition
    processor – 3.2 GHz, 45nm, 125W
  • AMD Phenom II X4 910e processor – 2.4 GHz, 45nm, 65W
  • AMD
    Phenom II X4 810
    processor – 2.6 GHz, 45nm, 95W
  • AMD Athlon II X4 635 processor – 2.9 GHz, 45nm, 95W
  • AMD Athlon II X4 630 processor
    – 2.8 GHz, 45nm, 95W
  • AMD Athlon II X4 610e processor – 2.6 GHz, 45nm, 95W
  • Asus M4A78T-E motherboard
    – 790GX chipset

Measurement and Analysis Tools

  • CPU-Z
    to monitor CPU frequency and voltage.
  • SpeedFan
    to monitor CPU temperatures.
  • Real Temp
    to monitor CPU temperatures.
  • CPUBurn
    processor stress software.
  • Prime95
    processor stress software.
  • Media Player
    Classic – Home Cinema
    to play x264/MKV video using DXVA.
  • 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
  • Adobe
    Photoshop CS2
    as an image manipulation benchmark.
  • PCMark05
    as a general system benchmark.
  • Seasonic
    Power Angel
    AC power meter, used to measure the power consumption
    of the system.
  • Digital Multimeters used to measure the current and voltage across
    the AUX12V connector
  • 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).

Our first test procedure is designed to determine the overall system power
consumption at various states (measured using a Seasonic Power Angel). To stress
CPUs we use either Prime95 (large FFTs setting) or CPUBurn depending on which
produces higher system power consumption. For video playback we use a x264-encoded
1080p clip played with Media Player Classic – Home Cinema.


1080p | 24fps | ~14mbps

x264 1080p: Spaceship is a 1080p x264 clip encoded from the
Blu-ray version of an animated short film. It features a hapless robot
trying to repair a lamp on a spaceship.

Secondly, we run a series of benchmarks, most of which are timed tests of real-world
applications. During the first 10~15 seconds, AC system power is measured at
the wall (and later converted to DC) and current/voltage measurements are taken
at the AUX12V connector to determine how much power is drawn by the CPU and
VRM circuitry alone.

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.

TEST RESULTS

Operating Voltages

Comparison: Operating Voltages (CPU-Z)
Model
Clock Speed
Idle
Load
Athlon II X4 610e
2.4 GHz
0.853V
1.059V
Athlon II X4 630
2.8 GHz
1.035V
1.287V
Athlon II X4 635
2.9 GHz
1.085V
1.372V
Phenom II X4 810
2.6 GHz
1.059V
1.317V
Phenom II X4 910e
2.6 GHz
0.982V
1.147V
Phenom II X4 955 BE
3.2 GHz
0.982V
1.305V
Phenom II X4 965 BE
3.4 GHz
0.959V
1.311V
Core i5-661
3.33 GHz
0.936V
1.168V
Core i5-750
2.66 GHz
0.864V
1.240V

Before we jump into power consumption numbers, please note the operating voltages
of the processors tested today as sort of a disclaimer. Different samples of
the same processor often run at slightly different operating voltages which
can greatly affecting power consumption. With Cool’n’Quiet and C1E enabled,
most AMD chips typically run with a core voltage of about 1.0V when idle and
1.3V on load.

System Power Consumption

The X4 610e comes out on top in general system power consumption, but it only
beats the Core i5-661 by a small margin. The X4 910e ranks third, sandwiched
between the two Core i5s and uses about 30W less on load than its closest analog,
the X4 810. The majority of the rest of the lineup has remarkably similar power
consumption when idle and during video playback. On load, the X4 955 and 965
are extremely energy demanding, using 42~54W more than the Core i5-750.

CPU + VRM Power Consumption

When we measure power directly from the AUX12V connector, the X4 610e is the
thriftiest once again at idle and during video playback. On load, all the processors
except the X4 635 use less than their rated thermal design power, though if
you included the unknown amount of loss to the VRMs, it certainly does not exceed
95W.

Thermals

For our thermal test, each CPU was paired with the Scythe Kabuto with its stock
fan spinning at ~800 rpm. AMD temperatures were taken using the motherboard’s
sensor via SpeedFan and Intel temperatures were taking with RealTemp.


Note: direct comparisons between Intel and AMD processors are not applicable.

No surprises here, the CPU load temperature is proportional to power consumption.
As such, the X4 610e and X4 910e run very cool compared to the 95W and 125W
AMD quads.

