iXBT Labs - Computer Hardware in Detail

Platform

Video

Multimedia

Mobile

Other

AMD Phenom X4 Memory Controller in the Ganged/ Unganged Mode

In real-life applications.

September 16, 2008



<< Previous page

     Next page >>

Professional photo processing


Test Ganged Unganged Ratio
Photoshop
Blur 0:06:14 0:06:12 -0.5%
Color 0:01:24 0:01:24 0.0%
Filters 0:06:06 0:06:06 0.0%
Light 0:02:14 0:02:14 0.0%
Rotate 0:02:48 0:02:46 -1.2%
Sharp 0:02:48 0:02:46 -1.2%
Size 0:01:02 0:01:02 0.0%
Transform 0:02:22 0:02:22 0.0%
Group Score 94.1 94.4 -0.4%

In this group the Ganged mode is either on par or losing by similarly negligible values.

Scientific suites


Test Ganged Unganged Ratio
Maple 0.025711665 0.0258 -0.2%
Mathematica
Internal 3.0290 3.0390 -0.3%
MMA 1.1388 1.1242 1.3%
MATLAB
LU 0.0851 0.0833 -2.1%
FFT 0.1593 0.1582 -0.7%
ODE 0.2710 0.2697 -0.5%
Sparse 1.7875 1.8073 1.1%
2D 0.3338 0.3368 0.9%
3D 0.5745 0.5702 -0.7%
Group Score 65.1 65.1 -0.1%

MATLAB gets close the "record" of 3ds max, but in general it's all same as in the previous tests.

Web server


Test Ganged Unganged Ratio
PHPCalculator 176 176 0.0%
PHPSpeed
Synthetic PHP 2866 2746 4.4%
SyntheticMySQL 2629 2624 0.2%
SyntheticRead/Write 1898 1884 0.7%
Real WorldPHP 4266 4051 5.3%
Real WorldPHP & MySQL 2472 2466 0.2%
Server 1877 1881 -0.2%
Group Score 99.2 97.7 1.5%

At last we can see some more or less significant values of up to 5%. It's interesting that they don't speak well for the Unganged mode.

Archiving


Test Ganged Unganged Ratio
7-Zip 0:02:33 0:02:32 -0.7%
WinRAR 0:01:23 0:01:23 0.0%
Ultimate ZIP 0:02:19 0:02:19 0.0%
Group Score 96.2 96.4 -0.2%

The difference is negligible.

Media encoding


Test Ganged Unganged Ratio
FLAC 0:01:13 0:01:13 0.0%
LAME 0:02:04 0:02:04 0.0%
Musepack 0:02:17 0:02:17 0.0%
Vorbis 0:04:42 0:04:42 0.0%
CanopusProCoder 0:05:17 0:05:17 0.0%
DivX 0:01:21 0:01:21 0.0%
x264 0:02:44 0:02:44 0.0%
XviD 0:09:42 0:09:42 0.0%
Group Score 93.3 93.3 0.0%

And in this group there's no difference at all.

Games


Test Ganged Unganged Ratio
Call of Duty 4 116 124 -6.5%
Company ofHeroes 55 55 0.0%
Call ofJuarez 46 46 0.0%
Crysis 19.14 19.69 -2.8%
S.T.A.L.K.E.R. 112 112 0.0%
UnrealTournament 3 96 100 -4.0%
World inConflict 45 45 0.0%
Group Score 93.0 94.8 -1.9%

Games definitely favor the Unganged mode. At that 3 of 7 show rather considerable values. Call of Duty is this article champion, showing the record 6.5% difference between the Ganged and Unganged modes (Unganged wins).

Non-professional photo processing


Test Ganged Unganged Ratio
ACDSee 0:12:12 0:11:28 -6.0%
IrfanView 0:22:33 0:22:47 1.0%
Paint.NET 31151 30972 -0.6%
xat.com Image Optimizer 0:40:57 0:41:58 2.5%
XnView 0:25:24 0:25:53 1.9%
Group Score 82.9 83.1 -0.3%

ACDSee is another application rather sensitive to the mode of AMD Phenom memory controller. And it also likes the Unganged mode.

Conclusions


Test Ganged Unganged Ratio
PRO SCORE 89.4 89.0 0.5%
HOME SCORE 91.3 91.9 -0.6%
OVERALL SCORE 90.4 90.5 -0.1%

In general, this is a draw. With certain rare exceptions, we can't state an obvious advantage of either mode. Moreover, there are no trends or regularities related to application class, including single- and multi-threaded nature, which could at least conform to the official positioning of these modes. The 0.1% total loss of the Ganged mode doesn't indicate anything, since another application can turn it into a complete draw or even a victory.

Thus, our conclusion is brief: in terms of real-life desktop software, it's generally the same, in which mode Phenom's memory controller is working. There are certain rare exceptions, but their character can hardly be predicted (at least, based on our test).

As for the question if we were right, the 0.1 point (0.1%) difference is hardly a reason to start using the Unganged mode in primary tests. Especially as the Ganged mode is motherboard default, and even the 10 times higher value (1%) would still be negligible as a possible measurement error.


Memory modules provided by Corsair Russia.

Write a comment below. No registration needed!


<< Previous page

Article navigation:

Page 1: Introduction, tests

Page 2: More tests, conclusions



blog comments powered by Disqus

  Most Popular Reviews More    RSS  

AMD Phenom II X4 955, Phenom II X4 960T, Phenom II X6 1075T, and Intel Pentium G2120, Core i3-3220, Core i5-3330 Processors

Comparing old, cheap solutions from AMD with new, budget offerings from Intel.
February 1, 2013 · Processor Roundups

Inno3D GeForce GTX 670 iChill, Inno3D GeForce GTX 660 Ti Graphics Cards

A couple of mid-range adapters with original cooling systems.
January 30, 2013 · Video cards: NVIDIA GPUs

Creative Sound Blaster X-Fi Surround 5.1

An external X-Fi solution in tests.
September 9, 2008 · Sound Cards

AMD FX-8350 Processor

The first worthwhile Piledriver CPU.
September 11, 2012 · Processors: AMD

Consumed Power, Energy Consumption: Ivy Bridge vs. Sandy Bridge

Trying out the new method.
September 18, 2012 · Processors: Intel
  Latest Reviews More    RSS  

i3DSpeed, September 2013

Retested all graphics cards with the new drivers.
Oct 18, 2013 · 3Digests

i3DSpeed, August 2013

Added new benchmarks: BioShock Infinite and Metro: Last Light.
Sep 06, 2013 · 3Digests

i3DSpeed, July 2013

Added the test results of NVIDIA GeForce GTX 760 and AMD Radeon HD 7730.
Aug 05, 2013 · 3Digests

Gainward GeForce GTX 650 Ti BOOST 2GB Golden Sample Graphics Card

An excellent hybrid of GeForce GTX 650 Ti and GeForce GTX 660.
Jun 24, 2013 · Video cards: NVIDIA GPUs

i3DSpeed, May 2013

Added the test results of NVIDIA GeForce GTX 770/780.
Jun 03, 2013 · 3Digests
  Latest News More    RSS  

Platform  ·  Video  ·  Multimedia  ·  Mobile  ·  Other  ||  About us & Privacy policy  ·  Twitter  ·  Facebook


Copyright © Byrds Research & Publishing, Ltd., 1997–2011. All rights reserved.