GeForce GTX 550 Ti

The new GeForce GTX 550 Ti model, based on the improved GF116 graphics processor, will not replace the GeForce GTS 450 video card on the market. At the beginning of their journey, these two models will coexist in the company's lineup together. Updated GPU It is manufactured using the same 40 nm process technology as other video chips of this generation, and its performance improvements were made possible by debugging the 40 nm process technology and deep processing. GPU.

The naming principle of the video card has changed a little again. The updated model now not only belongs to the new fifth generation, and compared to the GTS 450, not only the Ti suffix has been added. For some reason, NVIDIA decided to transfer the released solution based on the GF116 chip to the higher GTX class, compared to the previous similar card, which belonged to the GTS class.

Like the previous GeForce GTX 500 series models, the announced GTX 550 Ti uses an improved graphics processor, already familiar to us from the previous series, and only slightly modified in order to increase performance and improve energy efficiency.

The modifications to GF116 are aimed at reducing leakage and increasing yield at higher frequencies compared to GF106. As a result, the GTX 550 Ti has better performance while running at almost 120 MHz higher than the GTS 450 GPU, while consuming only 10% more power. The GeForce GTX 550 Ti uses the same multiprocessor configuration as the GTS 450. GF116 contains one GPC cluster, four SMs, and three 64-bit memory controllers interlocked with LXNUMX cache and eight ROPs each. In total, the new GPU includes 192 stream processors, assembled into four multiprocessors with 48 pieces each.

But the GF116 memory subsystem has undergone some changes. Each multiprocessor has the same 64 kilobytes of on-chip memory, and GPU In total, it contains 384 KB of unified L550 cache, which serves all requests for loading and saving data. But in the GTX 450 Ti, three memory controllers are already active, not two, as was the case in the GTS 192, which in total makes up a XNUMX-bit bus.

But even this is not all. It's not just the frequencies and bus width that have changed with the GF116. As you remember from the technical specifications of the solution, with a 192-bit memory bus, the GTX 550 Ti somehow contains exactly 1 GB of video memory, although usually in such cases it should be 768 or 1536 MB. If you use the same GDDR5 chips, of course.

So here is the memory controller of the new one GPU supports the use of mixed-density memory chips. Which allows you to dial exactly 1024 megabytes using a 192-bit interface. Four chips of the same density (32M×32) with a total volume of 512 megabytes are used, and two more chips of twice the density (64M×32), which together also make up 512 MB. And in total, you get exactly 1 GB.

The biggest change to the GTX 550 Ti, from a performance perspective, is the significant increase in clock speeds. GPU and memory, as well as expanding its bus. Unlike the reference frequency of the GTS 450 chip, equal to 783/1566 MHz, in the GTX 550 Ti it was increased to 900/1800 MHz, which is 15% higher. With video memory it is even better, its frequency was increased from 3608 MHz to 4104 MHz, that is, by 14%, and the bus was expanded by one and a half times. As a result, the overall increase in the throughput of the GTX 550 Ti compared to the GTS 450 exceeds 70%.

Such an increase in theoretical characteristics causes an increase in practical rendering performance in games, especially in cases where the speed is limited by video memory bandwidth. Judging by NVIDIA's own tests, with typical settings for the GTX 550 Ti (1680x1050 and MSAA 4x), just expanding the bus to 192-bit gives about 14% on average, and in the case of some games even higher (StarCraft II - 31% , Aliens vs Predator - 23%, Batman: Arkham Asylum - 21%).

The GF116 is a slightly improved and refined GF106, just as the GTX 550 Ti is a slightly improved GTS 450. The updated GPU from the upper part of the lower price range compares favorably with its predecessor due to significantly higher clock speeds, an increased number of ROP units and noticeably higher video memory bandwidth with only slightly increased power consumption.