|
|
Nvidia vs ATI Midrange PCIe GPU Comparison
[Abstract]
The New Cards On The PCIe BlockWith the availability of NVIDIA's nForce4 Media Communication Processor (MCP), the third generation I/O interface (PCI Express bus) has finally arrived on the AMD...
[Content] PCDigitalMobileGame
Overall GPU / VPU Comparison
On this page, we listed the complete specifications of the GeForce 6600 GT, RADEON X700 XT and the RADEON X700 PRO for direct comparison. Additionally, we have also listed the previous generation of mid-range PCIe graphics cards, the GeForce PCX 5900 and RADEON X600 XT for further comparisons.
As we have seen in the high-end graphics processors where the GeForce 6800 series has a much higher transistor count over the RADEON X800 series, this is once again repeated on the mid-range as well. Supporting Shader Model 3.0 requires a lot more registers, buffers and these directly translate to a much higher transistor count on the GeForce 6 series. The difference this time between the GeForce 6600 series and the RADEON X700 series is slightly subdued and that's mainly because the RADEON X700 VPUs have six vertex shaders ?double the amount in the GeForce 6600 GPUs. Still, the GeForce 6600 series has a higher transistor count.
Interestingly, NVIDIA's GeForce PCX 5900 had the largest memory bandwidth among all the cards compared here, but as you will see from the results, this is largely overshadowed by the limited processing capability of the previous generation GPU.
GPU/VPU Core Comparison
| GPU/VPU | NVIDIA GeForce 6600 GT (PCIe) | NVIDIA GeForce PCX 5900 | ATI RADEON X700 XT | ATI RADEON X700 PRO | ATI RADEON X600 XT | | Core Code | NV43 | NV35 | RV410 | RV410 | RV380 | | Transistor Count | ~145 million | 130 million | 120 million | 120 million | 75 million | | Manufacturing Process (microns) | 0.11 | 0.13 | 0.11 | 0.11 | 0.13 | | Core Clock | 500MHz | 350MHz | 475 MHz | 425MHz | 500MHz | | Rendering (Pixel) Pipelines | 8 | 8 | 8 | 8 | 4 | | Pixel Fill Rate (MPixels/s) | 4,000 | 2,800 | 3,800 | 3,400 | 2,000 | | Texture Units per Pipeline | 1 | 1 | 1 | 1 | 1 | | Peak Texture Fill Rate (Mtexels/s) | 4,000 | 2,800 | 3,800 | 3,400 | 2,000 | | Vertex Shaders | 3 | Array | 6 | 6 | 2 | | Memory Clock | 500MHz (1000MHz DDR3) | 275MHz (550MHz DDR1) | 525MHz (1050MHz DDR3) | 430MHz (860MHz DDR3) | 370MHz (740MHz DDR1) | | DDR Memory Bus | 128-bit | 256-bit | 128-bit | 128-bit | 128-bit | | Memory Bandwidth | 16.0GB/s | 17.6GB/s | 16.8GB/s | 13.8GB/s | 11.8GB/s | | NVIDIA CineFX Engine | 3.0 | 2.0 | - | - | - | | NVIDIA Intellisample Technology | 3.0 | 2.0 | - | - | - | | NVIDIA UltraShadow Technology | II | I | - | - | - | | ATI SMARTSHADER | - | - | HD | HD | 2.1 | | ATI SMOOTHVISION | - | - | HD | HD | 2.1 | | ATI HYPER Z | - | - | HD | HD | III+ | | Pixel Shader Support | 3.0 | 2.0a | 2.0b | 2.0b | 2.0 | | Vertex Shader Support | 3.0 | 2.0a | 2.0 | 2.0 | 2.0 | | DirectX Support | Up to DirectX 9.0c | Up to DirectX 9.0 | Up to DirectX 9.0b | Up to DirectX 9.0b | Up to DirectX 9.0 | | FSAA Mode | Multi Sampling (up to 6x) | Multi Sampling (up to 6x) | Multi Sampling (up to 6x) | Multi Sampling (up to 6x) | Multi Sampling (Up to 6x) | | Anisotropic filtering Modes | Up to 16x | Up to 16x | Up to 16x | Up to 16x | Up to 16x | | Other Features | On-Chip Video Processor NVIDIA HPDR Technology DVC 3.0 Support for NVIDIA Cg NView | DVC 3.0 Support for NVIDIA Cg NView | VIDEOSHADER HD 3Dc HYDRA VISION | VIDEOSHADER HD 3Dc HYDRA VISION | VIDEOSHADER TRUEFORM 2.0 HYDRA VISION | | RAMDAC | Dual 400MHz | Dual 400MHz | Dual 400MHz | Dual 400MHz | Dual 400MHz | | TV Output | Int., 1024x768 | Int., 1024x768 | Int., 1024x768 | Int., 1024x768 | Int., 1024x768 | | TMDS transmitter | Int. 165MHz | On-board IC | Int. 165MHz | Int. 165MHz | Int. 165MHz | | PCIe Interface | x16 | x16 (with HSI) | x16 | x16 | x16 | | Molex Power Connectors | NIL | NIL | NIL | NIL | NIL |
|
|
|
