Graphics Cards Suppliers May Dislike the Upgrade of Nvidia’s New Chip

From X-bit Labs: Nvidia Corp. does need to offer something more competitive compared to its GeForce GTX 260 product, but introduction of a new model may actually get its partners among suppliers of graphics cards upset as they think that the improved version of the GTX 260 may cause overstock problems.

Presently Nvidia and its add-in-board partners sell two versions of the GeForce GTX 200: the premium-class GTX 280 with 240 shader processors (SPs), 80 texture units (TUs), 32 render back ends (RBEs) and 1GB of GDDR3 memory connected using 512-bit interface as well as high-end GTX 260 with 192 SPs, 64 TUs, 28 RBEs and 896MB of GDDR3 memory connected using 448-bit interface. The GeForce GTX 280 has GPU clocked at 602MHz, SPs operating at 1296MHz and memory working at 2214MHz, whereas the GeForce GTX 260 is clocked at 576MHz, 1242MHz and 1998MHz, respectively.

But in September the company plans to add another board into the lineup, which will have the same clock-speeds with the GTX 260 model, but with 216 SPs, 72 TUs, 30 RBEs and 896MB of GDDR3 memory connected using 448-bit interface. It is projected that the novelty will retain the GeForce GTX 260 model number, which will cause a lot of misunderstanding on the market as different 260s will have completely different performance despite of similar price around $299.

It is unclear why Nvidia wants to use the old model number for a new graphics card as this not only causes a lot of misunderstanding on the market, but also puts pressure on graphics cards suppliers who have the “old” GeForce GTX 260 products in stock. Naming the new product as the GeForce GTX 270 would simplify the transition considerably.

View: Article @ Source Site