NVIDIA announced the new card on its website just hours after verifying the existence of the RTX 2060 12GB variant through the drivers.
As previously stated, the RTX 2060 12GB will have 2,176 CUDA cores. The base and boost clock speeds for this model are 1470 MHz and 1650 MHz respectively. The CUDA core clock rates are the same for the RTX 2060 12 GB and RTX 2060 SUPER. The new model will include 12 GB of GDDR6 memory connected to a 192-bit memory bus. This implies that the card’s maximum memory bandwidth will be 336 GB / s, a drop of 112 GB / s compared to the SUPER.
The card has a 12nm Turing architecture and a TU106 GPU with ray tracing and DLSS capabilities. However, several current amperage generation capabilities, such as the resizable bar, PCI Express 4.0, and next-gen RT and Tensor cores, will not be available for this card.
Due to the improved capacity of the memory subsystem, the board has an official TDP of 185W, which is 25W more than the 6GB vanilla variant and 10W more than the RTX 2060 SUPER.
NVIDIA has yet to release pricing for the new card, however, a brief statement has been sent to The Verge:
“This is a premium version of the RTX 2060 6GB and we expect the price to reflect that.”
– NVIDIA spokesperson
This implies that it will most likely cost more than the official MSRP of $ 299 for the 6GB model, which went down from $ 349. This price has nothing to do with the actual sale price, but there is good news. The card would be available in bespoke and Founders Edition designs. Since Founders Editions are now available, players can purchase these cards without paying a retail premium. The card will be available on December 7.
NVIDIA GeForce RTX x60 Series | |||||
techno | GeForce RTX 2060 SUPER | GeForce RTX 2060 12 GB | GeForce RTX 2060 6 GB | GeForce RTX 3060 Ti | GeForce RTX 3060 |
Architecture | NVIDIA Turing | NVIDIA Turing | NVIDIA Turing | NVIDIA Ampere | NVIDIA Ampere |
Plank | PG150 | PG161 | PG160 | PG190 / PG142 | PG190 |
GPU | 12 nm TU106-410 | 12 nm TU106 | 12 nm TU106-300 | 8 nm GA104-200 | 8 nm GA106-300 |
Boost the clock | 1650 MHz | 1650 MHz | 1680 MHz | 1665 MHz | 1777 MHz |
CUDA: RT: Tensor | 2176: 64: 136 | 2176: 64: 136 | 1920: 48: 120 | 4864: 38: 152 | 3584: 28: 112 |
Memory | 8 GB of GDDR6 | 12 GB of GDDR6 memory | 6 GB of GDDR6 | 8 GB of GDDR6 | 12 GB of GDDR6 memory |
Memory clock | 14 Gbit / s | 14 Gbit / s | 14 Gbit / s | 14 Gbit / s | 15 Gbit / s |
Memory bus | 256 bits | 192 bits | 192 bits | 256 bits | 192 bits |
Bandwidth | 448 GB / s | 336 GB / s | 336 GB / s | 448 GB / s | 360 GB / s |
TGP / TBP | 175W | 185W | 160W | 200W | 170W |
MSRP | 399 USD | To confirm | 349 -> 299 USD | 399 USD | 329 USD |
Release date | July 9, 2019 | December 7, 2021 | January 14, 2019 | December 2, 2020 | End of February 2021 |
Read more:
Qualcomm to launch new Snapdragon 8 Gen 1 now
Sapphire’s GOPRO X080 will be an AMD Navi 22 GPU carrying 10 GB of GDDR6 RAM clocked at 16 Gbps
The source
– Advertising –