We will start by talking about its chips, its die, the GPU itself, where the younger of the two sisters integrates a GA106-300-A1. This would not be more relevant without the fact that the RTX 3060 Ti incorporates a GA104-200-A1, a superior chip which although cut for the occasion is the same as that used by the RTX 3070 and RTX 3070 Ti.
Therefore, initially there are two different performances and with a slight gap between them. Likewise, both belong to the Ampere range and are made in 8nm from Samsung, but the number of transistors and the area are completely different from what is mentioned above. The minor reached 13,250 million transistors and an area of 276 mm2, the oldest has 17.4 billion transistors Yes 392 mm2, therefore proportionally and thermally they should be equal for the same radiator, but it will be necessary to check the consumption to discern it, as well as the frequencies.
Likewise, both graphics cards support PCIe 4.0 x16 and lower, so they have the same available bandwidth. It seems obvious but it is not, because in some models they are cut in x8, especially low and mid-range, although this is not the case here.
GA104 vs. GA106
What does each of the tokens consist of? Well there are quite a few differences between them, starting with the RTX 3060 and its 3584 shades which are distributed in 28 SM At the rate of 128 Shaders for each of them (in fact there are 64, but Nvidia doubles the quantity because it considers double FP32s). Her sister gets 4864 shades Distributed in 38 SM, so for simple CUDA units there is already a clear winner, although not everything is said far from it.
They accompany the first 112 TMU, 48 ROP, 112 Tensor Core and 28 RT Cores, the second sees the numbers increase to 152 TMU, 80 ROP, 152 tensor cores and 38 RT cores. What they logically share are the 128 KB of L1 for each SM that they integrate, so the more there are, the more the top-level cache is important.
This also happens in its L2 cache, where the RTX 3060 gets 3 MB while the RTX 3060 Ti goes up to 4 MB.
Same type of memory, different buses and speeds
Looking at the chips in their entirety, we need a clear picture of what VRAM is going to offer in order to accurately render a verdict on FP32, as the type, speed, bus, and frequencies of the chips almost determine. all.
The muscle is accompanied by frequencies which for GA106 are 1.320MHz in the base and 1.777MHz in Boost, while the older of the two gets 1.410 MHz in the base and 1.665 MHz In Boost, the numbers in both cases will increase depending on the quality of the ASIC and the heat sink dissipation in service, while in the case of Founders Edition they will surely be very uniform in said increase.
What they share is undoubtedly their VRAM, although there are some differences as well. Both graphics cards contain GDDR6 to their credit, but the smaller one curiously receives more VRAM (12 GB) while the larger and more powerful con forms to 8 GB
Because? Well, for a simple reason the RTX 3060 Ti has 256-bit bandwidth, which gives it 448 GB / s for the 360 GB / s RTX 3060 due to 192 bits which integrates.
Power, consumption and dimensions
Starting from the base that being the same series, they support the same technologies like DX12 Ultimate, OpenGL 4.6, OpenCL 3.0, Vulkan 1.2, CUDA 8.6 and Shader Model 6.6, after seeing all the above we have some interesting consumption differences.
For the 242 x 112 mm measured by the two cards with their radiators, the RTX 3060 consumes 170 watts for the 200 of her older sister, which does not change that they both need a 12 pins from NVIDIA (Founders Edition) or 8 pin if we have a custom model. Either way, they both have one HDMI 2.1 and three DisplayPort 2.0b.
Beware of waterblocks because the PCBs are marked differently by NVIDIA: PG190 SKU 60 for the minor and PG142 SKU 20 for the RTX 3060 Ti. That said, the result if we take into account the previous data and still speaking of raw power which is not effective, gives us as the result FP32 in floating point of 12.74 TFLOPS for RTX 3060 and 16.20 TFLOPS for the RTX 3060 Ti.
Having seen all this, we can only compare the prices. NVIDIA RTX 3060 is priced at MSRP of $ 329, while the RTX 3060 Ti goes up to $ 399.
Conclusion: RTX 3060 vs RTX 3060 Ti
Is it worth paying more? Well, in this case yes, because despite 4 GB less VRAM, performance climbs between a 22% and 26% roughly in favor of the Ti, so the price GAP is more than justified as long as you’re not playing at 4K resolutions in Ultra, where the RTX 3060’s 12GB would cut the distance at one point, but not too much for its lower overall bandwidth.
Table of Contents