Well ATI is back with a vengeance by releasing its eagerly awaited (by non-nVidia employees) R300 chipset. The R300 chipset is being used to make several different cards for different market segments, but today we will be focusing on the 9700 Pro configuration produced by Sapphire.
The Radeon 9700 Pro sports 8 pixel pipelines with 1 texturing unit per pipeline while the GF4 Ti series sports 4 pixel pipelines with 2 texturing units per pipeline. In terms of theoretical pixel processing power, these two chipsets will be similar. At 300Mhz (Ti4600 speeds), GeForce4 Ti can do 2400 million texels (textured pixels) per second in raw fillrate. At 325MHz, Radeon 9700 Pro can do 2600 million texels per second. In terms of raw pixel fillrate (without being passed through a texture unit), GeForce4 Ti at 300Mhz can do 1200 million pixels per second and Radeon 9700 Pro can do 2600 million pixels per second. Due to having 8 pixel pipelines over the GeForce4 Ti's 4 pipelines, raw pixel fillrate is over 200% higher on Radeon 9700 Pro.
However, memory bandwidth also plays an important role when the GPU is getting taxed, mostly when the resolution is high. Anisotropic texture filtering and full scene anti-aliasing are also features that take up large amounts of bandwidth. The Radeon 9700 Pro has a 256-bit memory bus width and a 620Mhz(310Mhz DDR) memory clock speed while the Geforce4 Ti 4600 has a 128-bit memory bus width and a 650Mhz(325Mhz DDR) memory clock speed. By computing the memory bandwidth, this shows the Radeon has 19.4GB/sec of raw memory bandwidth while the Geforce has 10.2GB/sec. Please note these are the peak theoretical numbers.
Page 1 of 6
