NVIDIA GeForce GTX 280 and GTX 260 Unleashed

Iria · Jun 16, 2008, 02:13 PM

“The high-end 280 card is powered by NVIDIA's 1.4 billion transistor GT200 GPU, produced on TSMC's 65nm process node and is the largest, most complex chip TSMC has ever manufactured. Beyond just a larger number of stream processing units, the GT200 also supports three times the number of threads in flight, as NVIDIA's previous G80, at any given time. It also has a new scheduler design that is up to 20% more efficient, along with a wider 512-bit memory interface with improved z-cull and compression technology. Perhaps what's most interesting about this GPU however, are its two operational states, one as a 3D Graphics processor and another standard compute mode configuration. NVIDIA's new behemoth chip not only lights up the gaming benchmarks but also can be specifically reprogrammed for more efficient general purpose GPU (GPGPU) computing usage models like Folding and supports atomic read-modify-write operations in memory.”
_______________
Source: HotHardware
Check out our own review here.

Jun 16, 2008, 02:13 PM	#1
Iria DriverHeaven Extreme Member Join Date: Apr 2004 Posts: 7,275 Rep Power: 89	NVIDIA GeForce GTX 280 and GTX 260 Unleashed “The high-end 280 card is powered by NVIDIA's 1.4 billion transistor GT200 GPU, produced on TSMC's 65nm process node and is the largest, most complex chip TSMC has ever manufactured. Beyond just a larger number of stream processing units, the GT200 also supports three times the number of threads in flight, as NVIDIA's previous G80, at any given time. It also has a new scheduler design that is up to 20% more efficient, along with a wider 512-bit memory interface with improved z-cull and compression technology. Perhaps what's most interesting about this GPU however, are its two operational states, one as a 3D Graphics processor and another standard compute mode configuration. NVIDIA's new behemoth chip not only lights up the gaming benchmarks but also can be specifically reprogrammed for more efficient general purpose GPU (GPGPU) computing usage models like Folding and supports atomic read-modify-write operations in memory.” _______________ Source: HotHardware Check out our own review here. Last edited by Iria; Jun 16, 2008 at 02:34 PM.