NVIDIA’s GeForce RTX 3090 graphics card is the most famous and is also the company’s flagship GPU offering top-notch performance.
However, according to the latest information the green team wants to take things up a notch with its upcoming SUPER refresh series. The NVIDIA GeForce RTX 3090 SUPER is expected to feature the full-fat GA102 GPU core, faster clocks, and will come with a huge requirement for power.
coming to the GPU, NVIDIA’s GeForce RTX 3090 SUPER graphics card will have the GA102 GPU at its hard. The GPU is manufactured using Samsung’s 8nm custom process node which is designed specifically for NVIDIA as well as features a total of 28 Billion transistors.
This card will be the first-ever gaming card to feature the full GA102 GPU with a total of 84 SM units that would result in a total of 10752 CUDA cores. According to reports, the card will also feature a TGP of over 400W.
The GPU will also rock a 24 GB GDDR6X memory interface, however, NVIDIA will still offer boost memory clocks to 20 Gbps (19.5 Gbps on RTX 3090) for 960 GB/s bandwidth.
According to reports, it’s expected that the NVIDIA RTX 3090 SUPER will have a 5% performance increase than the previous ‘SUPER’ launches from NVIDIA. NVIDIA could likely cut MSRPs on existing cards and replace them with these newer SUPER refreshes at the same MSRP.
NVIDIA GeForce RTX 30 ‘SUPER’ Series Graphics Card Specifications (Rumored):
| Graphics Card Name | NVIDIA GeForce RTX 3060 | NVIDIA GeForce RTX 3060 Ti | NVIDIA GeForce RTX 3070 | NVIDIA GeForce RTX 3080 | NVIDIA GeForce RTX 3090 | NVIDIA GeForce RTX 3090 SUPER |
| GPU Name | Ampere GA106-300 | Ampere GA104-200 | Ampere GA104-300 | Ampere GA102-200 | Ampere GA102-300 | Ampere GA102-350? |
| Process Node | Samsung 8nm | Samsung 8nm | Samsung 8nm | Samsung 8nm | Samsung 8nm | Samsung 8nm |
| Die Size | TBC | 395.2mm2 | 395.2mm2 | 628.4mm2 | 628.4mm2 | 628.4mm2 |
| Transistors | TBC | 17.4 Billion | 17.4 Billion | 28 Billion | 28 Billion | 28 Billion |
| CUDA Cores | 3584 | 4864 | 5888 | 8704 | 10496 | 10752 |
| TMUs / ROPs | 112 / 64 | 152 / 80 | 184 / 96 | 272 / 96 | 328 / 112 | 336 / 112 |
| Tensor / RT Cores | 112 / 28 | 152 / 38 | 184 / 46 | 272 / 68 | 328 / 82 | 336 / 84 |
| Base Clock | 1320 MHz | 1410 MHz | 1500 MHz | 1440 MHz | 1400 MHz | TBA |
| Boost Clock | 1780 MHz | 1665 MHz | 1730 MHz | 1710 MHz | 1700 MHz | TBA |
| FP32 Compute | 13 TFLOPs | 16 TFLOPs | 20 TFLOPs | 30 TFLOPs | 36 TFLOPs | TBA |
| RT TFLOPs | 25 TFLOPs | 32 TFLOPs | 40 TFLOPs | 58 TFLOPs | 69 TFLOPs | TBA |
| Tensor-TOPs | 101 TOPs | 192 TOPs | 163 TOPs | 238 TOPs | 285 TOPs | TBA |
| Memory Capacity | 12 GB GDDR6 | 8 GB GDDR6 | 8 GB GDDR6 | 10 GB GDDR6X | 24 GB GDDR6X | 24 GB GDDR6X |
| Memory Bus | 192-bit | 256-bit | 256-bit | 320-bit | 384-bit | 384-bit |
| Memory Speed | 16 Gbps | 14 Gbps | 14 Gbps | 19 Gbps | 19.5 Gbps | TBA |
| Bandwidth | 384 Gbps | 448 Gbps | 448 Gbps | 760 Gbps | 936 Gbps | TBA |
| TGP | 170W | 175W | 220W | 320W | 350W | ~400W |
| Price (MSRP / FE) | $329 US | $399 US | $499 US | $699 US | $1499 US | $1499 US |
| Launch (Availability) | 25th February 2021 | 2nd December 2020 | 29th October 2020 | 17th September 2020 | 24th September 2020 | 2022? |
