NVIDIA GeForce RTX 3060 3DMark & Superposition Synthetic Benchmarks Leak Out, On Par With RTX 2070


More benchmarks of the NVIDIA GeForce RTX 3060 graphics card have leaked out just a few hours before its launch by Videocardz. The GeForce RTX 3060 will be the latest and most entry-level card within the RTX 30 series so far with an MSRP of $329 US.

NVIDIA GeForce RTX 3060 3DMark & Superposition Benchmarks Leak Out, Performance On Par With An RTX 2070

The NVIDIA GeForce RTX 3060 AOTS benchmarks leaked out a while back and showed it on par with an RTX 2070 SUPER however that should be the best-case scenario for Ampere. When comparing it in synthetic benchmarks, the results are very similar to the RTX 2070 or the RTX 2060 SUPER while the GeForce RTX 3060 TI maintains a 40% lead on average which is quite a huge gap for the NVIDIA 3060 series lineup.

NVIDIA Launches GeForce GTX 1630: An Entry-Level Graphics Card For The Sub-$150 US GPU Segment

So talking about performance numbers, the NVIDIA GeForce RTX 3060 scores 10284 points in the 3DMark Fire Strike Extreme, 5012 points in the 3DMark Fire Strike Ultra, 8757 points in the 3DMark Time Spy, 4111 points in the 3DMark Time Spy Extreme and 5072 points in the Port Royal benches. As for Superposition, the GPU scores 5073 points in the 1080p Extreme preset and 2832 points in the 8K optimized preset.

Now given these performance metrics, the NVIDIA GeForce RTX 3060 comes out as a decent card and a replacement for the RTX 2060 SUPER which retailed at $399 US. The graphics card offers a higher VRAM capacity of 12 GB versus 8 GB and has a much lower power draw thanks to the GA106 GPU. Then again, the $329 US MSRP is a far cry from where the card will actually be priced and we are already looking at $450 US+ retail prices for various custom models.

Now mining can be used as a reason to justify the higher prices but it has more so to do with the price gouging tactics of retailers and distributors who have been taking advantage of the pandemic situation. Do note that NVIDIA has already nerfed the mining performance of the RTX 3060 & has issued a statement that the mining potential of the card is pretty hard to bypass as detailed here.

NVIDIA GeForce RTX 3060 Graphics Card Specifications - GA106 GPU & 12 GB GDDR6 Memory

At the heart of the NVIDIA GeForce RTX 3060 graphics card lies the GA106 GPU. The GA106 is one of the many Ampere GPUs that we will be getting on the gaming segment. The GA106 GPU is the third gaming GPU that NVIDIA has produced.

Colorful’s GeForce GTX 1630 Graphics Card Leaks Out, Entry-Level Design With Dual-Fan Cooler

The new shader core on the NVIDIA Ampere architecture is 2.7x faster, the new RT cores are 1.7x faster while the new Tensor cores are up to 2.7x faster than the previous generation Turing GPUs. The 2nd Generation RT core delivers dedicated hardware-accelerated ray-tracing performance & features twice the ray/triangles intersection with concurrent RT graphics and compute operations.

For the GeForce RTX 3060, NVIDIA has enabled a total of 28 SM units on its flagship which results in a total of 3584 CUDA cores, 112 TMUs. In addition to the CUDA cores, NVIDIA's GeForce RTX 3060 also comes packed with next-generation RT (Ray-Tracing) cores, Tensor cores, and brand new SM or streaming multi-processor units. The graphics card has a TDP of 170W.

In terms of memory, the GeForce RTX 3060 comes packed with 12 GB of GDDR6 memory. The memory runs across a 192-bit bus wide interface and features an effective clock speed of 15.00 Gbps which delivers a cumulative bandwidth of 360 GB/s which is slightly higher bandwidth than the RTX 2060 6 GB model & you also end up with twice the memory while Ampere's brand new compression architecture makes sure that the GPU isn't bandwidth starved.

NVIDIA GeForce RTX 30 Series 'Ampere' Graphics Card Specifications:

Graphics Card NameNVIDIA GeForce RTX 3060NVIDIA GeForce RTX 3060 TiNVIDIA GeForce RTX 3070NVIDIA GeForce RTX 3080NVIDIA GeForce RTX 3090
GPU NameAmpere GA106-300Ampere GA104-200Ampere GA104-300Ampere GA102-200Ampere GA102-300
Process NodeSamsung 8nmSamsung 8nmSamsung 8nmSamsung 8nmSamsung 8nm
Die SizeTBC395.2mm2395.2mm2628.4mm2628.4mm2
TransistorsTBC17.4 Billion17.4 Billion28 Billion28 Billion
CUDA Cores358448645888870410496
TMUs / ROPs112 / 64152 / 80184 / 96272 / 96328 / 112
Tensor / RT Cores112 / 28152 / 38184 / 46272 / 68328 / 82
Base Clock1320 MHz1410 MHz1500 MHz1440 MHz1400 MHz
Boost Clock1780 MHz1665 MHz1730 MHz1710 MHz1700 MHz
FP32 Compute13 TFLOPs16 TFLOPs20 TFLOPs30 TFLOPs36 TFLOPs
Tensor-TOPs101 TOPs192 TOPs163 TOPs238 TOPs285 TOPs
Memory Capacity12 GB GDDR68 GB GDDR68 GB GDDR610 GB GDDR6X24 GB GDDR6X
Memory Bus192-bit256-bit256-bit320-bit384-bit
Memory Speed16 Gbps14 Gbps14 Gbps19 Gbps19.5 Gbps
Bandwidth384 Gbps448 Gbps448 Gbps760 Gbps936 Gbps
Price (MSRP / FE)$329 US$399 US$499 US$699 US$1499 US
Launch (Availability)25th February 20212nd December 202029th October 202017th September 202024th September 2020

The NVIDIA GeForce RTX 3060 12 GB graphics card launches on the 25th of February at 6 AM Pacific Time. The graphics card will be positioned against AMD's Radeon RX 6700 6 GB graphics card & will be offering performance close to the RTX 2070 SUPER.