The notebooks are both equipped with discrete graphics cores, ATI Mobility Radeon X1700 and Nvidia GeForce Go 7400. We tested them in three versions of 3DMark: 3DMark 2003 3.6.0, 3DMark 2005 1.2.0 and 3DMark 2006 1.0.2.
Compared with the previous version, 3DMark 2005 uses Shader Model 2.0x/3.0 instead of Shader Model 1.x, provides full compatibility with Shader Model 2.0, includes more complex tests (over a million polygons per each frame), and employs normal maps. 3DMark 2006 brings support for HDR, Uniform Shadow Maps, and multi-core CPUs. It is overall oriented at Shader Model 3.0, but two out of its four graphics tests work within the Shader Model 2.0 framework.
The difference in the notebooks’ graphics performance is remarkable. The ATI Mobility Radeon X1700 shows itself as a rather fast GPU when powered from the mains (it is also aided by the high-frequency system memory). But as soon as the notebook is disconnected from the wall outlet, it suffers a fourfold performance hit. Note that the CPU tests from 3DMark 2003 and 2005 depend on GPU performance whereas 3DMark 2006 benchmarks the CPU only.
You can also note that the A8Jp didn’t pass the Shader Particles Test. It is because the ATI Radeon X1000 architecture doesn’t support the vertex texture fetch feature, which is only available on the GeForce 6 and 7 architectures.
Next, we tested the notebooks in two modes in Quake 3 :
- 640x480; 16 bit; Vertex Lighting; Low Detail; 16-bit Texture Quality; Bilinear Texture Filter
- 1024x768; 32 bit; Lightmap Lighting; High Detail; 32-bit Texture Quality; Trilinear Texture Filter
And in one mode in Quake 4 :
- Overall Quality – High; Resolution – 1024x768; Format – 4:3; Multi-core Optimization – Yes. Other settings were left default
There was no standard demo record in Quake 4 , so we had to create one by ourselves. We will use it in every following review of notebooks on our site so that different notebooks could be compared under identical conditions.
Again, the ATI Mobility Radeon X1700 is very good when powered from the mains, but not from the battery. Nvidia’s GeForce solutions do not suffer a big performance hit on switching to the battery even in Quake 4 , which is not tied to CPU performance. ATI’s solution behaves differently. It slows down greatly to save more power, and even the faster system memory of the A8Jp cannot help it here.