Nvidia Corp. has published official specifications of its forthcoming GeForce models GT 220 and G210 on its web-site. Currently the company only positions such graphics cards for original equipment manufacturers since their performance is very unlikely to be truly high or sufficient even for occasional gamers. Still, formally, models GT 220 and G210 are Nvidia’s first desktop 40nm chips that support DirectX 10.1.
The GeForce G210 (GT218 chip) OEM graphics solution has only 16 stream processing units, which means that its performance should be comparable to Nvidia GeForce 9400M core-logic with built-in graphics core. The GeForce GT220 (GT215 chip) OEM graphics processor sports 48 stream processors, therefore, its performance will be higher compared to GeForce 9500-series discrete graphics chips.
Specifications of Nvidia GeForce 200-Series OEM Graphics Chips | |||||
| G210 | GT 220 | |||
Stream Processors | 16 | 48 | |||
GPU Clock-Speed | 589MHz | 615MHz | |||
SP Clock-Speed | 1402MHz | 1335MHz | |||
Memory Clock-Speed | 1000MHz | 1580MHz | |||
Memory Type | 512MB DDR2 | 1GB GDDR3 | |||
Memory Bus Width | 64-bit | 128-bit | |||
Both new chips support DirectX 10.1 and OpenGL 3.0 application programming interfaces, hardware decoding and post-processing of high-definition videos as well as other functionality of contemporary graphics processing units. The GT218 and GT215 chips are made using 40nm process technology at Taiwan Semiconductor Manufacturing Company’s production facilities.
It is doubtful that going forward these products will find themselves in the channel after the currently available graphics cards become outdated due to various reasons. Firstly, performance benefits of the GeForce GT 220 and G210 compared to currently available graphics solutions are not going to be substantial. Moreover, actual performance of desktop GeForce G210 is lower compared to mobile GeForce G210M due to lower clock-speed. Secondly, TSMC will not be able to supply its 40nm chips in sufficient quantities for some time and it will hardly make sense for Nvidia to sell such chips on different markets, whereas OEM only positioning allows Nvidia to ship the processors only to select partners.



