According to an earlier post, NVIDIA will release 8 GB and 16 GB versions of their upcoming GeForce RTX 4060 Ti graphics card, with the 8 GB model going on sale later this month and the 16 GB model following in July. We are finding out what else distinguishes the two. By turning on 34 out of 36 SM physically present on the silicon, which equates to 4,352 out of 4,608 CUDA cores present, both are based on the 5 nm “AD106” silicon. The ASIC code for the 8 GB model is “AD106-350,” while the ASIC code for the 16 GB model is “AD106-351.”

Additionally, the RTX 4060 Ti 16 GB variant’s TDP is somewhat higher, rated at 165 W as opposed to 160 W for the 8 GB model. Instead of TGP (typical graphics power), which accounts for power used by the complete board, this is the TDP of the silicon. Since the 16 GB model has higher-density memory, it will undoubtedly have a higher TGP. In order to accomplish 16 GB, NVIDIA will probably use four 32 Gbit (4 GB) GDDR6 memory chips (instead of eight 16 Gbit ones with two chips piggybacked per 32-bit path).