ASUSA has confirmed the two upcoming NVIDIA G. Force RTX 30 series graphics cards, GeForce RTX 3080T and JForce RTX 3060. Both graphics cards are set to launch in early ’21, but ASUS has already listed them under the Support and Services page.
ASUS GeForce RTX 3080T20GB and GeForce RTX 3060 12GB ROG Strix Custom Graphics Cards Confirmed
Both were spotted by the graphics card HXL (9550 Pro) Within the ASUS Support and Services page for graphics cards. The list includes the GeForce RTX 30 Series ROG Strikes variants and also mentions the unpublished GeForce RTX 3080T and GeForce RTX 3060 graphics cards that we talked about in the previous leak.
The leak confirms that ASUS is working on both overclocked and non-overclocked types for their ROG Strikes series. The GeForce RTX 3080T will have 20GBDDR6X memory and the GeForce RTX3060 will be equipped with 12GB of GBDDR6 memory. Since both of these cards are custom types, they will use the brand new ROG Streaks cooling solution seen on the existing GForce RTX 30 ROG Streaks lineup C, which expects cooling performance top and beef custom PCB design.
NVIDIAA GeForce RTX 3080T ‘Rumor’ Graphics Card Specifications
Kopit k km details indicate that the Jeff Orce RTX 3080 TEF (Founder Edition) graphics card will feature PG133-SQ1515 PCB design and GA102-250-KD-A1 graphics core. The GA102-250 GPU has also changed since the last time we saw them and now it is similar to the GeForce RTX 3090 on 10496 FP32CUDA cores.
Speaking of memory, this card will have 20 GBDDR6X memory. In contrast to the RTX 3090’s 19.5 Gbps speed, the GeForce RTX 3080 T is said to maintain memory speed up to RTX 3080 at 19 Gbps. Since we are getting a 20GB bus interface, NVIDIA will use a 320-bit bus interface which is equivalent to a total bandwidth of 760GB / s. The TGP for the card is set at 320 watts similar to the RTX 3080. It is definitely necessary to feed the extra cores so NVIDI has to optimize the watches a bit here.
NVIDIA will be able to cut the AMD Raden RX 6900 XT which is a ટી 999 US graphics card with the RTX 3080 TI, although it remains to be seen how much of a difference the TI variant will be, as it will be a sandwich between the RTX 3080 and RTX 3090. There are reports that the RTX3080TI could end faster than the RTX3090 due to its clockwise clock speed while the RTX3090 could be a product of the Titan class serving the vibrant and content creation market with its high VRAM buffer.
NVIDIA GeForce RTX 3060 ‘rumor’ graphics card specifications
The GForce RTX 3060 is suggested to feature GA 106-400 GPU with a total of 30 SM units expected to feature 3840 FP 32 cores. The graphics card PG190 SKU will be used by board partners that are both GA104 and GA106 pin-compatible. The graphics card has seen a range of memory specs updates starting at 6 GB (14 Gbps) and is now listed at 12 GB (16 Gbps).
It is rumored that the card will feature a huge 12GB memory buffer that is even larger than the RTX 3060T, RTX 3070, and RTX 3080. This has not been confirmed yet but the reason NVIDIA really wants to face. AMD in the mainstream segment. Again, we also saw rumors of 20GB and 16GB for the RTX 3080 and RTX 3070, which may not have been true due to the early designs being later canceled.
The card will have a 192-bit bus interface and 16 Gbps, with bandwidths of up to 384 GB / s. The GeForce RTX 3060 could be priced at US 299 which would make it a formidable graphics card.
NVIDIA GeForce RTX 30 Series ‘Ampere’ Graphics Card Specifications:
| Graphics card name | NVIDIAA GAFORCES RTX 3050 | NVIDIAA GForce RTX 3050T | NVIDIAIA GForce RTX 3060 | NVIDIAIA GAFORCES RTX 3060T | NVIDIAA GAFORCES RTX 3070 | NVIDIAAA GFORCES RTX 3070T? | NVIDIAA GAFORCES RTX 3080 | NVIDIA GForce RTX 3080T? | NVIDIAA GFORCES RTX 3090 |
|---|---|---|---|---|---|---|---|---|---|
| GPU name | Ampere GA 107 | Ampere GA 106? | Ampere GA 106? | Ampere GA 104-200 | Ampere GA 104-300 | Ampere GA 102-150 | Ampere GA 102-200 | Ampere GA 102-250 | Ampere GA102-300 |
| Process node | Samsung 8nm | Samsung 8nm | Samsung 8nm | Samsung 8nm | Samsung 8nm | Samsung 8nm | Samsung 8nm | Samsung 8nm | Samsung 8nm |
| Die size | TBA | TBA | TBA | 395.2 mm | 395.2 mm | 628.4 mm | 628.4 mm | 628.4 mm | 628.4 mm |
| Transistor | TBA | TBA | TBA | 17.4 billion | 17.4 billion | 28 billion | 28 billion | 28 billion | 28 billion |
| CUDA cores | Is 2304 | 3584 | 3840 | 4864 | 5888 | 7424 | 8704 is | Is 10496 | Is 10496 |
| TMU / ROP | TBA | TBA | TBA | 152/80 | 184/96 | 232/80 | 272/96 | 328/112 | 328/112 |
| Tensor / RT cores | TBA | TBA | TBA | 152/38 | 184/46 | 232/58 | 272/68 | 328/82 | 328/82 |
| Base clock | TBA | TBA | TBA | 1410 MHz | 1500 MHz | TBA | 1440 MHz | TBA | 1400 MHz |
| Boost the clock | TBA | TBA | TBA | 1665 MHz | 1730 MHz | TBA | 1710 MHz | TBA | 1700 MHz |
| FP32 count | TBA | TBA | TBA | 16.2 TFLOPS | 20 TFLOPS | TBA | 30 TFLOPS | TBA | 36 TFLOPS |
| RT TFLOPS | TBA | TBA | TBA | 32.4 TFLOPS | 40 TFLOPS | TBA | 58 TFLOPS | TBA | 69 TFLOPS |
| Tensor-tops | TBA | TBA | TBA | TBA | 163 top | TBA | 238 Tops | TBA | 285 tops |
| Memory capacity | 4GB GDDR6? | 6GB GDDR6? | 6GB GDDR6? | 8GB GDDR6 | 8GB GDDR6 | 10GB GDDR6X? | 10GB GDDR6X | 20GB GDDR6X | 24GB GDDR6X |
| Memory bus | 128-bit | 192-bit? | 192-bit? | 256-bit | 256-bit | 320-bit | 320-bit | 320-bit | 384-bit |
| Memory speed | TBA | TBA | TBA | 14 Gbps | 14 Gbps | TBA | 19 Gbps | 19 Gbps | 19.5 Gbps |
| Bandwidth | TBA | TBA | TBA | 448 Gbps | 448 Gbps | TBA | 760 Gbps | 760 Gbps | 936 Gbps |
| TGP | 90 W? | TBA | TBA | 180 W? | 220W | 320W? | 320 W. | 320 W. | 350 W. |
| Price (MSRP / FE) | 9 149? | $ 199? | 9 299? | 9 399 US? | 9 499 U.S. | 9 599 US? | 9 699 U.S. | 99 899 US? | 99 1499 U.S. |
| Launch (availability) | 2021? | 2021? | 2021? | November 2020? | 29 .October | Q42020? | September 17 | January 2021? | September 24 |
