Now let’s take a deeper look at the new GPU. First of all, it has 240 Shader Processors! Yes, that’s almost doubled amount than G92 had. Of course that is not all. The number of texture units is also increased to a number of 80, and so is the situation with ROP’s which can be numbered at total of 32. Impressive isn’t it? Well, we have to say we are impressed. It takes a lot of guts to take a risk and double the complexity of a GPU, making it a lot more vulnerable to mistakes. Fortunately so far there are no mistakes, and new GPU can pull out maximum of 933GFlops. Next are memory and memory bus. There is 1GB of RAM which is supported by 512bit memory bus. There are 16 memory chips, arrayed in 8x64bit. This is something that ATI tried first with the R600 but we think that they were ahead of time. This solution has real application just now as we are getting prepared for usage of GDDR5 which will result in demands for memory bandwidth rocketing sky high. Other than the improvements regarding numbers, there are a lot of new features that are exclusive to the GTX280 architecture that can make it a great gaming tool. For example the texture scheduler is greatly improved, and so are the ROPs. One thing that NVIDIA is proud of is a new feature for NVIDIA cards - PhysiX support. Having this support, you will get every single benefit from realism in games as if you had a separate PhysiX card. We cannot really tell much on those benefits because there are still a small number of games that use full potential of these features. All these improvements make the new GTX280, not just a gaming tool. With CUDA, which nVIDIA considers the technology with great amount of potential, this GPU has even wider application. With CUDA technology we could easily forget our today opinions of CPU, and never think of it the same way we do it today, because CUDA is a language that uses parallel processing, making use of all of GPU power, and can reduce time in performing general-purpose mathematical calculations. CUDA is a software and GPU architecture that makes it possible to use as many processor cores (and eventually thousands of cores) in a GPU to perform general-purpose mathematical calculations. CUDA is accessible to all programmers through an extension to the C and C++ programming languages for parallel computing. This is possible because the architecture of the GTX280 delivers 16kb of cache for every group of 8 Shader Processors. Considering new features regarding 3D rendering quality, there is support for 128bit floating point, 16xAntiAliasing, GigaThread technology and Lumenex Engine which we first encountered with G80 GPU.

Video decoding capabilities of the GTX280 card aren’t just transferred from the previous generation, but they are remarkably improved. There is still Pure VideoHD and dual link DVI, but there is also Dual Stream Hardware Acceleration technology that supports picture-in-picture content for the Blu-ray videos.  There is also a Discrete Programmable Video Processor which is also called nVIDIA PureVideo and this is, as its name suggested, a discrete programmable processing core in nVIDIA GPUs that provides improved picture quality and smoothing playback of the movies with 100% offload of H.264 video decoding from the CPU with significant reduction in power consumption.