Nvidia has confirmed that its Fermi-based GF100 GPU will feature 512 CUDA processors, 16 geometry units, 384-bit GDDR5 memory bus and 48 ROP engines.
According to ZDNet’s Adrian Kingsley-Hughes, the GF100 is also expected to include 4 raster units, 64 texture units and DirectX 11 support.
“Each GPU is made up of four separate GPC (Graphics Processing Clusters). Breaking this down further, we can look at each of the streaming multiprocessor (SM) cores within the GPCs separately,” wrote Kingsley-Hughes.
Nvidia details Fermi-based GF100 specs
“Each of the SMs comprise of 32 CUDA cores, 16 or 48KB of shared memory, 16 or 48KB of L1 cache, 4 texture units and a polymorph engine.”
However, Kingsley-Hughes emphasized that there were still “more questions than answers” about Nvidia’s next-generation Fermi architecture.
“Well, it’s all interesting technology, but there are still a lot of unanswered questions. First, there’s practical stuff such as price, power consumption, clock speeds, and thermal issues,” opined Kingsley-Hughes.
“Then there’s the all important question of how much benefit will this offer end users, especially gamers.”
He added that most games weren’t “pushing” graphics as hard as they once did, and that for most people a $100 graphics card was “more than enough.”