NVIDIA seems to have taken note of the recent AMD RX 6000 leaks and has allegedly started work on two brand new graphics cards, the GeForce RTX 3080 Ti and the GeForce RTX 3070 Ti. The new graphics cards will be positioned against the Radeon RX 6800 series graphics cards and will feature much powerful specifications than the standard non-Ti variants that are already announced and in the market.
NVIDIA GeForce RTX 3080 Ti & GeForce RTX 3070 Ti Allegedly Being Prepped To Tackle AMD Radeon RX 6800 Graphics Cards, Beefed-Up Specifications
The latest details come from none other than Kopite7kimi who has been spot on with the majority of his leaks in the past and has now tweeted out two brand new GA102 GPU based graphics cards that NVIDIA is working on to tackle AMD’s Radeon RX 6800 series cards which most likely pose an imminent threat to NVIDIA’s recently released GeForce RTX 3080 and the upcoming GeForce RTX 3070 based on the recent performance leaks.
The NVIDIA GeForce RTX 3080 Ti and GeForce RTX 3070 Ti are the expected names for the new line of graphics cards and its also isn’t the first time we have heard about these names. The RTX 3080 Ti and the RTX 3070 Ti have long been rumored and was actually in the plans at one point before the green team decided to move things up to the RTX 3090. The RTX 3080 Ti and RTX 3070 Ti were also going to feature much different specifications before but it looks like things have changed and both graphics cards will feature the GA102 GPU. Following are the specifications that we can expect from the new graphics cards (if they make it to market and not get cancelled like their previous entrants):
NVIDIA GeForce RTX 3080 Ti (GA102-250-A1) Graphics Card To Tackle AMD’s RX 6800 XT?
First up, we have the NVIDIA GeForce RTX 3080 Ti which has been upgraded to the GA102-250-A1 GPU core. This graphics card should feature performance in between the GeForce RTX 3080 & RTX 3090, both of which are also GA102 based GPUs. In terms of specifications, the GeForce RTX 3080 Ti graphics card with its GA102 GPU will feature 9984 CUDA cores which is super close to the RTX 3090’s 10496 CUDA Cores.
That’s a 15% core count increase over the RTX 3080 and just 5% less cores than the RTX 3090. We have already seen that the GeForce RTX 3090 is only able to show greater than 10-15% gains in a handful of titles which are able to utilize its insane core architecture. If that’s the case with the RTX 3080 Ti, then it should end up around 5-10% faster than the RTX 3080 which isn’t a huge gain over the existing graphics card but should be enough to outperform the Radeon RX 6800 XT or Radeon RX 6900 XT.
The graphics card is also said to feature a 384-bit bus interface and will come with GDDR6X VRAM. Since there’s a 384-bit bus, we are most likely looking at either 12 GB of 24 GB of capacity. The 24 GB VRAM should be slightly unlikely since NVIDIA would want to tune this card at more enthusiasts than the RTX 3090. One of the reasons that the RTX 3090 costs $1499 US is due to its VRAM since it makes use of 12 additional GDDR6X memory chips on the backside to make up the 24 GB VRAM. A 12 GB RTX 3080 Ti should cut down the costs a lot and should be available in better quantities than the RTX 3090 at around $899-$999 US.
NVIDIA GeForce RTX 3070 Ti (GA102-150-A1) Graphics Card To Tackle AMD’s RX 6800 XL?
The NVIDIA GeForce RTX 3070 Ti on the other hand is expected to feature the GA102-150-A1 GPU. The graphics card is also going to feature the GA102 GPU which will be a huge upgrade from GA104 that the standard offering would feature. The GA102 GPU should also result in higher power input which should directly affect its TGP.
According to Kopite7kimi, the NVIDIA GeForce RTX 3070 Ti will utilize the PG132-SKU35 board which is different than the previous SKU that the graphics card was going to be based upon. The GPU will feature 7424 CUDA cores or 58 SMs. These are 26% more CUDA cores than the GeForce RTX 3070 and around 9% lower cores than the GeForce RTX 3080.
Previously, the GeForce RTX 3070 Ti was going to feature the full GA104 GPU with 6144 CUDA cores. The NVIDIA GeForce RTX 3070 Ti was also supposed to feature a 256-bit bus with a 16 GB GDDR6 VRAM but that has changed too. As per the new specs, the card would most likely come with a 320-bit bus interface and as such, we can expect a 10 GB GDDR6X VRAM.
NVIDIA GeForce RTX 30 Series ‘Ampere’ Graphics Card Specifications:
| Graphics Card Name | NVIDIA GeForce RTX 3050 | NVIDIA GeForce RTX 3050 Ti | NVIDIA GeForce RTX 3060 | NVIDIA GeForce RTX 3060 Ti | NVIDIA GeForce RTX 3070 | NVIDIA GeForce RTX 3070 Ti? | NVIDIA GeForce RTX 3080 | NVIDIA GeForce RTX 3080 Ti? | NVIDIA GeForce RTX 3090 |
|---|---|---|---|---|---|---|---|---|---|
| GPU Name | Ampere GA107 | Ampere GA106? | Ampere GA106? | Ampere GA104-200 | Ampere GA104-300 | Ampere GA102-150 | Ampere GA102-200 | Ampere GA102-250 | 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 | 395.2mm2 | 395.2mm2 | 628.4mm2 | 628.4mm2 | 628.4mm2 | 628.4mm2 |
| Transistors | TBA | TBA | TBA | 17.4 Billion | 17.4 Billion | 28 Billion | 28 Billion | 28 Billion | 28 Billion |
| CUDA Cores | 2304 | 3584 | 3840 | 4864 | 5888 | 7424 | 8704 | 10496 | 10496 |
| TMUs / ROPs | 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 Clock | TBA | TBA | TBA | 1665 MHz | 1730 MHz | TBA | 1710 MHz | TBA | 1700 MHz |
| FP32 Compute | 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 TOPs | TBA | 238 TOPs | TBA | 285 TOPs |
| Memory Capacity | 4 GB GDDR6? | 6 GB GDDR6? | 6 GB GDDR6? | 8 GB GDDR6 | 8 GB GDDR6 | 10 GB GDDR6X? | 10 GB GDDR6X | 20 GB GDDR6X | 24 GB 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 | 90W? | TBA | TBA | 180W? | 220W | 320W? | 320W | 320W | 350W |
| Price (MSRP / FE) | $149? | $199? | $299? | $399 US? | $499 US | $599 US? | $699 US | $899 US? | $1499 US |
| Launch (Availability) | 2021? | 2021? | 2021? | November 2020? | 29th October | Q4 2020? | 17th September | January 2021? | 24th September |
The GPU is said to be rated at 320W and would be just slightly slower than the GeForce RTX 3080 which should be enough to tackle the Radeon RX 6800 (Navi 21 XL) graphics card while the GeForce RTX 3080 (or Ti) would compete against the Radeon RX 6800 XT (Navi 21 XT). In terms of pricing, we can expect the card at around $599 US which should slot in perfectly between the RTX 3070 and the RTX 3080.
Once again, these cards and the GPUs they are derived from are mostly in testing phase right now at NVIDIA. It all depends on the AMD Radeon RX 6000 series graphics cards, how well they perform and how well they are priced against the competition, for NVIDIA to give these a proper greenlight to be produced for the mass market. If the RDNA 2 lineup fails to meet NVIDIA’s internal estimates, then there’s a chance that these cards would also get cancelled in favor of something new for later.
