Po onome sto ATI kaze, teselacija zapravo ne trosi ekstra bandwidtha, da citiram:
Citiraj:
•
Bandwidth is improved because, instead of transferring all of the vertex data for a high-polygon mesh
•
over the PCI-E bus, we only supply the coarse mesh to the GPU. At render time, the GPU will need to
fetch only this mesh data, yielding higher utilization of vertex cache and fetch performance. The
tessellator directly generates new data which is immediately consumed by the GPU, without additional
storage in memory.
|
A druga stvar o kojoj pricas, da ATI samo placa funkcionalne chipove, ne mora biti istina, jer se ista stvar prica i za NVdiju.
Treca stvar, nece nikad Nvidia u Globalfoundries dok je AMD u vlasnickoj strukturi iste. Barem po CEO-u Nvidije.
Prije ce ATI preci tamo. GF je pokazao na CES-u 28nm wafer tako da je moguce da ce biti ispred TMSC-a sto se tice prelaska na novi node.