Asus EN7950GX2 - GeForce 7950 GX2


Our Test System and 3DMark06

HOW WE CONFIGURED THE TEST SYSTEMS: We tested the NVIDIA based cards on an Asus A8N32-SLI nForce 4 SLIX16 chipset based motherboard. The ATI powered cards, however, were tested on an A8R32-MVP motherboard based on the CrossFire Xpress 3200 chipset. Both systems used the same AMD Athlon 64 FX-60 dual-core processor and 2GB of low-latency Corsair XMS RAM, though. The first thing we did when configuring these test systems was enter each BIOS and loaded their "High Performance Defaults."  The hard drive was then formatted, and Windows XP Professional with SP2 was installed. When the installation was complete, we installed the latest chipset drivers available, installed all of the other drivers necessary for the rest of our components, and removed Windows Messenger from the system.  Auto-Updating and System Restore were also disabled, the hard drive was defragmented, and a 1024MB permanent page file was created on the same partition as the Windows installation. Lastly, we set Windows XP's Visual Effects to "best performance," installed all of the benchmarking software, and ran the tests.

The HotHardware Test Systems
AMD Athlon 64 FX Powered

Processor -

Motherboard -





Video Cards -



Memory -


Audio -

Hard Driv
e -

 

Hardware Used:
AMD Athlon 64 FX-60 (2.6GHz x 2)

Asus A8N32-SLI
nForce4 SLIX16 chipset

Asus A8R32-MVP
ATI CrossFire Xpress 3200

GeForce 7950 GX2
GeForce 7900 GTX (x2)

Radeon X1900 XTX (x2)

2048MB Corsair XMS PC3200 RAM
CAS 2

Integrated on board

Western Digital "Raptor"

74GB - 10,000RPM - SATA

OS -
Chipset Drivers -
DirectX -

Video Drivers
-



DirectX -
DirectX -
DirectX -
DirectX -
DirectX -
OpenGL -
Relevant Software:
Windows XP Professional SP2
nForce Drivers v6.85
DirectX 9.0c (March Redist)

NVIDIA Forceware v91.29

ATI Catalyst v6.5

Benchmarks Used:
3DMark06 v1.0.2
Splinter Cell: Chaos Theory v1.05
Oblivion*
F.E.A.R. v1.05
Half Life 2: Lost Coast
Quake 4 v1.2*

* - Custom Test (HH Exclusive demo)

Quad SLI Modes -
Before we jump into our benchmarking scores, let's talk a bit about NVIDIA's Quad SLI technology, which we know isn't quite ready for prime time since it is not "officially" supported for the end user DYI market.  Quad SLI is available in NVIDIA's drivers however, when two GeForce 7950GX2s are installed and the driver check box enabled.  There are three new multi-GPU modes that are made available when Quad SLI is enabled.

Quad AFR:
This is a straight forward method of Alternate Frame Rendering for the four GPUs, in an dual GeForce 7950 GX2 configuration, to render alternate frames.  GPU1 takes the first frame, GPU2 the next, GPU3 the third and so on.  All GPUs in the system are more or less equally load-balanced and this is actually the most efficient method of employing Quad SLI.  As we'll show you in the benchmark testing ahead, to our knowledge Quake 4 is the only the game engine NVIDIA is running Quad AFR mode.

Quad SFR: 
Quad SFR load balances four GPUs, splitting up the a single frame into four individual slices (Split Frame Rendering).  This algorithm is the lest efficient and also takes the most resources since the machine must calculate on a continual basis how much of the frame workload is to be allocated to each GPU in the system.

AFR of SFR:
As the name implies, this is a combination of the two methods detailed above.  GPUs are paired on each board in SFR mode and then each pair is allocated workload in Alternate Frame Rendering mode.  This is the method we noted was being run in most any DirectX game engine we tested and in the 3DMark06 tests you'll see here next.

Performance Comparisons with 3DMark06 v1.0.2
Details: http://www.futuremark.com/products/3dmark06/

3DMark06
Futuremark recently launched a brand-new version of their popular benchmark, 3DMark06. The new version of the benchmark is updated in a number of ways, and now includes not only Shader Model 2.0 tests, but Shader Model 3.0 and HDR tests as well. Some of the assets from 3DMark05 have been re-used, but the scenes are now rendered with much more geometric detail and the shader complexity is vastly increased as well. Max shader length in 3DMark05 was 96 instructions, while 3DMark06 ups the number of instructions to 512. 3DMark06 also employs much more lighting, and there is extensive use of soft shadows. With 3DMark06, Futuremark has also updated how the final score is tabulated. In this latest version of the benchmark, SM 2.0 and HDR / SM3.0 tests are weighted and the CPU score is factored into the final tally as well.


NVIDIA's Quad SLI
AFR of SFR Mode In 3DMark06

 

3DMark06 is not an application that benefits much from Quad SLI.  As you can see we only gained about 5% more performance according to 3DMark's scoring methods which is hardly worth the cost of another Asus N7950GX2.  However, we're also very certain that no one would buy a second 7950 GX2 card simply for the purposes of running 3DMark06.  Beyond that the Asus EN7950GX2 scored in line with our stock reference GeForce 7950 GX2, offering  approximately 30+% more performance over the fastest single high-end card in these tests.  The EN7950GX2 does fall short of both the Radeon X1900 XT CrossFire setup and the GeForce 7900 GTX SLI setup by about 15%, but when you consider the cost of each of those dual-card combinations (~$800-$900), the EN7950 GX2's performance seems a bit more cost efficient overall.  Let's move on to some real-world game testing next.


Tags:  Asus, GeForce, X2, force, 7950

Related content