|
Page 1 of 4 One of the most expected products in hardware industry these days is nVIDIA GTX 280. nVIDIA says that it’s a new generation product, which is not just a simple upgrade from G92 chip. It is still based on the same technology but it has some major upgrades that make it worthy of calling it the next gen graphics card. If you want to know if this is true or just simple marketing, continue reading the next few pages.
In past two years we were witnesses of great improvements in graphics card segment, when we got G70 and R520 card which evolved later in G71 and R580. Both ATI and nVIDIA after the release of the new gen products, released practically the same chips with some new features but most important is that those chips were made in a smaller production process, consumed less power and worked on higher clocks. That is the step that was repeated since then, so we got G92 after G80, and R670 after R600. Why is this important? You will see that later, in the conclusion, but now let’s talk about new technologies integrated in GTX 280.
GTX 280 is card that finds its roots in the G80 chip, and similar to it in a few ways. The most significant similarity of those chips, including G92, is that they work on same principle and that is - parallel computing, where every single Shader Processor can become a single core in a 240 core system, making use for CUDA Technology for example. There are lots of new features that this GPU introduces for the first time. It has 1,4 billion transistors that have the total size of a 576 square millimeters which is huge, and there we have the first similarity with the G80 GPU. When G80 came out, everybody was impressed with its huge size, and so is the situation with GTX280. The reason why this GPU is so big is because nVIDIA used old 65nm production process, which resulted in huge GPU size, and also made it impossible to integrate NVIO chip into GTX280. NVIO chip is yet again out of the GPU, and finds itself in front part of the board. When this fact came to public, a majority of media and users expressed negative critique because, by their opinion that wasn’t the smartest thing to do and by doing that nVIDIA makes a huge mistake regarding GTX280. In fact there is pretty reasonable explanation why nVIDIA did this. Production process in 65nm is the safest route for NV and they are using this process because they want to have as small yield as possible. That can make the new product “more power hungry” but if you look the targeted audience of this GPU, the quality power supply is something that is not standing in the way.
|