The tag Ti for almost two decades is for NVIDIA synonymous with models much more powerful than the number that precedes it. In the case of the RTX 30, when they were launched in 2020 many of us wondered why there is an RTX 3090 and not an RTX 3080 Ti. Well, finally and almost a year after the launch of the GeForce Ampere, we finally have the RTX 3080 Ti accompanied by the RTX 3070 Ti. Within the range of NVIDIA products based on its RTX 30 architecture or known as GeForce Ampere, there were two gaps to fill due to the distance in price between the RTX 3070 and the RTX 3080 on the one hand and from the RTX 3080 to the RTX 3090 The new RTX 3070 Ti is placed in the first hole and the RTX 3080 Ti in the second. Thus
NVIDIA GeForce RTX 3080 Ti
Specifications
- GPU: GA102-225
- Family: RTX 30
- Architecture: GeForce Ampere
- SM/CUs: 80
- ALUs FP32 por SM/CU: 128
- Ray Tracing Unit: 1 per SM (80 total)
- Tensor Units: 4 per SM
- TFLOPS FP32: 34
- TFLOPS Tensor (FP16): 273
- VRAM speed: 19 Gbps
- Type VRAM: GDDR6X
- VRAM quantity: 12 GB
- VRAM bandwidth: 912 GB / s
- Base speed: 1365 MHz
- Boost speed: 1665 MHz
Details
The first and most powerful of the new NVIDIA graphics cards is the MSI GeForce RTX 3080 Ti, which is the third model based on the NVIDIA GA102 GPU after the RTX 3080 and RTX 3090. In this case it is based on the chip. GA102-225 Starting with the VRAM, the graphics card has a 384-bit GDDR6X bus like the RTX 3090, but with 12 GB of capacity, since NVIDIA has decided not to use the x8 mode of and therefore has installed half the memory chips than his older sister. Another change is the somewhat slower memory usage of 19 Gbps instead of 19.5 Gbps. So it uses the same type of GDDR6X as the GeForce 3080.
However, its consumption is 350 W, the same as that of the RTX 3090 and 30 W above the 320 W of the NVIDIA RTX 3080 graphics cards. If we go to the GPU itself, we are faced with a configuration of 80 SM at a base speed of 1365 MHz with a Boost to 1665 MHz, which gives it a power of 34 TFLOPS in FP32, 67 RT-FLOPS and 274 Tensioner TFLOPS. The first measure comes from the calculation capacity of the 128 ALUs in FP32 of each of the SMs, the second is a comparative measure created by NVIDIA for its graphics cards and the third is the calculation capacity of the Tensor Cores. As for its price, it has an of $1,199, which puts it almost double the RTX 3080’s of $699.
NVIDIA RTX 3070 Ti
Specifications
- GPU: GA104-400
- Family: RTX 30
- Architecture: GeForce Ampere
- SM/CUs: 48
- ALUs FP32 por SM/CU: 128
- Ray Tracing Unit: 1 per SM (48 total)
- Tensor Units: 4 per SM
- TFLOPS FP32: 22
- TFLOPS Tensor (FP16): 174
- VRAM speed: Gbps
- Type VRAM: GDDR6X
- VRAM quantity: 8 GB
- VRAM bandwidth: 609 GB / s
- Base speed: 1580 MHz
- Boost speed: 1770 MHz
Details
The second model that NVIDIA has presented at Computex 2021 is the RTX 3070 Ti, which immediately becomes the most powerful mid-range graphics card in the RTX 30. Since like the RTX 3060 Ti and RTX 3070 Ti this is based on a variant of the GA104, specifically the GA104-400.
Compared to the RTX 3070 this has the full 48 SM configuration, in addition to running at a base clock speed of 1580 MHz and a Boost of 1770 MHz, both of which are higher than the RTX 3070. surprise? NVIDIA has replaced the 256-bit GDDR6 memory interface with one of the same width GDDR6X, so the VRAM capacity is still 8 GB, but in exchange it has gone from having 14 Gbps memory to 19 Gbps, increasing its bandwidth from 448 GB / s to 609 GB / s. However, this increase in bandwidth comes with an increase in the energy consumption of the card, which goes from 220 W to 290 W in total.
Regarding the technical specifications, we find a power of 22 TFLOPS in FP32, 42 RT-TFLOPS and 174 Tensor TFLOPS. The meaning of the three performance parameters being the same that we have discussed in the RTX 3080 Ti and therefore the same as the rest of the NVIDIA RTX 30 range.
The NVIDIA recommended retail price for the RTX 3070 Ti? 599 dollars.
Based on GeForce Ampere architecture
Like the rest of the RTX 30 that have appeared on the market in recent months, both the RTX 3080 Ti and the RTX 3070 Ti are based on the GeForce Ampere architecture, which relies on three different pillars to achieve its enormous power:
- Enhanced CUDA cores, now with a second set of FP32 ALUs switched with the integer ALUs, allows you to perform up to twice as many operations per SM and clock cycle as GeForce Turing, aka RTX 20s.
- New RT Cores, the ability for Ray Tracing has been improved, to the point of doubling the performance in the face of the beam-box and beam-triangle intersection.
- Third-generation Tensor Cores, which allows acceleration of DLSS algorithms and the ability to use NVIDIA Broadcast at high speed and without compromising game performance.
These three elements make the RTX 30 graphics architecture today the most powerful on the market to date, making both the RTX 3080 Ti and RTX 3070 Ti graphics with unprecedented power in their respective price ranges.
To all this we must also add support for NVIDIA Reflex, which is designed to reduce gaming latencies as much as possible, allowing refresh rates of up to 360 Hz on G-SYNC monitors.
RTX 3070 Ti and RTX 3080 Ti join the 2nd of the RTX 30
NVIDIA has completely renewed its RTX 30 in a second generation that is designed especially for the gaming market. The way to achieve this has been by cutting Ethereum’s mining capacity in half, a change that has been carried out by NVIDIA also in the RTX 3080 Ti and the RTX 3070 Ti.
Another change that NVIDIA has implemented is the serial support for the standard Resizable BAR, this technology consists of giving access to the VRAM to the CPU. Its utility? It allows GPUs to read information from command lists much faster by not having to access the system RAM to obtain them. Which translates into a cut in frame time and therefore more FPS in games.
Also See:
Nvidia GeForce RTX 3080 Ti Review – An Elite Graphics Card for 4K Gaming
Author
-
by
-
