New specifications for NVIDIA's upcoming GeForce RTX 3080 Ti graphics card have emerged from Kopite7kimi. The graphics card has been in the rumors even before the GeForce RTX 30 series was introduced and has seen various spec changes but it looks like as we get close to an introduction, NVIDIA has decided to revise the core and memory configuration once again.
NVIDIA GeForce RTX 3080 Ti Now Rumored To Feature 12 GB GDDR6X Memory & 10240 CUDA Cores, Expected To Launch in April
In his latest tweet, Kopite7kimi reports that NVIDIA has once again modified the specifications of the GeForce RTX 3080 Ti graphics card. We know that GPU makers plan out several GPU configs prior to their release but in this generation, we have got to see just how much different variants a graphics card sees prior to finalization and production. This has to do with GPU yields, production & stock, & the latest specifications, while a small downgrade core-wise, will still net almost similar performance as the GeForce RTX 3090.
NVIDIA GeForce RTX 3080 Ti / RTX 3080 20 GB 'Rumored' Graphics Card Benchmarks
NVIDIA's GeForce RTX 3080 Ti / RTX 3080 20 GB FE (Founders Edition) graphics card is expected to feature the PG133-SKU15 PCB design and the GA102-250-KD-A1 graphics core. The GA102-250 GPU has also changed since the last time we saw them and will now feature 10240 CUDA cores within a total of 80 SM units. Since the main core configuration is entirely brand new, the previously leaked clocks & TMU/ROP counts should now be dismissed.
Things have changed.
Maybe the current spec is 10240FP32, 12G GD6X.— kopite7kimi (@kopite7kimi) February 6, 2021
As for memory, the card will feature 12 GB of GDDR6X memory. Unlike the 19.5 Gbps speeds of the RTX 3090, the GeForce RTX 3080 Ti is said to retain the same memory speeds as the RTX 3080 at 19 Gbps. Since we are getting 12 GB memory, the NVIDIA GeForce RTX 3080 Ti will be using a 384-bit bus interface which equals a total bandwidth of 912 GB/s. So while we see an 8 GB VRAM reduction over the previous spec, the larger bus interface should drive higher memory bandwidth. The TGP for the card is set to be the same as the RTX 3080 at 320 Watts. That's definitely needed to feed the extra cores so NVIDIA might have to optimize the clocks a bit here.
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 GA106-300 | 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? | 3584 | 4864 | 5888 | 7424? | 8704 | 10240? | 10496 |
| TMUs / ROPs | TBA | TBA | 112 / 64 | 152 / 80 | 184 / 96 | 232 / 80 | 272 / 96 | 320 / 112? | 328 / 112 |
| Tensor / RT Cores | TBA | TBA | 112 / 28 | 152 / 38 | 184 / 46 | 232 / 58 | 272 / 68 | 320 / 76? | 328 / 82 |
| Base Clock | TBA | TBA | 1320 MHz | 1410 MHz | 1500 MHz | TBA | 1440 MHz | TBA | 1400 MHz |
| Boost Clock | TBA | TBA | 1780 MHz | 1665 MHz | 1730 MHz | TBA | 1710 MHz | TBA | 1700 MHz |
| FP32 Compute | TBA | TBA | 12.7 TFLOPs | 16.2 TFLOPs | 20 TFLOPs | TBA | 30 TFLOPs | TBA | 36 TFLOPs |
| RT TFLOPs | TBA | TBA | 25.4 TFLOPs | 32.4 TFLOPs | 40 TFLOPs | TBA | 58 TFLOPs | TBA | 69 TFLOPs |
| Tensor-TOPs | TBA | TBA | 101 TOPs | 129.6 TOPs | 163 TOPs | TBA | 238 TOPs | TBA | 285 TOPs |
| Memory Capacity | 4 GB GDDR6? | 6 GB GDDR6? | 12 GB GDDR6 | 8 GB GDDR6 | 8 GB GDDR6 | 10 GB GDDR6X? | 10 GB GDDR6X | 12 GB GDDR6X? | 24 GB GDDR6X |
| Memory Bus | 128-bit | 192-bit? | 192-bit | 256-bit | 256-bit | 320-bit | 320-bit | 384-bit | 384-bit |
| Memory Speed | TBA | TBA | 15 Gbps | 14 Gbps | 14 Gbps | TBA | 19 Gbps | 19 Gbps | 19.5 Gbps |
| Bandwidth | TBA | TBA | 360 Gbps | 448 Gbps | 448 Gbps | TBA | 760 Gbps | 912 Gbps | 936 Gbps |
| TGP | 90W? | TBA | 170W | 200W | 220W | 320W? | 320W | 320W | 350W |
| Price (MSRP / FE) | $149? | $199? | $329 | $399 US | $499 US | $599 US? | $699 US | $999 US? | $1499 US |
| Launch (Availability) | 2021? | 2021? | February 2021 | December 2020 | 29th October | Q4 2020? | 17th September | April 2021? | 24th September |
As for the launch date, Videocardz has suggested that the NVIDIA GeForce RTX 3080 Ti is likely to launch sometime in April though the date isn't 100% guaranteed. The reason stated is that the same roadmap that lists down April also has other dates mentioned. But with that said, the RTX 3080 Ti still aims to compete against the Radeon RX 6900 XT at a lower price point. It is likely the changes to the memory design could've been made to offer the card at a more attractive price point which will still end up being closer to $999 US, the same as AMD's flagship graphics card.
The post NVIDIA GeForce RTX 3080 Ti Now Rumored To Launch in April With 12 GB GDDR6X Memory & 10240 CUDA Cores by Hassan Mujtaba appeared first on Wccftech.
source https://wccftech.com/nvidia-geforce-rtx-3080-ti-12-gb-gddr6x-10240-cuda-cores-april-launch-rumor/