Photos and performance of the upcoming NVIDIA RTX 3060 Ti graphics card leaked from Videocardz (who else). NVIDIA’s RTX 3060 Ti graphics card will include 4864 CUDA cores and will be faster than the older RTX 2080 SUPER and will cost only $ 399 USD. The card is expected to be released in early December and should make a killer from the middle spectrum of NVIDIA.
NVIDIA RTX 3080 Ti Photos and performance expired, landing on December 2, 2020.
NVIDIA GeForce RTX 3060 Ti will include a GA104-200 graphics processor with 4864 CUDA cores and 8 GB of GDDR6 memory with a clock speed of 14 Gbps. The bus width will be 256 bits for a total bandwidth of 448 GB / s bandwidth. As the board will require less power, the TBP should be around 200W, with a reference design of around 180W. This is a 5W increase over the RTX 2070 at 175W and in line with how we expect Samsung’s node to stand in terms of energy efficiency compared to TSMC’s nodes. It is expected to have an MSRP of $ 399 and will start on December 2.
The Founder publishing version features the familiar metallic black aesthetics of the RTX 3080 and RTX 3090, but both fans are at the top of the card with a small cutout in the back panel, visible from the back. The overall shape and form of the founder’s edition is very compatible with the RTX 3000 series. The fan at the end of the card will expel air through the radiator and below – which is an interesting design philosophy compared to graphics cards from earlier this decade.
In terms of benchmarks, the RTX 3060 Ti is faster than the RTX 2080 SUPER and should trade in the RTX 2080 Ti. It is faster not only in shader performance, but also in rendering (calculation) and beam tracking capabilities and completely destroys the older replaced RTX 2060 SUPER. Of course, as with any benchmark comparison, always keep in mind that you must wait for results from third parties before making a decision.
Note that the reason why NVIDIA RTX 3090 and RTX 3080 cards do not scale linearly in performance is due to obstacles from game engines and binaries that are not designed to handle so many cores. We know that the hardware and driver stack scale linearly due to the benchmark performance of these cards in software such as vRAY and Octane – which are designed to handle a huge amount of graphics power. This means that as you shrink the cores, performance will no decreases linearly. In fact, I totally expect the RTX 3060 Ti to beat the RTX 2080 clock by clock! If we accept a little less watches than the RTX 2080 Ti, it should easily trade with the former flagship.
The yield for the initial few batches of Samsung’s 8-nanometer unit is lower than expected and expensive. The company expects yields (and therefore wafer prices) to improve significantly in the coming months. To protect its shareholders, the company was cautious about its orders for the first few lots and placed a significantly larger volume of orders for later lots. This problem was further exacerbated by the bots, which cleared the entire volume from the beginning. We expect group orders to start hitting by the end of the year (assuming once again that miners don’t eat it, of course).
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 You?||NVIDIA GeForce RTX 3080||NVIDIA GeForce RTX 3080 You?||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|
|Matrix size||TBA||TBA||TBA||395.2 mm2||395.2 mm2||628.4 mm2||628.4 mm2||628.4 mm2||628.4 mm2|
|Transistors||TBA||TBA||TBA||17.4 billion||17.4 billion||28 billion||28 billion||28 billion||28 billion|
|TMU / ROP||TBA||TBA||TBA||152/80||184/96||232/80||272/96||328/112||328/112|
|Tensor cores / RT||TBA||TBA||TBA||152/38||184/46||232/58||272/68||328/82||328/82|
|Basic 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 Calculation||TBA||TBA||TBA||16.2 TFLOP||20 TFLOP||TBA||30 TFLOP||TBA||36 TFLOP|
|RT TFLOP||TBA||TBA||TBA||32.4 TFLOP||40 TFLOP||TBA||58 TFLOP||TBA||69 TFLOP|
|Tensor-TOP||TBA||TBA||TBA||TBA||163 TOPS||TBA||238 CANNON||TBA||285 CANNON|
|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 bits||192 bits?||192 bits?||256 bits||256 bits||320 bits||320 bits||320 bits||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|
|Price (MSRP / FE)||$ 149?||$ 199?||$ 299?||$ 399 US?||$ 499 United States||$ 599 US?||$ 699 United States||$ 899 US?||$ 1499 United States|
|Launch (availability)||2021?||2021?||2021?||November 2020?||October 29||Q4 2020?||September 17||January 2021?||September 24|