Introduction

Here you are, the NVIDIA GeForce GTX 560 Ti. Making use of NVIDIA's proven second generation Fermi architecture, this GPU targets a key price point which has been NVIDIA and ATI's hunting ground since 2006. The sub-$300 segment is where customers have learned to expect high-end like performance and features at compelling prices. Price-performance ratio is the king here. The performance-price sweet spot is a virtual G-spot for GPU manufacturers. Whoever hits it right, gets loads of...sales. Veterans in this segment include the GeForce 7900 GT, Radeon X1950 Pro, GeForce 8800 GT, Radeon HD 4850, GeForce GTX 260-216, GeForce GTX 460, Radeon HD 6870, and in comes the latest contender, the GTX 560 Ti. The model name invokes some nostalgia as SKUs carrying the "Ti" marker were some of NVIDIA's first with programmable shaders. While the GTX 560 Ti isn't a "first" in anything as far as feature-set goes, I think "Ti" has more to do with shaping up the brand, telling buyers that the product has a little more to offer for its price, and that it's a step above the price point GTX 460 set for itself, while remaining to be a performance segment model.

Getting into the fine print of NVIDIA's offer, the GeForce GTX 560 Ti is based on NVIDIA's new GF114 chip. As far as specifications and transistor-count go, it is identical to the GF104 on which GTX 460 was based, except that it has all 384 of the CUDA cores physically present enabled, and that it uses the same secret-sauce (read: electrical enhancements) that made GF110, an evolved clone of the GF100, totally rock with power consumption figures. 384 CUDA cores apart, there's a 256-bit wide GDDR5 memory interface, 32 ROPs, branched geometry processing, and the immediate fruition of the electrical enhancements, clock speeds: 822 MHz core, 1640 MHz CUDA cores, and 1000 MHz (4.00 GHz effective) memory. As far as features go, the GTX 560 Ti doesn't come with anything we haven't seen already with the GTX 460, it's all about performance per watt/dollar in this round.