PERFORMANCE

PCMark05

PCMark05 is a synthetic benchmark suite with aggregates the results of 11 different
system tests including web page rendering, file decryption, video and audio
editing, and 3D usage.

The i5-661 surprisingly well in PCMark05 despite being only a dual core processor
— its high clock speed definitely makes up for it. The X4 810 comes in
last which is a bit of anomaly, though individual runs of PCMark can vary by
as much as 200 points in our experience.

Adobe Photoshop

Our Photoshop test is a variation of Driver Heaven’s Photoshop Benchmark V3
using a smaller test image. A variety of filters are applied to the image and
the times elapsed are added up.

The more expensive X4 955/965 and Core i5s dominate this test, with the rest
of the field a step behind.

ESET NOD32

Our NOD32 test is a simple anti-virus scan of a folder with dozens of files
including archives. It’s one of the most common tasks performed on a PC.

The X4 610e completes this test with very low power consumption, but its lack
of cache and low clock speed makes its inferiority fairly noticeable, coming
in well behind the X4 810 and the rest of the pack.

WinRAR

Our WinRAR test consists of archiving a RAR file with a large group of files
of varying size.

WinRAR is a program that loves cache, so the Phenom II X4 910e and X4 810 jump
ahead of the Athlon IIs with higher clock speeds.

iTunes

Our iTunes test is simply the encoding of a MP3 file into AAC.

After several split decisions between the X4 955/965 and i5-661/750, the Intel
chips finally best their AMD opponents outright.

TMPGEnc

Our TMPGEnc test consists of encoding an XVID-encoded AVI to VC-1.

Few surprises here except for the i5-661 using a ridiculous 27W less than the
i5-750 during encoding.

HandBrake

A similar test to TMPGEnc, only this time encoding to H.264.

This is the only test where the i5-661’s lack of 4 proper cores really hurts
its performance, falling behind the entire field except for the lowly X4 610e.

Benchmark Power Consumption

Our timed benchmark results multiplied by the average power consumption during
each test gives us an estimate of how much total power was used during testing.

As far as total system power is concerned, none of the AMD processors tested
today come close to the Core i5-661 and i5-750. The 610e and 910e are about
7 watt-hours behind, and the rest of the field are 11 watt-hours back.

FINAL THOUGHTS

Processor Value

To judge performance/value, we gave all our timed benchmarks equal weighting
with the top performer, the Core i5-750, given a score of 100 on our performance
scale. Similarly, the CPU that delivered the best performance per dollar, the
X4 635, was given a score of 100 on the value scale.


Performance and value, sorted by CPU pricing.

On our test suite, the X4 630 and 635 gave us 75~80% performance of the i5-750
but at half the price, confirming that AMD can’t be beat when it comes to pure
value. The Core i5s were very fast, but also priced high, resulting in a much
poorer value score. The two energy efficient X4s are even worse, an undesirable
combination of lukewarm performance and high price.

Unfortunately, just as there is no such thing as a free lunch, there is no
ideal combination of both value (performance/dollar) and power consumption.
The best value processors, the X4 635 and 630 are among the most energy inefficient,
while their low power brothers, the X4 610e and X4 910 only offer minor power
savings coupled with poor value. The Intel chips fare better in this regard,
but keep in mind that we did not factor in motherboard cost. Intel boards typically
the cost 20~40% more than comparable AMD boards, driving down their value further.

Athlon II X4 610e & Phenom II X4 910e

The "E" processors total power consumption during our timed benchmarks
was only 10% better than the faster 95W and 125W AMD quad cores, and still a ways off from Intel’s Core i5s. They are slightly more energy efficient
when idle, but the difference is too small to make a real impact in the grand
scheme of things. In addition, the price premium attached to these processors
makes them rather poor values; equivalent or better performance can be found in the ~$100 Athlon II X4 635. From what we can
gather, the only real selling point of these CPUs is the lower operating temperature… but then again, most standard AMD processors undervolt fairly well.

Our thanks to AMD
and for Athlon/Phenom II X4 samples used in this review.

* * *

Articles of Related Interest
AMD’s
890GX Chipset

Asus P7H55D-M EVO LGA1156 microATX Motherboard
Intel Core i5-661: A 32nm CPU with
Integrated Graphics

Athlon II X4 630 & 620: Affordable
Quad Cores

Intel’s LGA1156 and Lynnfield
core

AMD’s 785G Chipset Boards: 780G Evolved

* * *

Discuss
this article in the SPCR forums.

Leave a Comment

Your email address will not be published. Required fields are marked *